-
Notifications
You must be signed in to change notification settings - Fork 0
/
content.json
1 lines (1 loc) · 761 KB
/
content.json
1
{"pages":[{"title":"","path":"archive/index.html","permalink":"http://sx349.github.io/archive/index.html","content":""},{"title":"","path":"categories/index.html","permalink":"http://sx349.github.io/categories/index.html","content":""},{"title":"","path":"tags/index.html","permalink":"http://sx349.github.io/tags/index.html","content":""},{"title":"","path":"about/index.html","permalink":"http://sx349.github.io/about/index.html","content":"<p><img src=\"avatar.png\" alt=\"The Lament for Icarus\"></p><p><a href=\"https://artsandculture.google.com/asset/the-lament-for-icarus/wwGsH3KJkvD1gA\" target=\"_blank\" rel=\"noopener\"><em>“The Lament for Icarus” by Herbert James Draper</em></a></p><p>USC经济学博士在读,人和写出来的东西一样不拘小节。</p>"},{"title":"Workings","path":"workings/index.html","permalink":"http://sx349.github.io/workings/index.html","content":"<h1 id=\"Workings\"><a href=\"#Workings\" class=\"headerlink\" title=\"Workings\"></a>Workings</h1><p>“SCARE OR SPREAD: Spill-over Effect of High Crime Rates on Nearby Cities’ Housing Prices”(term project), May 2015.<br><a href=\"crime0511.pdf\">Download</a></p>"}],"posts":[{"title":"今天的中国,绝对不是“公车上书”时的中国","path":"2022/11/28/gongcheshangshu/","permalink":"http://sx349.github.io/2022/11/28/gongcheshangshu/","content":"<p>读明叔杂谈公众号所刊《今天的中国,绝对不是“五四运动”时的中国》一文,如鲠在喉。我设想,大概在“五四运动”前后,也有类似的“明叔”,试图劝说学生不要参与运动吧?倘若如此,实乃国之大不幸。本文假想明叔穿越时空回到“五四”前夕,回顾“公车上书”,劝说学生不要参与“五四运动”,以为反讽之意。</p><p>阅读本文前,请务必先一览雄文<a href=\"http://www.kunlunce.com/ssjj/guojipinglun/2022-11-28/165647.html\" target=\"_blank\" rel=\"noopener\">《今天的中国,绝对不是“五四运动”时的中国》。</a></p><hr><p>一九一九年五月三日/本报电</p><p>这几天,巴黎和会的协议传来之后,国内一些地方的形势变得有些复杂。一些缺乏上下文的大字报在城内外传播,引起了大家的关注。在几所学堂里,也出现了一些聚集性活动。但是,情况越是复杂、敏感,我们越是要保持清醒的头脑。第一次世界大战宣告结束,解决战争所造成的问题,奠定战后的和平,仍然是我们共同追求的目标。如何在与英美法等协约国保持良好关系的同时,争取我国作为战胜国的正当权利,这是全社会的共同期待,也是全社会需要共同面对的挑战。</p><p>当代中华民国的年轻人勇于表达自己,有强烈的、朴素的正义感,也有很强的行动能力,这不完全是坏事。我跟同年进士老刘曾多次交流,辛亥革命之后成长起来的年轻人,整体上“一代更比一代强”,他们成长在一个蓬勃发展、欣欣向荣的中华民国,他们没有上几代人的思想包袱,他们在潜意识里就认为“没有什么不可能”。当这一代年轻人成长起来后,他们不再受制于强烈的历史屈辱意识和苦难意识,很多人将有机会不再需要为了生存而放弃自己热爱的东西,民国社会将因此展现出更大的自信心和创造力。一个前《申报》同事告诉我,她家的儿子很喜欢“内联升”这样的国产“千层底”布鞋,而且对这些国产布鞋的设计特色一清二楚,他们甚至都“不屑于”去买我们这一代人特别喜欢的洋装皮质马靴。我听了之后很受触动。新一代年轻人成长起来后,对自己的国家、对自己国家的企业和产品,拥有很强的自信心。从某种程度上来说,他们将成为革命以来第一代拥有健康的、“大国国民”心态的中国人。</p><p>年轻人关注国家和社会大事,希望我们这个国家能变得更好,这也是好事。很多年轻人热血沸腾的时候,往往会谈及“公车上书”。但是,我们必须清楚,今天的中国,绝对不是“公车上书”时期的中国。一八九五年的中国,内部清政府腐败无能,外部列强肆意欺侮,强意逼迫在甲午战争中战败的清政府与日本签订《马关条约》。毫无疑问,“公车上书”是一起学生爱国运动,它的出发点是为了捍卫国家主权和民族尊严,从结果上来说,它为辛亥革命的成功起到了积极作用,成为群众政治运动的开端。</p><p>反观今天的中华民国,虽然从微观视角切入,依然可以发现各种各样的问题,还有很多可以改进的空间;但是,如果我们把目光放得稍微长远一点,就可以看到,过去几十年,每隔五年、十年,甚至每隔两三年,社会都会发生很多明显的、积极的变化。这种变化不仅仅体现在数字上,更体现在每一个老百姓的生活都在以肉眼可见的方式,发生向好、积极的变化上,也体现在今天作为一个中华民国国民在国际社会享有的尊严上。不管我们说北洋政府还存在多少具体的问题,但中华民国的进步和改善都是有目共睹的。尽管在西方“主流媒体”的眼中,中华民国依然贫穷、落后,甚至仍未摘掉“东亚病夫”的牌子,但国际社会已经有越来越多具有独立思考能力的人,认识到北洋政府取得的巨大成就,并对中华民国表达出了由衷的敬佩和赞叹。</p><p>在最近的巴黎和会上,北洋政府取得的巨大成就和当今中华民国的国际影响力都得到了高度认可,我在这里只举几个例子:</p><p>——美国总统盛赞了中华民国在巴拿马万国博览会上斩获二百五十八枚金奖,一改清政府在世界列强眼中愚昧、落后的印象,展现出积极上进的新姿态。</p><p>——中方代表顾维钧的发言受到了美国总统与英国首相的握手肯定,更被法国总理克里蒙梭形容为“有如灵猫戏鼠,尽显擒纵之巧技”,受到除日本以外各大国代表的一致赞扬。</p><p>这样的例子还有很多。</p><p>我们并不是说我们要对中华民国存在的问题视而不见,而是说,我们要知道,过去几十年来,我们总体上是成功的。我们要做的恰恰就是,把成功的、好的做法坚持下去,把不好的东西一点一点改过来,这才是让中华民国变得更好的正确选择。</p><p>今天的大学生,到底该以一种什么样的姿态参与到民国的社会热点议题中来?我觉得,每一个有知识、有文化、有理性的师长,都应该告诉今天的大学生,要努力做一个“温和的改良者”,如果有能力,再努力做一个“建设性批评者”。我一直反对,任何人鼓动自己或别人的孩子,去做一个“激进的对抗者”。因为历史已经证明,这将对孩子本人、家庭和社会,都带来巨大的代价。</p><p>美国、英国这些西方列强,从内心深处充满了对中华民国发展的恐惧,他们还充满了对黄种人的傲慢和种族主义态度。他们要遏制民国,不仅仅是在战场上跟民国分个高下,还把目光瞄准了民国社会内部,他们会利用每一起发生在民国的不幸和悲剧,他们会放大民国社会的每一个缺点和不足,他们会想尽一切办法,鼓动民国的年轻人“要勇敢”,但他们的终极目的,不过是要利用年轻人的冲动,来搞乱中华民国。对于美国、英国这些本质上依然相信“白人高人一等”的盎格鲁-撒克逊民族来说,一个软弱的、贫穷的、分裂的、混乱的中华民国,才是一个好中华民国。</p><p>那些情绪激动、觉得自己是在努力捍卫“正义”的大学生,一定会觉得自己有充足的判断力,自己绝对不会受到所谓“境外敌对势力”的教唆和挑动,我也相信他们能做到这一点。但我想说的是,你们的原意是想解决民国社会存在的真实问题,是想让中华民国社会变得更好,但那些“坏人”却是想利用你们,攻击和抹黑北洋政府的政权、政治制度和发展道路,终极目标就是要在民国搞“颜色革命”。如果你们真的想让民国变得更好,现阶段就踏踏实实把自己的本领学好,未来有大把报效祖国、改变国家面貌的机会。</p><p>对于最近几天发生的事情,我也总结一下,我希望,绝大多数中华民国国民,都能达成一个最大或者最小的“公约数”:</p><p>——巴黎和会达成的协议,是众矢之的,中方代表仍在据理力争,我们慢慢就会看到效果;</p><p>——当年签署各种丧权辱国条约的政府官员,接下来也会成为“过街老鼠,人人喊打”;</p><p>——所谓的各类“运动”,充满了煽动性,但在今天的中华民国绝对成不了气候。民国的外交政策一定会不断优化、不断改进,这是我们坚持实事求是原则的内在要求,但任何人要想利用中华民国的问题,来搞乱中华民国,都是痴心妄想;</p><p>——坚持“公理战胜强权”,这是我们每一位国民的共同心愿。要做到这一点,非常不容易。但我们看到,民国社会自上而下,正在朝着这个方向努力。</p><p>战争对我们每个人都有影响,这是毫无疑问的。但是,我们一定要知道,我们曾经经历过没有战争的日子,我们也终将再次迎来没有战争的日子。每一代人都会与时代的宿命不期而遇,抱怨是没有用的。年轻人,静下心来多读点书,多掌握一些技能,即便外边的世界再乱,我们也能更好地保护自己和家人。</p><p>让我们一起为中国的外交官加油吧,最终终结这场闹剧的,一定还是对“公理战胜强权”的坚定信念。我一直觉得,这一天已经不会太遥远了!</p>"},{"title":"有限理性、AdaBoost与博弈论","path":"2019/12/11/column61/","permalink":"http://sx349.github.io/2019/12/11/column61/","content":"<p><strong>这是本专栏的第 <em>61</em> 篇日记</strong></p><p>在本学期最后一次Theory Seminar上,In-Koo Cho教授present了一篇与本校Jonathan Libgober教授合作的论文,标题为:”Machine Learning for Strategic Inference in a Simple Dynamic Game”. 目前并未公开放出working paper,但是可以发邮件向Jonathan索要(就是下图在自拍的这位,是我今年4月去Boston开会时在Harvard经济系偷拍的)。</p><center><iframe src=\"https://drive.google.com/file/d/1Rh2RXGIycF7U-Prxsee-2KRcawioB6ju/preview\" width=\"640\" height=\"480\"></iframe></center><p>有限理性(Bounded Rationality),顾名思义,指的是人在进行决策时并不像传统经济学所假设的那样具有“完全的理性”。对于有限理性的建模有很多种方式,在本文中作者采用的是Rubinstein(1993)的想法:</p><p>一名卖家向一名买家出售一种商品,由于信息不对称,买家不知道商品的真实价值,而卖家知道且能够根据商品的真实价值选择不同的定价策略。进一步地,买家有两种不同的类型,卖家生产商品的成本由真实价值和买家类型共同决定,因此卖家有动力去区分不同的买家。但在买家完全理性的情况下,卖家并不能有效地将高成本的买家有效地排除出去,因此无法实现理论上的最高利润。</p><p>我们注意到,如果买家是完全理性的,那么他总可以根据卖家的定价策略和实际给出的售价,通过贝叶斯更新得出商品价值的后验分布,并理性地决定买或者不买。由于卖家的定价策略的任意性,这种理性的买或不买的决定并不一定是单调的:比如说,也许存在高中低三个价格,当观察到高价时,商品的期望价值很高因此可以买,当观察到低价时,商品的期望价值较低但仍高于售价,因此也可以买,但在中间这个价格时,商品的期望价值反而低于售价因此不能买。</p><p>因此,为了让卖家实现有效排除并提高利润,Rubinstein进一步假设,高成本的买家受到“有限理性”的限制,只能采取“阈值策略”(Threshold Strategy):这类买家不能任意地根据售价决定买还是不买,而是只能选定一个“阈值”,在阈值以上总是做同一个决定,在阈值以下也总是做同一个决定。在这一前提下,卖家可以任意地接近理论上的最高利润,即高成本的买家几乎都被有效地排除了。</p><p>回到Cho&Libgober(2019)这篇文章,“有限理性”和机器学习,特别是AdaBoost又是怎么挂上钩的呢?</p><p>AdaBoost是Adaptive Boosting的简称,粗略地说,AdaBoost的想法是,通过不断增加“弱分类器”(分类效果仅略优于随机分类)的数量并给予合适的权重,提升分类的准确程度。在本文的情境中,给定卖家的定价策略,买家的任务就是实现一个将“价格”映射到“买/不买”的决策上的分类器,其中完全理性的买家的决策被视为正确的分类器,而有限理性、采用“阈值策略”的买家则只能实现一个“弱分类器”。换言之,<strong>本文给予了AdaBoost(以及其它机器学习方法)一个新的行为经济学的解读:给定无穷多具有某种有限理性特征的个体,能否通过自适应的做法调整个体决策的权重,使得最终构成的整体决策符合完全理性的特征?</strong></p><p>具体来说,在Rubinstein(1993)的基础上,本文假设高成本的玩家仍然只限于使用“阈值策略”,但可以选择任意数目的此类策略并且赋予不同的权重,并根据加权平均的结果决定最终是买还是不买,具体的策略数目及权重的选择,则按照AdaBoost的算法进行。当然,为了让AdaBoost能够“学会”正确的分类,此时卖家和买家必须重复进行博弈,因此卖家和其它低成本的买家的“完全理性”需要额外考虑到动态博弈时对未来所有收益的折现,以及高成本玩家使用AdaBoost算法这一事实。</p><p>为了衡量AdaBoost算法的效果,作者引入了Uniformly PAC(Probably Approximate Correct) Learnability的概念。PAC大致的意思是,训练所得的分类器与正确分类器的差别超过任意小的epsilon的概率不超过任意小的delta。对于一族正确分类器,如果算法总能够在有限步迭代中达成这一目标则称为PAC Learnable,如果进一步地,所需迭代步数不依赖于真实分类器的选择,则称为Uniformly PAC Learnable。</p><p>为了简化问题,我们假定卖家能够给出的价格只有有限个(作为对比,通常的做法是允许卖家在连续的区间内任意选择价格),作者证明了如下两个定理:</p><ol><li><p>如果卖家可以任意选择定价策略,且在选定后公开,那么AdaBoost是Uniformly PAC Learnable的。</p></li><li><p>如果卖家可以任意选择定价策略,但在选定后不公开,那么AdaBoost不是Uniformly PAC Learnable的。</p></li></ol><p>定理1比较好理解,如果卖家公开定价策略,那么“正确分类器”,也即给定价格下完全理性的高成本玩家所应该做出的决策是确定的,问题转化为如何让算法去拟合一个给定的函数(不禁让人想起“神经网络能拟合任意函数”的事儿);定理2则说明,如果卖家不公开定价策略,那么存在一部分定价策略具有“迷惑性”以至于AdaBoost学不会完全理性的决策。</p><p>到此为止,似乎本文的核心已经转向了纯粹的计算理论,只不过披了一层行为经济学“有限理性”的皮。但是让我们思考这样一个问题:什么样的定价策略才能顺利地骗过AdaBoost呢?作者于是给出了定理3:</p><p><strong>3. 如果卖家只能选择那些对AdaBoost构成Best Response的定价策略,那么即使卖家不公开,AdaBoost也是Uniformly PAC Learnable的。</strong></p><p>也就是说,如果卖家想要采用定理2当中那一小撮让AdaBoost学不会的定价策略, 是需要付出利润损失的;但是对于卖家来说,买家能否学会理性决策根本不重要,重要的是买家学到什么程度以及相应的自己能获得多少利润。如果买家学不会的时候自己要亏钱,学会了反而赚得更多,那何乐而不为呢?</p><p>相信读者已经意识到,这就是为什么本文标题还要加上 “博弈论”这个关键词的原因:基于机器学习本身的讨论只能把我们带到定理1和定理2,只有理解博弈论,特别是“最优反应”的概念,才能让我们进一步得到定理3。</p><p>Reference:</p><p>Cho, I. & Libgober, J. (2019). <em>Machine Learning for Strategic Inference in a Simple Dynamic Game</em>. Working Paper.</p><p>Rubinstein, A. (1993). On price recognition and computational complexity in a monopolistic model.<em>Journal of Political Economy,101</em>(3), 473-484.</p><hr><p>(2019.12.11 @ 知乎专栏 @ 《有限理性、AdaBoost与博弈论》)</p>"},{"title":"大象公会《穹顶之下》文字实录(转载)","path":"2017/12/18/fulltext/","permalink":"http://sx349.github.io/2017/12/18/fulltext/","content":"<blockquote><p>柴静的《穹顶之下》昨天在网上以现象级传播,我们今天刊载《穹顶之下》的文字版,既是追热点,也是对同行的致敬。</p><p>我是刚做大象公会不久知道她准备做雾霾调查的,当时没太在意,多少怀疑她是低估了这个选题难度的一时心血来潮,因为雾霾成因并无公论,媒体上专家们众说纷纭,而且涉及到大量枯燥数据资料,我也想不出向大众传播时该以什么方式来呈现。因为这两个原因,虽然雾霾是个周期性的热门话题,但我还在做新闻时一直没有去触碰它。去年夏天,意外得知她真是打算拍一部叫《雾霾》的纪录片,虽然刚辞职但已自费带着小团队做了很长时间,已完成了大量前期采访,她预计大概年底作品能出来。我不怀疑柴静的勤奋,但很怀疑最后是什么效果,多少觉得她的野心太大了。</p><p>去年12月6日,终于看到了《雾霾》的PPT大纲,当时极为震撼,有一种实实在在的被密集信息碾压的感觉,作为同行,我深知柴静团队背后采访付出的心血和耐心。我深信它会有震撼性的传播效果,不过,当时柴静并不像今天一些聪明人论断的那样成竹在胸,她十分怀疑雾霾已成老生常谈时,还有多少人像她一样愿意关心它。她的这种怀疑,甚至持续到今年初的小范围试讲,这是每个做新闻的人都共有的困惑:你对一个话题投入的心血和热情越多,越容易怀疑自己是不是错判了大众的关注热心。</p><p>《穹顶之下》的传播影响没有辜负柴静和她的同事们倾注的心血,这或许对同行是一种精神鼓励。</p><p>最后,作为一个看过它的几个版本的观众,多少觉得的遗憾是,网上正式版本内容经过几次不小调整,舍弃了很多非常好的素材,虽然大家已经习惯了。</p><p>大象公会创始人、总编</p><p>黄章晋</p></blockquote><p>转载链接:<a href=\"http://chuansong.me/n/1181431\" target=\"_blank\" rel=\"noopener\">http://chuansong.me/n/1181431</a></p><p><strong>1、引子·动机</strong></p><p>2013年1月份的北京,一个月连续25天雾霾:</p><p>我当时在北京,但当我这一年里反复想回忆当时有什么印象、什么感觉?却记不起来了。因为当时大家都说好像这场雾霾是偶然的气象原因导致的,就没当回事。</p><p>那个月里头我还去了四个地方出差,河南、陕西、江西、浙江,回头看镜头里的天,当时的中国正被卷入一场覆盖了25个省市,和6亿人的大雾霾,但我置身其中,浑然不觉,只有我的嗓子有印象,在西安那天晚上就咳的睡不着觉,我就切了一只柠檬放在枕头边上。</p><p>回到北京之后,我知道我怀孕了,我第一次见到她就觉得她应该是小女孩,因为我觉得她像个童花头,听到她的心跳的那一瞬间,我觉得对她没有任何期望,健康就好。</p><p>但是,她被诊断为良性肿瘤,在出生之后就要接受手术,做麻醉之前,医生对我说,她这么小的年纪做全身麻醉是有可能醒不过来了,你要有个心理上的准备,我还没有来得及抱她一下,她就被抱走了。</p><p>后来护士在我手里放了一只小熊,那是用来安慰小孩的,她用来安慰我。我再见女儿的时候,她还在昏迷,医生对我说手术很成功,但有一件事情你要原谅我,她说她太胖了,所以刚才麻醉的时候,我们扎了好多针眼才找着静脉。</p><p>我把她满是针眼的小手放在我脸上,叫她的名字,一直到她睁开眼睛看我一眼。我是一个非常幸运的人,后来我辞职,陪伴她、照顾她,只要一家人在一起平安就好、健康就好。</p><p>但是回家的路上,我就已经开始感到害怕了,全是烟熏火燎的味,我就拿一个手绢捂在她鼻子上,这样做很蠢,因为她会挣扎,就会呼吸得更多。以前我从来没有对污染感到过害怕,去哪我都没戴过口罩,现在有个生命抱在你怀里,她呼吸、她吃、她喝都要由你来负责,你才会感到害怕,(2013年底)那场雾霾持续了差不多两个月,它让我意识到这件事情不是偶然发生,也不可能很快过去,它就是十年前我在山西时看到的天空:</p><p>2004年我采访这个6岁的小女孩的时候,无论如何我都没想到,她所说的,就是我女儿可能会面临的一个世界。</p><p>这个,2014年全年的北京,只有空气优和良的时候我才能带她出门,但是这样的天能有多少呢?污染天数175天,这意味着一年当中有一半的时间,我不得不把她像囚徒一样关在家里面。十年前,山西环保局长说,孝义是山西缩影,山西是中国缩影,现在我眼睁睁地看着他的话成为现实:</p><p>以前我看过一个电视剧叫《穹顶之下》,它说的是一个小镇上被突然天外飞来的一个穹顶扣在底下,与世隔绝不能出来。有一天我发现我们就生活在这样的现实里。早上起来有时候我会看到我女儿站在玻璃窗前用小手拍着,用这个方式告诉我她想出去玩:</p><p>她总有一天会问我,妈妈为什么你要把我关起来,外面到底是什么?它会伤害我吗?这一年当中我做的所有的事情就是为了回答将来她会问我的问题,雾霾是什么?它从哪来?我们怎么办?</p><p><strong>2、雾霾是什么?</strong></p><p>那么雾霾是什么?</p><p>有时候我在黑暗中会把灯关掉,我想端详一下PM2.5到底长什么样子——但我知道我看不见它,它们是一些空气中悬浮的直径小于2.5微米的细颗粒物,所以它们会把大量的可见光都散射、折射跟吸收掉,留给我们一个能见度很低的世界,但也因为这个原因,我能肉眼看到的颗粒物最少也都是它的20倍,换句话说,这是一场看不到敌人的战争。</p><p>但是有一个办法我能把它拽在眼前,就是PM2.5的采样仪。</p><p>我可以把一个非常干净的采样膜放在采样仪里,我就背着这个仪器24小时,让它模拟我的呼吸。</p><p>这是那个采样膜最初与24小时之后的样子。如果我不对自己跟孩子加以保护的话,所有的东西都会呼吸进去。</p><p>它有多少呢?305.91微克每立方米。</p><p>各国安全值是这样——美国日平均35微克每立方米,欧洲年平均25微克每立方米、中国是日平均75微克每立方米。</p><p>我很想知道那个采样膜里面到底是什么物质,就请北大邱兴华博士给我做了一个测试,他分析出来结果是这样:其中有15种致癌物,其中世界上最强的致癌物之一苯并芘,它是国家标准值的14倍。</p><p>我不太相信这个数字,因为我曾站在四五个焦化厂中间的时候,呼吸的苯并芘才超标9倍,现在我在一个国际大都市的中心,举目四望我根本看不见工厂跟烟筒,我怎么可能超标14倍呢?</p><p>邱博士自己也怀疑了,他说不太可能这么高吧,我再算一次。第二天他给我打了一个电话说:没错,就是14倍。大量的样本显示,有很多人比你更高,超标20倍。</p><p>所有这样致癌物都附着在就你刚才看到的那个黑色的膜表面,这种叫做黑碳的物质上,它非常小,只有0.2微米,但是它是一个锁链的结构,所以如果它打开的话,两克黑碳能有整个篮球场那么大。</p><p>所以它可以吸附很多的致癌物和重金属,在中国这样的黑碳有多少呢?2009年NASA做过一个测算。大气中黑碳含量最高、紫的发亮的地方就是中国,它像幽灵一样在我们的上空飘荡。</p><p>我又去了找了北大,我知道它有一个实验室,我问,能不能当一个志愿者,你把我放在一个高浓度的实验舱里面,测试一下我的身体的反应,这样我可以告诉大家。</p><p>他们犹豫了一下对我说,不好意思,这个实验没法做。你知道每一个实验都需要一个伦理安全值,这个安全值要保证受试者不能受到太大的伤害,但如果我设定这个值的话,外面大气比里面实验舱的空气(污染)浓度还要高。</p><p>换一句话说就是,我们每个人都生活在一个终生暴露的实验舱里。它的结果会是怎么样的?北大换了一种实验方式,他们就直接测试人们在日常生活中身体的反应,长达一年。我自己也参与和体验这个实验。</p><p>结果是:我自己有呼吸道炎症,是易感人群,细颗粒物会影响呼吸系统增加气道炎性反应还会增加进一步影响心血管系统,超细颗粒物升高跟心脏供血能力降低有关系。</p><p>在科学家的帮助下,我们把它做成了一个动画:</p><p>这是中科院做的一个测算,他们有一个简单的结论,PM2.5上升到200微克每立方米的时候,我们各种呼吸系统疾病,和心血管系统的疾病死亡率会增加14%到26%,如果这个数字听起来很拗口的话,那么最起码中国的前卫生长陈竺做出了一个非常简单的测算:在中国每一年,因为大气污染而死亡的人数有50万。</p><p>在这场跟人类战争当中,最脆弱和最容易受到伤害的就是我们的孩子和父母。这些小孩大都才两个多月大,还没有出过门,但是已经得了肺炎,他们在接受雾化治疗。</p><p>他们的父母让我拍下他们,怀疑是重雾霾期间没有防护导致的,但医生跟他们说,我只能给你诊断为不明原因肺炎,因为我们还没有目前的流行病学调查。</p><p>但是我可以告诉你的数据是这样的:2013年1月份,重雾霾期间,我们整个国家有27个城市都出现了急诊人数的爆发性增长(10%–150%)。</p><p>有的时候我们小区的妈妈们会聊起这件事情,有一个妈妈就问我说,我们该怎么办呢?我们要不要把小孩尽快地送到雾霾天里面去,让他们锻炼锻炼,适应适应,是不是不能让孩子输在起跑线上?</p><p>所以我就采访了曾经空气污染严重地区洛杉矶的一位预防学专家,让他来告诉我们,到底把小孩暴露在高浓度的空气里面,能不能产生适应性?</p><p>“我不认为有任何证据表明,让孩子多暴露在污染空气中,能帮助他们产生“适应性”。如果让他们暴露在污染中的第一天,他们受到一部分功能损伤,他们暴露的第二天,损伤没有第一天多,不是因为他们产生了适应性,而是损害已经发生了。”</p><p>前两天雾霾严重时候,AQI指数已经超过了500爆表,严重污染。可是,我楼下这个小学孩子们还在打球,跑步,玩耍,在运动的时候,人们的呼吸量会增加5到10倍。然后我看着他们,我意识到我不可能一直保护我的孩子,总有一天她要到社会当中去。</p><p>呼吸是没有办法选择,也没有办法逃避。你的每一口呼吸它都在,一个成年人像我们这样的人每天呼吸大概25000次,天长日久,日积月累,这些东西在我们的肺里面会是什么样子?</p><p>我们采访了一位北京肿瘤医院的患者,她是一个早期的肺癌患者,一位50多岁的女性,在医院的财务室工作,她跟家人都没有吸烟史。但是她却不得不在此接受治疗。</p><p>以下是我和医生、专家就此的对话:</p><blockquote><p>麻醉师:这个人抽烟吗?</p><p>杨主任:这个人不抽烟。但是她是山西的,是你们(柴静)老家的。</p><p>麻醉师:空气质量不太好。</p><p>杨主任:对,不干净。所以这可能是大气里面的东西。</p><p>杨主任:人体只有肺的淋巴结是黑的。就是因为人的肺吸了很多粉尘。淋巴系统是人的免疫功能,免疫功能就是防疫功能。</p><p>记者:那我问您,当这个黑色的物质在上面沉积之后,影响它对癌细胞的阻挡吗?</p><p>杨主任:所以是不是已经把它的免疫功能锻炼得疲劳了,所以你看今年的世界卫生组织WHO呢,已经把肺的慢性疾病划为肺癌的高危因素之一。</p></blockquote><p>让我们非常安慰的是这场手术很成功,患者在肿瘤摘除之后基本可以康复,医生就把这个黑色的淋巴结摘下来之后拿给她的家人看。</p><p>那家人问:您能告诉我我家人为什么会得肿瘤?</p><p>他说:我只能根据我的临床的经验和推断告诉你,我怀疑是空气污染,世卫组织告诉我,颗粒物被列为人类的一级致癌物。</p><p>我听他说的时候我有一个疑问。如果是肺癌的话,应该有一个潜伏期,起码应该在十年左右吧?但是雾霾我们这两年才听说,怎么能这么快就导致肺癌呢?</p><p>我就去找了我们国家奥运呼吸质量保障小组的组长唐孝炎院士,她给我看了一条曲线:</p><p>2004年,11年前,那时候我们还没有PM2.5的检测,但是我们有PM10,她说根据当时的估算,那个时候在污染期的话,PM2.5可以达到300到400,属于今天的严重污染。</p><p>我问她,我说那我失明了吗?2004年我在北京。</p><p>她说,你看看这张照片吧。</p><p>那是什么?我说那应该是霾吧,那我怎么不知道呢?她说你再查一下当天新闻报道的标题——《首都机场因雾出现今年最严重的延误》,那个时候我们一直认为那是雾,我们一直把它叫做雾。</p><p>看到这张照片,作为一个曾经的记者,我内心有些自责。那些年里面,我一直认为自己是在报道各类的污染事件。但是我一直感觉,好像听到矿山的炮响才叫污染,看到工厂的烟筒才叫污染,从来没有想到过我们生活在都市中心的人,每天看到的天空,就是污染。</p><p>那个时候,我已经不再是记者了,我早就是当事人。</p><p>但我还有一个疑问。如果是那样的话,小的时候家家户户都在烧煤油炉子,对吗?我们每天晚上都在呛人的味里睡着,再醒来。我怎么没有听到那时候有很多人得肺癌呢?</p><p>带着这个问题,我认识了我们国家从事这个工作三十多年的何兴舟教授,他给了我一本资料——中国26城市大气污染与居民死亡情况调查(1976到1981):</p><p>那个时候的大气污染,已经跟居民的肺癌死亡率分布一致,那时候的大气污染就是煤烟性污染,只不过这份材料在当年一直是内部资料。</p><p>一个人,别说是一个人了,一个活物,在我看来应该这么活着,春天来的时候门开着,风进来、花香进来。有雨、有雾的时候,人忍不住想要往肺里面深深地呼吸一口气,那种带着碎雨的那种凛冽的、清新的感觉。</p><p>秋天的时候你会想跟你喜欢的人一起,就一个下午什么都不干,懒洋洋的晒一会儿太阳,到了冬天你跟孩子一块出门,雪花飘下来,她伸着舌头去接的时候,你会教给他什么是自然和生命的美妙。</p><p>但现在呢?</p><p>这一年每天醒来,我做的第一件事情就是先看一下手机上的空气质量指数,用它来安排我一天的生活。我就靠盼着一点西北风过日子,我每天戴着口罩上街,带着口罩见朋友。我用胶条把我们家的门窗每个缝都给它粘上,带着孩子出门打疫苗的时候,她冲我笑我都会感到害怕。</p><p>我不是有多怕死,是不想这么活。我要知道它从哪来,我要弄明白这一切是怎么回事。</p><p><strong>3、复合污染</strong></p><p>我把采样膜放在科学家的电子显微镜底下,我就看到它的样子,它不是我想象中的一颗一颗一模一样的小米,它是各种各样的东西混合在一起,这个混合会增加它的毒性,还会让它迅速变得更多更大。它们之间可以发生化学反应。</p><p>中科院告诉我它的来源是什么?燃煤、燃油、生物质燃烧、工业、养殖、化肥、扬尘,绝大部分都跟人类的活动有关系。如果你对它加以归纳的话,你会发现它的大部分从哪来。</p><p>从燃煤和燃油当中来。也就是化石能源的燃烧。</p><p>我们中国的燃烧的强度有多大呢?中国比欧洲高出3到4倍,我们烧了这么多化石能源。</p><p>我采访了一些专家,他们如此回答道:</p><blockquote><p>王跃思:“我们燃煤的污染没解决,汽车污染叠加进来了”</p><p>郝吉明:中国现在的这个污染,基本是两种污染的叠加。</p><p>唐孝炎:多种多样的物质给你在大气里进行各种各样的反应,</p><p>王: 空气静稳条件(增加),这些污染物没扩散,它之间反应了,这就是二次粒子。</p><p>郝:复合污染</p><p>唐:吸附,粘连,碰撞,合并,全都混在一起了。</p><p>王:我们不可能改变自然条件,我们只能改变我们自己,这个自己的办法就是源头减排。</p></blockquote><p><strong>4、煤怎么了?</strong></p><p>全世界都要烧煤和油。我们的煤怎么了?</p><p>你们知道中国烧了多少煤吗——2013年的时候已经36亿吨。</p><p>你知道全世界其他国家烧了多少煤吗?——我们比全世界其他国家加起来烧的煤还要多,上一个达到过这样消费量的是1860年的英国,但在之后他们为此付出了沉重的代价。</p><p>所以,1960年代,大烟雾事件发生后,其他的国家纷纷减少和控制自己的煤炭用量。</p><p>然而,当时恰逢中国改革开放开始,这个已经封闭了多年和落后了多年的国家,迫切的需要一种巨大的能量来让自己起飞。</p><p>它的选择就是煤炭。跟随我们其后的是现在的印度,印度很快会成为世界上第二大煤炭消费的国家,而且印度的工业现在对重要的污染物的排放还没有规定的标准,印度也是世界上空气污染非常严重的一个国家。</p><p>那中国这么多煤用在了哪呢?</p><p>2013年的36亿吨——3亿8千万吨,烧在了京津冀,3亿8千万吨当中有3个亿烧在河北。</p><p>河北。</p><p>后来采访的时候,清华郝吉明院士告诉我说,你听没听过全世界钢铁产量排名?他说是中国第一,河北第二,唐山第三,美国第四。还补了一句说唐山瞒报产量第五。</p><p>我想用煤多工业多不一定代表污染,别的国家也有钢。他说你去一趟唐山看看就知道了,所以去年的10月份我就跟环保部的华北督查中心一起带着无人机去唐山,想去巡航一次。</p><p>我们确实采到了现场。但最让我意外的是,花了这么大的代价,这么多艰辛,我们也明明拍到这些排污的证据了,最终这个企业没有受任何处罚。我问大为为什么?他说你去问问我的领导吧。</p><p>他的领导就是环保部科技司熊司长。他说,环保部门无法取缔关停,就业与银行贷款因素很关键。</p><p>一个地方有大量的重工业布局集中,消耗大量的煤,而且还没有办法很好的控制污染,同时还无法取缔和关停会是什么结果?</p><p>我有一个好朋友叫老郝,也是我当年新闻调查的同事,十年前是他跟我一块去山西拍的污染,他知道我们家是临汾的,有时候会发张照片调侃我,说你们家现在是全球污染城市之首,我也无可奈何。</p><p>然后十年之后我就把这一张写有“为我市退出全国74个城市空气质量排名倒数第一而喝彩”的照片发给了他。</p><p>这是他们家河北的标语,很遗憾的是,这个城市三个月后又反弹回了第一名。但这种玩笑真的是让人心酸,而且对于一个居住在北京的人来讲,只要偏南风一吹,从河北来的污染会这样过来。空气是没有墙的,什么叫同呼吸共命运,这就是。</p><p>我还有一个朋友叫老范,江苏人,以前我跟老郝两个人争论敝省还是贵省更污染的时候,他就在旁边说哎呀我们南方人真的不知道什么叫做污染。</p><p>现在他们华东也沦陷了。但他很奇怪,他说不可能吧,我们华东至少跟煤应该没有关系,因为我从小到达都没有见到过煤,对,我给她看这张图:</p><p>这是你们家的钢铁厂,这是你们家电厂,这是你们家附近的水泥厂,在你们江苏,每30平方公里,就有一个电厂。</p><p>而被夹在中间的上海,烧的煤也不少,它是单位燃煤量最高的国际性大都市,每一平方米的土地上要烧掉10公斤煤炭。</p><p>我们的煤越烧越多, 会是一个什么样的后果?</p><p>好煤会烧得越来越少,对吗?</p><p>劣质的煤炭就会越来越多,劣质到什么程度?</p><p>这是我们常见的煤吗?我最初认为这肯定不是煤,这是木头吧,因为还看见纹路包括上面的化石,但是洗煤协会的人告诉我,这就是煤,是褐煤,是全世界最年轻的煤,年轻到了它的煤化程度特别低,一烧的时候将近有一半会变成黑灰全都飘在空中,那这样的褐煤我们用了多少呢?</p><p>这是这些年来,我们生产它的曲线:</p><p>它造成的后果是这样,2013年10月21日,第一天供暖的哈尔滨,它的雾霾爆表,PM2.5在1000以上,当天的哈尔滨人们是这样生活的。</p><p>我问过哈尔滨环保局,我说听说你们那两天车开出去,公共汽车都开丢了,他苦笑了一声说,别说公共汽车了,连我们局长都丢了。说他那天到现场去检查工作,没一个人认出他,因为雾霾太大。</p><p>当天的哈尔滨的雾霾就是因为他们烧了两千多万吨的褐煤,加上在2013年的时候全国几十万台这样的小锅炉,没有任何标准,也没有任何监管,就这么一烧,就这么一排,就这么一放,你在高空往下俯瞰的时候,1200万人,就像是被水泥砌在了底下。</p><p>煤本身并不一定意味着脏,德国人也用褐煤,但是德国人会把它们清洁,把它们进行提质,煤炭可以弄干净的,英国是95%以上都给它洗干净。但在中国,我们才洗了不到一半的煤。剩下的另外一半当中基本有几亿吨是没有任何用而且又污染又浪费的叫灰分的东西,我们就这么把它拉着,用汽车拉,用火车拉,满世界的跑,再放到煤炉子把它烧出来,毫无作用,造成的结果却是这样。</p><p>很多的煤是被我们普通的居民给烧掉了,在中国城市里面,我们现在都有四分之一的人是在烧散煤。他们烧的量不算大,只有20%,但是这20%排放出来的二氧化硫,跟所有的大电厂加起来一样多,就这么厉害。最要命是,它的毒性还非常大。</p><p>在北京冬天的时候你一定要小心,因为PM2.5的毒性远远大于夏天。光致癌物的含量就是夏天的25倍。</p><p>它为什么会这么高?散烧煤炭是主要原因之一,它的致癌物的排放因子是非常高的。中国人几十年来早就在承受这种散烧煤炭带来的痛苦。</p><p>云南的宣威,有个叫虎头镇的地方,散煤燃烧导致它是全世界肺癌最高发的地区之一。</p><p>这是一个肺癌晚期的病人,他在咳血,但是他已经没有力气去驱赶苍蝇了,所以他就请别人在他身上放了一张粘苍蝇的纸。这是这个老人和这些年他身边被肺癌剥夺走生命的人。而有的家庭已经空无一人。</p><p>我们总是听到一种声音说中国还是发展中国家,跟我们谈环保太奢侈了。但往往受污染最严重的就是最穷的人,他们最容易受到伤害,他们最需要发出声音,他们最需要得到帮助。</p><p>这些年来,中国消耗了这么大量的煤炭,我们的煤炭变得越来越劣质,我们缺少清洁,我们在排放的时候还缺少控制,造成的结果就是这样:在中国,煤炭消耗量越多的地方,它的PM2.5的浓度几乎就越高。</p><p>我们采访了发改委能源所负责人李俊峰,与中科院大气物理研究所研究院王跃思,他们说:</p><blockquote><p>李俊峰:我就说把煤最简单的办法,我们要降到20(亿)吨以下,我们的天一定是蓝的。我们做不到这一点,我们的煤能不能集中去燃烧去,把散煤降下来去集中发电。我们即使再退一步这一步你也做不到,我们把散烧的煤洗干净行不行。</p><p>记者:如果这个煤能够洗干净的话,现在排放?</p><p>李俊峰:至少减少一半。我们的改变就两条,必须把煤降下来,或者把煤弄干净。</p><p>王跃思:如果目前环保设施,100%按照国家的标准都到位了,我们的尘排放跟二氧化硫排放,应该比现在低60%。我们的氮氧化物应该低于35%。咱们中国的污染治理,第一得靠管理,第二还得靠管理,第三还得靠管理。习总说法律的尊严在于执行。</p></blockquote><p><strong>5、油怎么了?</strong></p><p>中国用三十年的时间走完别人一百年走过的工业化的道路,所以我们煤的污染之上还要再附加油的污染。</p><p>我们的油大部分烧在车上,中国的车辆十年之间增加了将近一个亿,光北京2010年一年增加80万。</p><p>80万,把所有的车头和尾这样连起来,可以从北京排到深圳,然后再从深圳排回来,这是一年增加的。造成的结果就是这样。</p><p>北京本地的污染源当中最大的就来自机动车。</p><p>还有一个城市出乎我的意料,“某城市一年的雾霾日达到200天以上”,我就查了一下。你可以猜一下这是哪个城市。</p><p>没错,杭州。</p><p>说实话我跟你一样吃惊,是因为我一直以为这叫山色空蒙雨亦奇。</p><p>后来我就打电话问杭州市环保局,他说是,因为它是全国人均机动车保有量最大的城市,平均两个人一辆车。所以它的机动车污染要占到接近40%。</p><p>那车多就污染吗? 东京比北京车多,为什么没有这么严重的污染?</p><p>北京交通发展研究中心给我的答案是这样的,东京每天90%以上的人是在轨道交通上,他们只有不到6%的人在开车。而北京34%的人开车,中心城区基本这么堵着,一堵的时候一怠速排放是2倍以上,一刹车重金属的排放会更多。</p><p>北京每天高峰期,六环以内每个小时的PM2.5的排放量是多少?一吨。</p><p>北京人在五公里以内开车将近一半,而且我们当中有12%的人是在两公里内开,我们有7%的人是在一公里以内开。</p><p>就我们爱开车?也许有一小部分人是这样,但不会这么多人都没有理性吧? 我们有辆车,但是有宝宝之后我们的约定是,除了老人小孩,机场医院我们基本不开,我先生每天要上班,然后他就骑自行车,骑自行车在我们家楼下是这样骑的,只能这么骑。</p><p>为什么呢?这是我们的自行车道,他基本上天天就被车这么压着,北京500多万车只有一半是有停车位的,但没有一辆停在半空中,所以它就压在自行车道上,压在人行道上,就这么开。</p><p>当然全世界人性都差不多,不管的时候都这样。我们看看伦敦,当年好像也这么停的,后来伦敦就装了计时收费器,然后是这么停车的,后来伦敦又提高了自己的停车费和拥堵费,然后车是这么停的。</p><p>五十年前东京也这样,不是很容易的,所以他们用了很长的时间才一辆一辆的把它清出去,这是对城市的精细化管理的一个考验。</p><p>爱车的人还会有另外一个疑问,就是这张照片,你看一看,北京的夜里没有车对吗?那为什么污染也这么严重呢?它真的是车带来的吗?</p><p>我的困惑一直到看到这张表才解决:</p><p>这是中科院给我的图,每一天晚上,你可以看看这张图,每一天晚上到零点的时候,有机碳的排放都会出现一个峰值,都会比当天下午三点钟要高出将近两倍,每天晚上固定出现,什么东西在北京的夜里12点定时出现呢?</p><p>我也想知道这个答案,我就去了延庆:</p><p>没有任何排放设施的时候,它的排放会是一个什么结果?</p><p>只这一辆车,只一项颗粒物的排放,它就是国Ⅳ车的500倍。所以柴油车的比例也不算高,它跟散煤差不多,并不多,只占17%,但是它的氮氧化物的排放,要占所有机动车的70%,它的一次性颗粒物的排放,要占到99%。</p><p>还有更可怕的事情是,柴油车的尾气排放出来的颗粒物,毒性远比一般的大。</p><p>多环芳烃是致癌物,对吗?但是,柴油车排放出来的是它的升级版,叫硝基多环芳烃,它的致癌性要高出一个量级,我在延庆的时候看到这些柴油车司机,他们本人就是这种柴油车尾气最大的受害者,是癌症的高危人群。而且去处罚他们显得很不合情理,他花的是真金白银自己的钱,他买的是贴了合格证的车,要找也要找那些造假的企业来问问吧:</p><p>不知道非洲兄弟听了什么感觉,企业一定会有自己的一个辩解和理由。但是我好奇的是,十几年来,如果说这种全面造假普遍造假都存在,而且监管部门知道的话,为什么不去处理他们呢?</p><p>我们不是有法律吗?我们不是有缺陷汽车产品召回的管理条例吗?你这样的产品是可以召回的呀。</p><p>但是从2004年到现在,这个法律在这类车辆上一次都没有用过。</p><p>为什么?我得到的答案是——只有当它处于“危及人身和财产安全的,不合理的危险”的时候才能召回。人家说了,这个环保设施不装跟安全有什么关系,又不会出人命。所以就没有召回的依据。</p><p>那我们有没有处罚它,让它销毁的权利呢?我们有,大气防治法已经规定了,你既可以让它停止违法行为,可以罚款,也可以让它没收销毁。那2002年这条法律到现在以来,十几年过去了,它却一次都没用。</p><p>为什么呢?我得到的答案是这样的——必须由“依法行使监督管理权”的部门来行使。每个人都说,我不知道这个部门是谁。那一共跟这个机动车管理的就那么几个部门,我就挨个问了一下,我先问了一下环境保护部,环保部说,听说不是我。我又问了一下工信部,工信部说绝对不是我,我又问了一下质监总局,质监总局说应该是我们三个吧。</p><p>也许他们的困惑有一定的理由,就是你这个法律怎么不说清楚谁管了?我又问了全国人大,他们给我的答案是这样的,这条法律看起来执法主体确实不明确。这在我们的立法当中是非常少见的情况,为什么偏偏在《大气法》的53条当中就会出现这样的情况呢?</p><p>我们采访了环保部机动车处处长丁焰,他这样说道:</p><blockquote><p>丁焰:当时我们问过,你《大气法》为什么要这么写,为什么不直接写我们?人家说了,就是因为当时定这条的时候很多部门反对,不能你环保管,但是最后这个法就过不去,过不去最后等于就用了一个模糊的写法,就是叫有执法权的去做</p><p>柴静:听上去就是你们三家都查环保?</p><p>丁焰:但是实际上你还是没管住啊。合格证全是真的,车型也全是真的,环保部门也发了那个绿标了,也是国四的,没有一个部门去看那个车到底是什么,只有那个车是国一的。</p><p>柴:你说你(环保部门)有执法权,也不会有人说有问题,那你就去执法?</p><p>丁:环保部门肯定是有惰政的嫌疑,这个我承认,肯定是这样,你也没有再追。</p><p>柴:那你们这些年来你们执法连点儿牙齿都没有?</p><p>丁:我现在都张不开嘴,我怕人看见我没牙,环保尴尬不就尴尬在这儿嘛。</p></blockquote><p>尴尬。</p><p>不光是环保部门,车企也挺尴尬的,那个造假的老板后来就跟我说,如果环保部能够去执法,去抓那些造假的车辆的话,我保证第二天就生产真的,否则的话,我生产真的,别人生产假的,明天我就垮了。然后我就问丁焰,我说你觉得人家说的合理吗?他说合理,不执法的结果就是逼别人作假。</p><p>车企还有一个说法,说我们国Ⅳ的标准早就已经确定,让大家上新车,但是国Ⅳ的油迟迟不供,而且供上来的油质量也不高,我们就在延庆的时候,随机抽查了这辆柴油车,这是在北京加的柴油,已经是非常好的,全国最高的水平了。但是它的柴油测出下来之后,它是欧盟和日本包括美国的25倍,这样的一个柴油品质。</p><p>那我们的汽油是什么品质呢?环科院给了我一个结论:我们国家油品长期比发达国家低两到三个等级,只要提高一个等级的话,我们排放可以减少10%。</p><p>我们看上去有好油,我们有国Ⅳ以及以上的油,但是这部分油只占3%的比例,这就是杭州的痛苦,杭州有钱,想买好油,但是2013年之前就是没有,费了九牛二虎之力,你只能烧国III的油。</p><p>即使在北京,看上去我们的硫含量跟欧洲美国已经接轨了,但是你再看看,要么它们就是增加大气的氧化性,形成颗粒物,要么就是增加毒性,最后一项蒸气压,它增加的是汽油蒸发出来到空中的强度。</p><p>我原来想:蒸发能蒸发多少?后来环科院的专家跟我说,你看看这照片吧:</p><p>这是用红外线拍摄的,加油站如果不加油气回收装置的话,它蒸发出来的量是这么大,每一升大概要蒸发出1.5克到大气空中来,而且我们平常开车的油箱也会自己蒸发出来,它蒸发强度大到什么程度?就是在北京我们光蒸发出来的碳氢化合物,就比汽车尾气排放出来的还要多。</p><p>而这一项物质是PM2.5的重要原料。那为什么我们不赶紧把油品弄的干净一点呢?石化行业的回答是这样的——因为国家标准不够高。这个说法看上去是合理的,你又没有理去要求企业生产比国家标准更高的油,对吧?</p><p>那我们就去问,环保部、环科院,参与制订油品标准的人,你为什么不把标准定高一点呢?我采访了环科院岳欣,他给我的答案是这样的:</p><blockquote><p>小标委(石油燃料和润滑剂分技术委员会)绝大部分,大部分是石化行业的人,这个委员大部分是石化行业的。</p><p>记者:多大比例?</p><p>岳欣:大概是67%左右,大标委(全国石油产品和润滑剂标准化技术委员会)更多,大标委几乎90%以上都是石化行业的人</p><p>记者:秘书处也是石化行业?</p><p>岳欣:标委会的秘书处也在石化行业内。</p><p>记者:我曾经还可能以为你们有一票否决权。</p><p>岳欣:照说应该,环保嘛</p></blockquote><p>岳欣在采访结束时的苦笑实在给我印象太深了。他回答我说,我唯一能体现我存在感的就是进门的时候在签到本上写我名字的时候。因为以前我连个投票权都没有。</p><p>他说他们也会极力地想要推动这个油品的升级,但是石化行业往往的解释是这样的,我们国家的石油很大一部分从中东进口来,对吗?</p><p>比如说伊朗的油很多是重硫油,在德黑兰,他们自己也承受着这种油品质量不好带来的困扰,雾霾很大。 但他们的问题是他们没有升级的工艺,可是中国的企业是不缺这个技术的,双方的分歧就在钱上。</p><p>国III升国Ⅳ的时候,环保部说一升(涨)7分钱就够了,石化行业说那不行,得5毛钱,我说你们双方就摊开成本算一算,到底是多少钱,很简单的事情。他又苦笑了一下,说别说是我了,你去问问发改委财政部,你看他们能知道吗?</p><p>我们于是采访了发改委某位官员:</p><blockquote><p>某:控制不了,比如说你是独生子,你这个小孩要学坏,你做母亲的有什么太大办法,只能全心全意打一顿呗,你能天天打啊 那也不行啊。人家有十个孩子,也不会调皮捣蛋的对吧。</p><p>柴:他们在你们面前?</p><p>某:人家根本不搭理我们,他副部级单位,你拿什么招啊?他搭理你啊,你价格司今年不涨价,明年逼上门来,你涨不涨,不涨我断供了,就这话。</p></blockquote><p>我不知道环保部和发改委说的这些话是不是真实,有没有依据。所以我需要查证,我就向我们国家的石油标准委员会主任,和中国石化集团前总工程师曹湘洪先生提出想见面跟他聊一聊问几个问题的想法,然后他同意了。</p><p>一问:为何石化行业主导标准?</p><blockquote><p>记者:大石化行业的人占(标委会)多数的话,有可能会损害掉了大众的公共利益</p><p>曹湘洪:这个坦率说组织石油产品标准的制定,不应该找一个不懂炼油企业的人来制定。</p><p>记者:环保部门的人不懂吗?</p><p>曹湘洪:环保部门我认为他不懂。</p></blockquote><p>二问:到底谁该来制定标准?</p><blockquote><p>记者:他是觉得如果标准是你定,然后升级的时候你又说升不了,然后到了期限你又供不了油,你可能说出很多原因,但是人们就不信任你们了。</p><p>曹湘洪:对,那不信任是很容易造成误解,但是回过头来,谁有能力来起草这个东西。</p><p>记者:那么有没有这种可能比如像其他国家的情况,第一它的环保部门的人,有懂行业的人他雇佣了不少这样的人。第二他在提交这个草案之前,他就会跟石油的行业有充分的接触、了解、沟通。然后到最后的时候他的健康报告,他的经济报告,他的炼油成本这块全部都提供给委员会。然后通过公开这个程序让大家来做研判,而且都有彼此发言,彼此辩论的机会。如果是这样的方式,您觉得是可以考虑接受?</p><p>曹湘洪:可以接受,我早就想卸任标准化委员会主任。 外面人就说你拐向中石化,中石化就说你吃里扒外。</p></blockquote><p>三问:为何不公开升级成本?</p><blockquote><p>记者:你们能不能把自己的投入还是升级的所有的成本能够公示,让大家有个评判呢?</p><p>曹湘洪:现在社会上自然的对中石化,中石油这样的企业有一种不信任感。</p><p>记者:您认为这种不信任是从什么地方来?</p><p>曹湘洪:我们很多人不讲辩证法,只要是有缺点就把你无限放大,越描越黑,还不如不说。</p></blockquote><p>四问:“是否提高标准倒逼升级?”</p><blockquote><p>记者:为什么我们不把标准提高一些,用它来倒逼油品的升级呢?</p><p>曹:油品事关安全,弄的不好就出大事了,不是阿猫阿狗谁都可以来搞经营的。</p><p>记者:他肯定会想着为什么中石化、中石油挣钱呢?大家也都可以挣钱。</p><p>曹湘洪:是大家都可以赚钱,所以我同意深化改革逐步放开,如果国家决定要推进这样的改革,那你国家去研究去吧,可能会有什么样的风险。</p></blockquote><p>五问:是否可以承担更多社会责任?</p><blockquote><p>记者:中石化已经是一个如此庞大的一个国有企业了,那么为什么不能够在关于环境问题上能够更多的承担起自己的社会责任呢?</p><p>曹湘洪:我个人觉得中石化是大,就像一个人很胖,但是虚胖。</p></blockquote><p>追求利润这是公司天性,无可厚非,但是我们目前为止查到的结论:澳大利亚、韩国、日本、墨西哥、加拿大基本都是环境部在主导标准,即使欧盟这样的欧洲标准委员会来定的时候它会有行业协会的声音,但是也没有出现过由石化行业来主导标准制定的情况。</p><p>在中国,除了我们的油品标准低之外,还有另外一个在我出发前远远没有想象到的问题,就是我们有一半的油,是完全失控的。</p><p>全球十大港口,现在有七个在中国,远洋货轮带来的污染之大,在靠近海岸线400米的地方,它排放的污染相当于五十万辆大卡车,所以在深圳66%的二氧化硫是来自轮船的。</p><p>也许你不在港口边上,但你可能生活在河流的边上。</p><p>中国20多万艘这样的船,他用的都是最差最差的那个重油,污染到什么程度?有的码头上你可以用手把黑碳一颗一颗的捡起来。也有可能你会生活在机场的边上,飞机一起飞一吨煤油没有了,一落地一吨煤油没有了。所以广州白云机场每天排污量相当于60万辆出租车的同时排放。</p><p>也许你既不生活在港口边上,也不生活在河的边上,也不生活在机场的边上,但你一定遇到过工程的施工车辆。</p><p>有一次我在河北看到他们的时候,我以为前面失火了,跑到前面一看是它在排放黑烟,我们就跟着它进入了其中一个村庄,看到村庄门口是这样的,这个小房间里面是村里人在这里打麻将、卖彩票、还卖柴油,我不知道这些柴油从什么地方来,当时我们跟环保部一起,环保部门就说我们去查一下看看吧。他们就现成买了一个柴油壶,然后就进了最近的一家民营加油站,我是最后一个下车的,等我下车的时候发现他的证已经被加油站的老板夺走了。</p><p>老板说,环保部不能查我们,你们“有义务没权力”。听完这句话之后,我们每个人都默默无语,然后就拎着柴油壶就散了。他说的太狠了,从煤到油一项一项下来,煤和油的消耗都这么大,我们的品质相对低劣,我们缺少清洁,我们还在排放的时候缺乏控制,我一直想知道为什么,一直到这个老板这句话我觉得他莫名其妙的道出了某种事物的本质。</p><p><strong>6、环保部怎么了?</strong></p><p>有一天晚上,有一个从事环保工作的人给我发了1999年的一张报纸:</p><p>当年他大学毕业,他说他受到感召,进入环保局,很想有所作为,他问我说,我怎么觉得这么多年下来,有时候我觉得自己是一个吉祥物呢?</p><p>一吨钢如果把它所有的环保成本省下来,不去装的话,他能够省100块钱,一吨煤能够省156块钱,一辆车如果不装环保设施的话,能够大概省两万。油品少升级一次,能够省500个亿。</p><p>10年前我问空气中是什么味道,我没有得到答案,现在我知道了,空气中是钱的味道。</p><p>曾经,我报道过的一位市委书记一定让我拍一拍他治理污染的成果,我们就拍了他炸烟囱的视频。他很满意,说了一句话,小柴啊,这个是为你们炸的。三年之后我又去了,原地建了一个更大的,倒是装了环保设施,我问我说运行过吗?老板搓了搓手说,哎呀,快了。</p><p>当时那个书记跟我说,都跟我讲环保,跟我讲环保,我问你,谁敢把中国的经济掉下来?</p><p>我也想知道这个答案,所以十年之后我就去了河北的钢材市场,进去之后,空空荡荡,这些粗钢都已经生了锈,埋在荒草里头,然后有的都做了鸟窝。</p><p>我就问这个经销商小伙子,我说生意怎么样,他说你看看旁边有车来拉吗,他说这么多企业,生产的都是一模一样的最低端的东西,而且只靠量竞争,只靠价格竞争,到最后我现在还不如去卖白菜呢。</p><p>我们生产一吨钢,要消耗多少能源呢?600公斤煤炭,3到6吨水,我们要排放多少呢,这么多二氧化硫,还有这么多烟尘,那生产这么一吨钢,现在利润有多少?你可以猜一下,多少?1000?</p><p>不,不到两块钱,连一只茶叶蛋都买不着。</p><p>烧一吨煤呢,大概得排放多少污染物?这么多二氧化硫,这么多二氧化氮,这么多一氧化碳还有这么多粉尘,你也可以猜一下一吨煤现在利润多少?多少?两个茶叶蛋?</p><p>你还真是高估了,它的结果就是连一杯饮料都买不着。</p><p>我们36个重工业当中,现在22个严重过剩。所以中石化的曹湘洪先生有一句说的是挺准确的,就像一个人胖,但是虚胖。</p><p>可是这些企业现在还在接受着大量的补助,其中一家企业政府为了避免让它退市,就隔一年给它补一次。</p><p>补多少呢?隔一年20亿。而有个已经严重亏损的企业,在拿到这个补助之后,在他们的办公室里面挂了这么一个条幅,贴身肉搏、刺刀见红、誓比某钢便宜100块。</p><p>所以央行的副行长刘士余把这些企业叫做僵尸企业,他们在消耗着大量的金融资源,他们将给我们的实体经济带来不可估量的风险。而这些企业到目前为止还在扩张。</p><p>前两天我收到一个小女孩的信,她跟我说,他们家边上的焦化厂,到现在还要扩建,要把他们家的房子拆掉,她爸不同意,然后她爸就被打伤了,她希望我能帮帮她。</p><p>这个女孩就是十年前那个看不到星星和白云的王慧卿。我问她你现在身体怎么样,她说阿姨我现在顾不上身体,我现在只想让我们全家能够有个住的地方。还好前两天我给她打电话,她说那个焦化厂的老板刚刚被纪委带走了。</p><p>但是这些已经被认为是严重过剩的行业,还有将近一半的省份在十二五规划当中仍然把他们列为支柱产业,理由也很简单,如果你城镇化还要发展,只要提升1%,这东西都能消化得了,只要你盖房子就需要钢铁,只要你修路就需要跑车,不是吗?</p><p>清华的江亿院士告诉我,中国的城市也已经到了严重过剩的程度,甚至比工业有过之而无不及。我不太相信,因为我觉得30年来,像我们这样普通家庭的变化,和对生活的改善,都是依靠城市化的建设来进行的。</p><p>我们家就是这样,城市化的建设不断改善着生活条件。</p><p>他问我,你是不是好几年没回山西了,我说对,他说你去看看吧。在今年我就带着孩子回一趟山西,沿途用手机拍下这段视频。</p><p>我为我的家乡能够有一些变化感到特别高兴,因为那里的人们需要富裕,需要发展。</p><p>但让我意外的是,我在这个县城里看到了那么多的房地产广告和那么多的楼盘,但是他们空着。我就问我一个亲戚,他投资了当地的房地产,我说这个楼盘到底卖出去多少,他说两三成吧,我说那为什么卖的少呢?他犹豫了一下说,好像官员实名制之后就卖不出去了。</p><p>他就把我们全家人安置进了当地的一家酒店里面,我一到酒店门口差点吓回来了,因为它是一个声称五星级酒店,而且是安排了一个总统套间,我说千万不要,他说没关系,才两百多块钱。</p><p>我就开车,我们家开车进这个停车场,这个停车场一盏灯都没有,乌漆抹黑,然后就颤巍巍过来一个大爷,拿了一个手电筒给我照着,然后一路又把我们送进酒店大堂,然后又照着手电筒,把我们送过走廊,送进房间,整个酒店连灯都不开,因为没有人住。</p><p>在全国每天要消失80个自然村,我们房屋的平均寿命只有30年,在中国一共有200多家地级市,其中184家要建成国际大都市,我们现在有14亿人,但是把这些城市规划的人口加起来,现在已经超过了34个亿。</p><p>1998年,我带着一个小箱子来到北京,在这儿上学,找到工作,我在这个城市找到我自己的价值跟归属,如果没有中国的城市化,我现在应该是在山西,我爸妈费劲千辛万苦给我找一个单位,然后我戴着蓝袖套,打着算盘渡过余生。</p><p>城市给了我们个人的自由,也给了这个国家30年来的繁荣,未来还会有三四亿人要进城,这个必然要到来,不可避免,他们会给这个国家带来不可思议的文明和财富。</p><p>但是假如用投资拉动工业和拉动城市发展的模式不改变的话,结果会是什么?</p><p>国务院发展研究中心做的测算是15年之后,我们的耗煤量会达到60亿吨,我们的用车辆会达到4亿辆,清华的倪维斗院士告诉我说,这意味着我们将在用光所有资源之前我们就用光所有的环境容量,这意味着在中国雾霾还只是刚刚开始,堵车只是刚刚开始。</p><p><strong>7、我们怎么办?</strong></p><p>APEC期间,我先生有一天早上带我去小的时候他常去的地方,他父亲在那里教会他滑冰、游泳、钓鱼。冬天的时候他说他最喜欢看着冰纹一轮轮向远去,一直到远处,故宫的角楼,那是这个城市古老的优雅之美。</p><p>我们俩看着这个景色,心情特别像小孩看着最后一颗糖,不吃你还怕它化了,吃了你就知道它没有了。那种又甜蜜、又忧愁、又气急败坏的感觉。</p><p>回来之后我就问了一下中科院的专家,我说您能告诉我,如果我们想要留住APEC蓝的话我们要付出多大代价吗?他说我们必须比2013年减排这么多二氧化硫,这么多氮氧化物,和这么多PM2.5,简单地说我们要减掉一半以上的污染物,才可能得到蓝天。</p><p>我们采访了发改委副主任解振华:</p><blockquote><p>APEC会议上,中美联合声明,中国政府承诺,2030年左右碳排放到峰值,非化石能源在能源结构中占20%。我国的能源结构以煤为主,碳排放与雾霾同源,有协同减排效果。</p><p>解主任:2030年左右,碳排放到峰值,实际上是一个倒逼机制。要走绿色低碳发展的这条道。</p><p>记者:大家会想说,如果我为了环保,会不会真的牺牲掉了发展?</p><p>解主任:这是我们必须要还的帐,现在的现实问题还有就业、还有提高收入,那怎么办?转型么!</p><p>记者:这个转型当中,可能意味着这个产业当中有一部分人会失去工作?</p><p>解主任:我国的节能环保产业现在年产值已达37000多亿,吸纳就业人口可能是3000多万人,这也是一个产业。</p><p>记者:未来还会不会投资拉动经济增长?</p><p>解主任:必须要继续发展经济,但实现的路径、途径是绿色、低碳、循环发展。西方国家解决伦敦烟雾事件或者是大气污染需要六七十年,它有经验教训,我们这个阶段会大大的缩短。</p></blockquote><p>之前我很担心,北京还在扩张,它的汽车的量还在增加,它的污染能降低吗?但在洛杉矶,一个跟北京非常像的地方,它也是三面环山,空气条件很不好,而且它的城市规划是著名的失败案例,加上错误地错过公交发展的时机,造成机动车无约束扩张。</p><p>它曾经发生过大规模的光化学污染烟雾事件,你可以看看1950年,这是在好莱坞的商店当中卖的一种罐头:</p><p>里面装着洛杉矶的空气,它上面有一行标语叫“如果有恨的人,你就买一罐送给他”,可以想像当时生活在其中人的感受和愤懑。但是从1970年以来,洛杉矶的车辆还在上升,上升了这么多,三倍,但他们排放呢?降低了75%,他们怎么做到的?我就去趟洛杉矶,在那里观察。</p><p>片尾的这张罚单给我的印象特别深,回来之后我仔细看的时候我才发现,被处罚的是从中国运来的车,同样是因为环保造假,但这次他们被逮住之后,罚了差不多四个亿人民币,其中两个负责人还进了班房。</p><p>全世界的人性都差不多,在美国你也处处可见想要逃脱惩罚或者想要挑衅法律的人,但是如果执法严格的话,是可以管住90%的人,而如果不执法的话,就像北京市环保局说的那样,我们90%的车在全面造假。今年《大气防治法》终于要修订了,目前草案当中它赋予了中国的环保部门,可以对假车进行召回跟处罚的权利。</p><p>但是最早到9月份它才能够实施。那么在此之前,每一天夜里,此时此刻这些大量的上百万辆的卡车,还在中国大地上奔驰,当地有多少是这样的假车,当中有多少还在排放这样的颗粒物和污染,是我们每一个人都在承受的。</p><p>那么我们的《中华人民共和国大气防治法》还没有退役之前,我们能不能够使用它一次,我们能不能够不要让它以这样的面目留在历史当中?</p><p>当年美国要实行严格的标准的时候,美国的车企一样哭哭啼啼,不答应,他们就天天告环保部门,然后不愿意改,就在这个时候,有外国的生产车说,我能做到,听到这句话之后,美国车从地上一跃而起立马就跟上了,但是很可惜,他们也失去了半壁江山。</p><p>我就问这个环保部门的人,那你们会不会有种压力说,你们在打压民族汽车工业呢?</p><p>他给我的回答是说,环保不是负担,而是创新,保护落后是没有办法创新的,政府的角色是制定好标准之后,你要保证整个市场的公平竞争,竞争本身就会赢得市场。</p><p>但是中国是一个煤炭的消费量占70%的国家,在我们国家能源结构里面。我们怎么才能减掉煤呢?我们来看看伦敦的曲线:</p><p>很多人都跟我讲说伦敦要等了40年、50年才把污染治好,我们也得等这么久,但真的是这样吗?看看这儿,从他们开始治理污染的前十年,他们就把污染物降低了80%,这是一个极大的改善。</p><p>英国是用更清洁的能源,石油和天然气,他们才取得了蓝天和白云。很有趣的是,这次我们在伦敦想要找到那些关于伦敦当年煤炭污染的资料的时候,做得最全、最充分、最权威的是这样一份资料,它是来自于天然气公司的制作,然后它的名字也很简单叫《罪恶的烟囱》。天然气是靠竞争迅速地推动了市场,把它的对手取代掉了。</p><p>英国能源大臣告诉我说,你知道伦敦在刚开始改革的时候是什么样一个经济水平吗?他说那时候人均GDP跟中国是一样的,政府最重要的是不要去补贴那些,已经要被时代淘汰掉的落后和污染还有亏损的产业,你要给新兴的产业,给他们公平竞争的机会,他们会带给你惊喜。</p><p>现在世界平均的天然气占能源结构的是很高的,大概24%,因为天然气相对清洁的能源。</p><p>但中国只有5%。为什么我们这么低?从小到大,我听到的标准答案是这样的,因为我们是一个煤多、贫油、少气的国家。但真的是这样吗?真的。</p><p>这是真的,但这完整吗?</p><p>我就写信给国土部矿产资源储量评审中心的主任张大伟,他给我回复了几个数字:中国天然气探明率22%——这意味着我们还有大量的天然气埋藏在中国大地之下,黑暗不知名的深处。</p><p>石油探明率38%。</p><p>天然气地质储量9000亿立方米左右,开采1200亿立方米(左右)。</p><p>有大量的资源,我们知道它在哪?但却没有去开采,为什么会这样?</p><p>我写信又问张主任,他给我的回复还是几组数字。他说拿世界上最大的天然气生产国美国来说,它们有6300家天然气石油公司,我们有几家呢?(主要公司)3家,其中70%(资源)集中在一家手里,中石油。</p><p>美国有160家天然气管道公司,我们(主体管道)有几家呢?3家。其中70%集中在一家手里,中石油。</p><p>这是他给我的答案,如果我们开放这个市场的话,几年以后,我国天然气的探明储量每年可以翻一倍,我们天然气的产量每年也可以翻一倍。</p><p>做这次演讲,我才第一次接触到中国的能源问题。我有一个很粗浅的感受,中国的改革开放不是像我原来想的那样,像一个小鸟破壳的过程,一出来是一个完整的活泼泼的生命,它更像是一个知了蜕壳的过程,一点一点一点从原有的壳里蜕出来,而能源就是留在这个壳当中的最后一个部分。</p><p>它一直被认为是事关这个国家的安全和经济的安全,所以它一直不被当作是商品,它一直很难受到真正的市场经济规律的支配。</p><p>这个壳本身也在付出代价。它成了一个腐败高发的区域,能源局前局长刘铁男在接受庭审的时候说过一句话,要想遏制高发的腐败,就要把本来应该属于市场的权利还给市场。</p><p><strong>8、普通公民能做什么?</strong></p><p>但是在等待一个国家,建立和完善一个庞大的体系之前,我们究竟能够做什么?世界上再强大的政府,也没有办法独立治理好污染。它要依靠就是每一个像你我这样的普通人,我们的选择、我们的意志。</p><p>2003年报道非典的时候,我得到的最宝贵的经验是,只有信息公开才是一切公众参与的基础。</p><p>三年来中国政府在向公众提供空气质量的预报,同时他们也建立了世界上最为庞大的在线监测系统,对于污染企业的。我们要做的就是把这些数据用起来,不要把钱浪费掉,让数据变得被我们看到,让它放在阳光底下来晒。</p><p>有一个手机的APP,是环保组织公众与环境研究中心做的。下载了这个软件之后,在手机上你就能够看得到,距离你大概5公里左右的地方,有哪些企业正在排放污染物?</p><p>红色的都是表示他们正在超标排放,中国的6000家重点企业,排放了65%的废气。你可以向环保部门去举报,你也可以在微博当中公开它,所有这些带着蓝色痕迹的,都是被公民举报之后,他们承诺的整改日期。</p><p>当然会有一个遗憾,有一些地方的数据如果根本就不公开,怎么办?</p><p>去年一年,我参加了一个地方首部环境公众参与立法的研讨会,这次研讨会花了半年的时间,最终的结论是我们应该把这些重点企业公开,不仅如此,公民可以要求他们公开,如果他们不公开的话,我们可以按日计罚,每天最高可以罚10万。</p><p>可是,谁来诉讼呢?中国有1%的环境争议事件进入了司法程序,为什么会这么少?你可以猜猜看,在今年之前有多少家中国的公益组织是具备司法诉讼主体资格的?</p><p>就是因为当时我们的《民事诉讼法》规定,只有有关组织才可以提起诉讼,至于谁是这个有关组织?没人知道。</p><p>但是1月1日开始,新环保法已经规定了,只要你从事环境公益活动五年以上,没有违法记录,你就可以承担这个诉讼主体的资格,现在700多家环保组织,他们已经有人在试,提出的标语是“新环保法给力不?小伙伴们正在试。”</p><p>像我们这样的普通人,如果你没有时间去参加一家环保组织,那么你还可以做什么?我们跟自然之友合作,做了一个小的动画:</p><p>当我以前看到类似的宣传片的时候,我只把它们当作是一个口号。我很难想象,我在生活当中会受到感染、会愿意去尝试一下。之前,我路过家门口裸露的工地的时候,我经常就是捂着鼻子就过去了,但现在我知道说在北京PM2.5当中至少15%左右是来自于扬尘的时候,我决定试一试:</p><p>前后这个过程不超过5分钟,然后我临走的时候,那个让我找他们大领导的工人对我说,你知道我们领导怕你什么吗?他指着我手里的手机,他怕你公开。</p><p>我们家楼下有个餐馆是做肉饼的,然后每到吃饭的时候,这个家里头和身上都是油烟的问题,以前我不觉得有什么问题,世界上中餐馆不都这样吗?但后来我知道,在北京餐饮带来的PM2.5的污染占到至少6%,而北京的餐饮业占到全国的12%,这么大量的一个餐馆。以前我觉得没有办法,后来等我到伦敦之后,去中餐馆吃饭的时候,我发现他们那没有味,为什么?这个老板就把我带进了他们的阁楼,然后看到了一个东西:油烟净化器,伦敦政府要求他们强制安装,而且有人定期来检查。</p><p>我觉得这个是不是高科技啊?后来我回北京一问说不光北京有,而且北京是要求餐馆装的,我就打了12369举报的电话,然后一个星期之后,我们家小区里面运来了这么个东西:</p><p>前后不超过半个小时。我拍的时候,这个餐馆老板就过来了,说谢谢您这个监督,欢迎您以后多吃我们肉饼。这一年我们的工作餐吃的全是肉饼。</p><p>有的时候我路过楼下加油站的时候,我们楼下就有加油站,我会闻到很强的那个汽油呛人的味道,但我以为全世界加油站都是这样,一直到我在洛杉矶的加油站,我在他们的油枪上看到了这个东西,一种叫油气回收装置的东西,我以为这是什么高科技?但回到北京我才发现,我们早就有了,2008年之后都要求装,只不过有的人根本就不维修,才让它排放出来很多气体。然后,我也给12369打了一个电话:</p><p>虽然他承诺100%会来,真的会来吗?我们不知道,但我们把这句话放在这里,视为一个检验,我们可以记住这几个数字,12369,如果你不打,它就永远只是一个数字。</p><p>我说过我这一年过得很不痛快,老觉得是在过一种很临时的生活,真的。雾霾天一来我就不知道我明天在哪?或者未来在哪?但是呢就在那个餐馆把那个油烟装好的那一会儿,我突然觉得我好像脚落实地,这种感觉很难说清楚,你明明知道说它对于改善大气污染作用是非常微乎其微的,但就是因为一个人知道了自己做的一点点事情,可以让事情本身变的更好,他心里面就能够踏实了。</p><p>回来来看,人类与污染之间的战争,历史就是这样创造的,就是千千万万个普通人,有一天他们会说不,我不满意,我不想等待,我也不再推诿,我要站起来做一点什么,我要做的事情就在此时,就在此刻,就在此地,就是此身。</p><p>如果你没有时间做更多的事情,在雾霾严重的时候,至少你可以做一件事,就是保护好你自己和你爱的人。</p><p>我会想起我女儿生病的时候,我所有担心失去她的恐惧,还有想要保护她的全部愿望,我希望所有的母亲都不必经历这一刻的感受。她现在已经痊愈,活泼、健康,还是胖乎乎的,特别喜欢大自然和小动物。但很多时候我没法带她出门,她呢就在自己的小花盆里面养了一只蜗牛,每天早上她醒来的时候,就跑到花盆那冲着蜗牛吹一口气,她觉得这样能帮着小蜗牛长大,她是那么地喜欢这个世界。</p><p>成千上万的孩子正在孕育、正在出生,这些河流、天空、大地是应该属于他们的,我们没有权利只知消费、不知克制,我们没有权利只知抱怨、不知建设,我们有责任向他们证明,一个被能源照亮的世界,同时可以是洁净和美好的。</p><p>每次在夜空中,看到这颗星球孤独旋转,我心中都会有一种难以名状的依恋和亲切,将来有一天我会离开这个世界,但是我的孩子还在其中生活,这个世界就与我有关。所以我才凝视它,就像我凝视你,所以我才守护它,就像我在守护你。</p><p><strong>(完)</strong></p>"},{"title":"法治社会和中国传统儒家社会的矛盾","path":"2016/01/20/lawtradition/","permalink":"http://sx349.github.io/2016/01/20/lawtradition/","content":"<p>如果用“法治”来描述西方社会,那么描述中国传统儒家社会,用“礼治”就再合适不过了。</p><p>的确,中国自古以来也有法律,然而中国的法律是作为底线的存在,触犯法律简直是一种耻辱,对于大家族的子弟,如果触犯法律,不仅仅是个人行为的不检点,对于其家族的声名也是一种损害,倘若不是家族的礼教做不到位,子弟何曾需要法律来规范其行为?所以,在中国传统的儒家社会中,“礼”是重于“法”的,“礼治”才是传统儒家社会的主流。<br>单看“法治”和“礼治”这两个词,两者都包含一个“治”字。“治”是指治理,而“治”的结果,无论是西方社会还是中国传统社会,都是整个社会的稳定运行。为了更好地理解这两者,我们不妨先来进行一场头脑风暴:现在,社会上的每个人都是一个点,这个点不包含你的生理特性,只包含你作为社会人的部分——你的身份、地位、家族与朋辈关系等等;然后,把任意两个存在某种社会关系的人连上线——父母与孩子、上级与下级以及亲属或朋友,甚至是两个偶然相遇的陌生人等等——于是我们就得到了整个社会的关系网络,而从这个角度来说,社会的稳定运行,就是要维护这个关系网络的稳固。</p><p>在做到这一点之前,先要解决的问题是:怎样的关系网络算作稳固呢?<br>一种想法是,社会关系中的每一条连线是稳固的,也就是两个人之间的关系是稳定的。这并不是说两个人之间的关系不会发生变化,毕竟无论朋友还是亲人都总会有聚散离合,而是说这种关系遵循着固有的模式,不会出现突然的崩裂。<br>在我看来,这就是法治社会所体现的思想。法治社会中,有三件事物环环相扣:个人权利、契约精神和法律。法律根植于个人权利和契约精神所构建的社会,又反过来保障个人权利和契约精神的实现。而无论是契约精神还是个人权利,都是在整个社会的个体与个体交际层面的重要要素,契约最终要落实到由具体的社会关系相连的个体,个人权利本质上是个体与社会或者社会中的其他个体发生交际时的权利,所以法治社会最终的关注焦点必然落在社会微观层面的每个角落、每个局部的稳定。简单粗暴地说,只要用法律保证了社会关系中任意两个个体之间的关系稳固,那么整个社会网络,无论呈现出一种什么样的形式,它总是稳固的。如此一来,也就不难理解为何法治社会往往倾向于代议制和程序正义了——代议制保证了选民和地方议员之间的有效沟通,使得议员代表选民利益,保证议员和选民关系的稳定,尽管议员在国会进行投票时未必能够站在整个国家或社会的角度考虑问题;程序正义使得辩方与控方处于平等的地位,保障了公民即使罪大恶极也尚未丧失的基本权利,尽管程序正义很有可能为了避免错杀一个而放走三千。</p><p>另一种想法是,我们维持某种结构上的稳定,而容忍个别关系的不稳定。倘若拿一个三角形来做类比,前一种想法是,三角形的三条边的出现与消失,都遵循一定的规则;而后一种想法是,无论这三条边到底如何出现与消失,这三条边作为一个整体需要满足某种性质,比如任意时刻都至少有两条边存在等等。<br>中国传统儒家社会,或者说礼治社会,所体现的思想恰恰如此。在这个社会体系中,某个结点或者某条线都是不重要的,重要的是更高层次的结构。《论语•颜渊》记载,齐景公向孔子问政,孔子的回答是“君君臣臣父父子子”,这里每两字一断,前一个词作名词,后一个词作动词,直译就是:君主要像君主,臣子要像臣子,父亲要像父亲,儿子要像儿子。《论语•泰伯》记载了孔子的另一句话,即“不在其位,不谋其政”,反之就是“在其位,谋其政”。这两句话说明了同样的一个意思,即个体在社会中的身份地位直接决定了其应当遵循的道德规范,进而囊括了他的个人权利与义务。尤其值得注意的是,这种义务不是由契约产生的,所以任何个体不是为了其它个体而在实现权利或履行义务,譬如作为君主,对于最初推举他做君主的人,或是现在作为他的臣子的人,他即使履行义务,也不是因为这些人的推举或是事奉所带来的隐形的“契约”,而是在他获得君主这一身份的时候,社会所自然而赋予他的。从这个角度出发,我们自然地可以理解为什么儒家要讲究“慎独”:在你独自一人时,并没有其他人需要你履行相应的义务,然而你也应当注重自己的道德品质,因为这是由你的身份而非与他人的交际所规定的道德规范。结果是,礼治社会发展出来了一套完整的整体秩序,而个体只要遵循其在整体秩序中的司职,即使出现个体关系的崩裂也不要紧。这大致可以解释为何元朝与清朝这两个少数民族入关所创立的朝代,在经历较为迅速的汉化后,也能够获得朝代更替的“正统”地位,并且将这种权威性的“正统”继续传递下去,这正是因为其接受了儒家所制定的整体秩序,传承了礼治理念,使得这数千年中国历史并没有发生彻底性的断代。</p><p>法治与礼治之间的这种区别,我认为肇始于对于社会起源的不同理解;尤其是对于权利、权力和制度之间关系的认识,两种社会体系经历了截然不同的发展过程。<br>对于法治社会来说,个体的权利是一种自然权利,即所谓“天赋人权”,这其中包含了财产权、生命权和政治自由等等。对于个人来说,这些权利只有在产生社会关系时才有意义,而为了在社会关系中进一步保障自己的权利,法治社会采用了契约这种制度,契约的双方或多方承诺如何在各方之间转移和分配各项权利,于是才有可能由多个个体组成政治共同体,由个体通过契约将自己的自然权利让渡一部分给共同体,而共同体则依靠这些让渡的权利形成权力。从先后关系上来说,先有权利,再有制度,最后才有权力。<br>然而,中国传统儒家社会的礼治并不依赖于自然权利。儒家所谓的“仁”与“礼”,并不包括具体的道德准则,而是一种先验的理念,独立于人、事、物而存在。不妨借用老子《道德经》中的一句话:“人法地,地法天,天法道,道法自然。”“仁”与“礼”也是一种“道”,这种理念先于人类社会的存在而产生,是人类社会的整体秩序的应有之义,而人们只能去从中探求“仁”与“礼”的具体做法,所以才有了理学的“格物致知”和心学的“反求诸己”,即通过研究外界事物或内心思索,来获取“知”,即对于各种不同的身份地位所应有的道德要求和道德规范,然后才相应地明确了个体的权利与权力。从先后关系上来说,是先有制度,然后才有权利和权力。</p><p>可以说,直到19世纪末,西方帝国主义进入远东,法治社会才与儒家文化圈开始发生思想与文化的正面碰撞。在法治社会成长的各国列强,在面对传统儒家社会熏陶的东亚各国时,这种社会体系的差异使得他们很难理解后者的一些行为。例如,鲁思•本尼迪克特教授所著的《菊与刀》一书,就形象地描绘了幕府时代末期至军国主义时代的日本及其国民在美国人眼中留下的形象。<br>这本书的前身是作者应国防部要求做的一份评估报告,评估的主题是:在日本投降后,民众会有怎样的态度?美国应该相应地采取怎样的举措加以应对?作者记录下了这样一个值得注意的细节:在天皇宣读投降诏书之后,除了个别少壮派军官外,日本军队迅速地放弃了抵抗;事实上,即使在投降之前,在美国的日本战俘就仿佛已经认命了一般,采取了非常配合的态度。<br>实际上,作者提到,日本战俘的数量其实是极少的。如果日本军人陷入被围攻的境地,他们遵循武士道思想的一面总会占据上风,若不能拼死抵抗,也总会选择切腹自尽以谢天皇,往往是伤病员无法自裁才不得不投降或被俘虏。但是,在投降前一秒还是铮铮铁骨极富武士精神,后一秒就可以向美军提供情报,这样的态度转变之快不免令人咋舌。<br>作者的观点正与上文类似:尽管日本军人发誓向天皇效忠,然而其本质上并不是为了天皇这个人而在战斗。在日本人心中,天皇的身份是高度抽象化的,他是国家的象征,已经脱离了其作为人的身份,这也是二战结束后,日本人民极少认为天皇应当为战争的挑起和战败负责的原因。所以天皇代表的是一种精神或者说规范,向天皇效忠,不过是说明你发誓遵守这种道德规范,在自己所处的身份和地位上做相应适合身份和地位的事——这也是维持日本社会稳定的重要原则。</p><p>这种做法在美国人看来是古怪的,在今日的中国看来也是古怪的,正是因为当今社会在继承了儒家传统的同时,也在西风东渐中不断引入法治的观念。如果我们回到春秋战国时期,古人却未必觉得这有什么古怪。比如,春秋五霸之首的齐桓公公子小白,曾与其兄长公子纠争夺齐国国君之位,后来齐桓公取胜,鲁国杀公子纠,辅佐公子纠的两位大臣,召忽随即自杀,管仲则被囚禁送往齐国,由于鲍叔牙的推荐,齐桓公不计管仲的一箭之仇,拜其为相,最终成就齐国霸业。“召忽死之,管仲不死”,对于召忽的所作所为,古人和今人都是认同的,而对于管仲则未必:这难道不是背信弃义的做法吗?然而,《论语•宪问》中,孔子却回答:“桓公九合诸侯,不以兵车,管仲之力也。如其仁!如其仁!”然后又进一步说:“微管仲,吾其被发左衽矣。”显然,孔子持有的是赞赏的态度。<br>如果我们以契约的角度看,将很难理解为何守约者和背约者都能够得到赞赏,今人的结论只能是,要么孔子赞颂管仲是由于其辅佐齐桓公有功,避免了“披发左衽”(古人以“左衽”为蛮夷),即结果导向的功利主义,要么就进一步地走向道德虚无主义,即认为古人并不存在普遍的道德标准。<br>这些想法显然是与传统儒家背道而驰:管仲作为臣子,只不过是在尽臣子的本分,在他辅佐公子纠时射了齐桓公一箭,辅佐齐桓公的时候成就了春秋五霸的事业,可以说是人臣之极,为什么不值得称赞呢?而且我们尽可以对比一下孔子在《季氏将伐颛臾》中的说法,他不仅对季康子的行为大加挞伐,同时也责备了作为季康子家臣的冉有和季路,这正是因为季康子作为鲁国的大夫,冉有和季路作为季康子的家臣,所作所为都不符合臣子的规范。<br>对于古人的道德观念缺乏正确的认识,也危害了当今时代的道德风气。中国民众的群体性意见和舆论导向充斥着道德虚无主义的思潮,在这种不认同普遍道德标准的情况下,依赖于结果的好坏来判断社会政策或是政治与法律实践,民粹主义抬头可以说是其必然的结果。</p><p>这种矛盾不仅间接影响法律的实践,而且对于法律本身也有直接的影响。<br>我们不妨回顾一下,在传统社会中,无论是法律的制定、执行还是监督,这些权利自始至终都属于皇帝以及士人阶级,而从未归属于民众。对于民众来说,一旦法律置自己与不利地位,他们从不曾也不可能想到能够修改法律,而往往选择振臂一呼:“今亡亦死,举大计亦死,等死,死国可乎?”<br>随着法治思想进入我国,一些新的概念,比如宪法(以及立宪、修宪、违宪)、三权分立、司法独立等等也在我国出现。官方一直坚持我国的政治体制并不是三权分立,但在实际操作的层面中,行政权划归政府,立法权划归人大,司法权划归最高法院,这种做法无疑是仿照三权分立进行的。然而,斟酌我国宪法和各项法律出台时的措辞,很容易发现,立法权只是名义上属于人大,实际上则民众自古以来就只有守法的义务而没有立法的权利,民众倾向于接受法律而不是创造法律。因此,无怪乎我国的法律往往是由党委、政府这些对应于原本负责制定法律的机构先行草拟并基本拍板,最后再经由人大这最后一道关卡正式通过。对于民众来说,固然节省了心力;而对于相关机构来说,这看起来似乎也不过是他们应尽的职责;然而这里的“应”字,绝不是法治所要求的(甚至是法治所希望避免的),而恰恰是中国传统儒家社会“在其位,谋其政”的要求。<br>进一步地,则执法、司法、监督等等都难逃这一现实。时至今日,仍然有民众在法院打官司的时候请求“青天大老爷”主持公道,这般荒谬的事实,从另一个侧面说明了民众并不了解其个人权利,而仍旧是依照古训,认为事情的真相应当交由管辖的父母官进行厘清。可以想见,这样的民众也决不会认为监督是自己的义务:监查官吏的事情,难道不是应该由钦差大臣来做吗?<br>在中央政法委书记仍担任政治局常委的年代,曾经有两项举措严重影响了我国的司法公正:一项是公检法联合办案,另一项是由政府或党委负责人兼任同级公安领导。有一个现实的例子:我国的正当防卫判例相对于其他国家而言比例过少,很大程度上是因为公检法内部同气连枝,许多实际上“正当防卫”的被告是以“故意伤害”的名义逮捕并起诉的,如果宣判正当防卫而无罪,那警察和检察院岂不是抓错人、告错人了?法院本应该与公安、检察院相互独立,然而这样的事情只限于理想之中,如上所述,如果公检法联合办案,或者由政府领导人担任公安领导,法院和公安、检察院之间还怎么能够独立呢?然而,这样的举措最初显然不是为了司法不公正而做出的,其初衷终究也是好的,这用另外一个例子就很好说明:公安系统曾一度流传几句口号,比如“命案必破”,再比如“限期破案”。道理上是一样的,就公检法的身份而言,使得犯罪分子得到惩治是其应尽的义务,这是千百年来未曾改变的道理,相反,保证公民的个人权利,则是在法治社会下才出现的社会要求,两相权衡,再加上民粹主义的抬头,国家暴力机关自然会选择前者而轻视后者。</p><p>总结而言,法治社会和传统儒家社会的根本矛盾,在于个人是应当遵循社会规范行动,还是应当遵循契约精神行动。契约精神推崇个人权利,而社会规范推崇集体秩序,两者之间的矛盾冲突,对于目前中国在舆论上出现道德虚无主义思潮,在社会风气上民粹主义抬头,以及在法律上立法权旁落和司法不独立,都难辞其咎。</p><hr>"},{"title":"非理文明","path":"2016/01/02/alien/","permalink":"http://sx349.github.io/2016/01/02/alien/","content":"<p>我有一个朋友,一个曾经在CERN实习过半年的哲学系研究生。</p><p>也许你们还记得,当我还在混迹人人的时候(其实也不过是一年多以前),我曾经和他有过一次关于乌鸦的谈话。</p><p>“你看了我给你发的知乎链接了吗?”我给他发了一条微信。<br>所谓知乎链接,其实是这个问题:“如果宇宙中存在超级文明,哪些是他们必备的技能?”</p><p>“还没。无聊的问题。”<br>我似乎能够看到,在另一个手机屏幕前,他打了个长长的哈欠。<br>“有没有人提到老农说江青太坏了,吃馍馍又蘸红糖又蘸白糖?”<br>“就最高票回答的打油诗里提了一句红糖白糖,大多数回答都说到了皇帝的金扁担。”</p><p>手机终于又震了一下。<br>“看完了,还是无聊。”<br>“这不是请您老发表一下高见吗,在CERN的时候没有趁机了解一些地外文明的秘辛?”我故意调侃了一句,因为我知道这家伙最禁不起调侃了。</p><p>然而他出乎意料地沉默了许久,然后才发过来一句话:<br>“太麻烦了,你得请我喝咖啡。”</p><p>“现在你可以说了吧?”<br>此时此刻,他又一次陷在沙发里,让阳光只照到他的上半身而找不到他的下半身。<br>“如果……你是一只阿米巴,”他拿起随咖啡附赠的白砂糖,撒在了面前的小碟子上。“你的活动范围只有眼前的这个碟子。”<br>“作为一只阿米巴,你的生存目标非常简单:趋利避害。白砂糖就是利,你需要糖分来维持生命;食盐……“他似乎想找找有没有合适的替代品,“……比如说食盐,就是害,会使得你脱水而死。”</p><p>“现在,对于你这只阿米巴来说,我就是一种超级文明。”他随手撒了一把糖下去,“你知道这个世界中有糖和盐,你会想:如果放任你变异、演化,也许能够发掘出更好的摄入糖的方式,或是发现如何才能更有效地防御盐的刺激。这就是‘皇帝的金扁担’式的超级文明。”<br>“但实际上,超级文明还可以这样……”他突然拿起咖啡杯,往碟子中倒了一些,看着白砂糖慢慢融化,“阿米巴呐,真是不可预料,也许下一秒你的世界中就充满了咖啡。”<br>“到此为止,你发给我的链接中的所有回答总结完毕。”</p><p>“没啦?”我勉强跟上了他的节奏。<br>“这些只能勉强算是Level 0而已。”他摇了摇手指,“这些超级文明,我姑且称之为‘合理文明’。”<br>“所谓‘合理文明’,包含的范围很广,《哈利·波特》这样的魔法文明,《圣经》、《古兰经》以及各国神话的上帝神佛,甚至是《三体》中那些扭转物理定律、降维降光速的神级文明,都只不过是‘合理文明’。”<br>“OK,你也听了半天了,你觉得‘不合理文明’是什么呢?”他冷不丁地将问题抛了过来。<br>“‘不合理文明’?就是和‘合理文明’相对的咯……”我陷入了沉思,“魔法文明……上帝……难道是……巨石悖论?”</p><p>巨石悖论,是我们曾经谈论过的一个话题。<br>如果上帝是全能的,他能否制造出一块自己举不起来的巨石?<br>如果他制造出来了,那么这块巨石他就举不起来,他就不是全能的。<br>如果他制造不出来,那他同样不是全能的。<br>悖论是一种可怕的武器,康德用悖论建造起了绝对律令的基石,罗素则用悖论差一点摧毁了数学大厦的基石。</p><p>“Bingo!”他举起咖啡杯,象征性地向我致意了一番。“‘合理文明’与‘不合理文明’的差别,就在于逻辑上的可能性与不可能性。笛卡尔就曾经认为,神的‘完全全能’即体现在他可以做违反逻辑规则而不在人的认识范围之内的事情,这就是‘不合理文明’。”<br>“还记得薛定谔的猫吗?它处于既死又活的叠加态。现在,想像有这样一只‘笛卡尔的猫’,它既死又活——并不是叠加态!即使你观测它,它也不会坍缩成死或活中的其中一种状态。”</p><p>“还没完。”他一定是看出来我还没消化完而故意抢先一步说道,“我记得你上过数理逻辑的课了吧?”<br>我点点头:“我刚想说,如果用M表示悖论,三值逻辑已经可以处理带有悖论的逻辑了。”<br>“没错,所有不同于经典逻辑的逻辑体系,都可以支持一类新的‘不合理文明’。那么,再进一步,如果一种文明,既不适用经典逻辑,又不适用任何非经典逻辑的逻辑,那就既不是‘合理文明’,又不是‘不合理文明’——这样的文明,我称之为‘非理文明’。”<br>“‘非礼’文明……?下次取名的时候可以不用那么劲爆吗……”我不禁擦了把汗。<br>“是‘道理’的‘理’,这是从马基雅维利主义的‘非道德’类比出来的。”他摆出一副恨铁不成钢的模样,“马基雅维利将帝王之术视为与道德无关的事物,认为其既非道德也非不道德,是所谓非道德,也即amoral的,那么,如果文明本身既非合理也非不合理,称之为‘非合理’,或者‘非理’也不为过了。”</p><p>我感到自己已经无法跟上他的思维了:“那么,‘非理文明’到底是一种怎样的文明呢?”<br>他举起咖啡杯,狠狠地喝了一大口。<br>然后又是一大口。<br>然后把咖啡全部倒入了嘴里。<br>“我不知道。当然,也没有人会知道,或者说,没有人能够描述出来。”<br>“如果我描述了一种文明具体如何,我们就能判断它是‘合理文明’还是‘不合理文明’,那么它就绝对不是‘非理’文明。”</p><p>也许这就是真正的超级文明吧。</p><hr><p>(2016.01.02 @ 知乎 @ 如果宇宙中存在超级文明,哪些是他们必备的技能?)</p>"},{"title":"一切“狂欢”,不过是“与我何干”","path":"2015/12/12/socialism_jokes/","permalink":"http://sx349.github.io/2015/12/12/socialism_jokes/","content":"<hr><p>今年年初,两名日本公民被ISIS绑架,ISIS向日本政府索取两亿美元,否则将杀害人质。</p><p>面对挑衅,有两件事夺人眼球:先是人质之一汤川遥菜的父亲向日本各界道歉,表示自己的儿子给国家添麻烦了;后是日本民众用Cosplay、P图甚至在恐怖分子Twitter下刷Lovelive的方式,向恐怖分子发起“反抗”。</p><p>这两者的反差,为“娱乐至死”的现代社会作了一个绝佳的注脚:越是与事件相关的人,事件发生之后,其行为就越是拘谨,其态度就越是严肃;而那些与事件八竿子打不着的家伙,则大可以发挥其娱乐精神,为段子界添砖加瓦。</p><p>无独有偶,ISIS近日袭击巴黎,又牵出了当年极端分子由于《查理周刊》时常揶揄、讽刺伊斯兰教创始人穆罕默德与现代的穆斯林领袖而血洗《查理周刊》编辑部的旧事。但这一次,舆论并没有倒向《查理周刊》这边,因为在巴黎恐怖袭击事件十天前,这家杂志的封面赫然是一幅嘲讽俄罗斯空难的下流漫画——要知道,在空难发生当天,ISIS即发表声明认领了这次袭击。现在这个“事不关己高高挂起”的《查理周刊》,简直让看客怀疑它在恐怖袭击中到底站在哪一边。</p><p>因此,同理而论,“社会主义梗”的“狂欢”,既没有什么“前三十年否定后三十年”,也没有什么“后三十年否定前三十年”,不过是因为,本应成为“共产主义接班人”的我们,却越来越疑惑:社会主义“与我何干”?</p><p>看看这些“梗”吧!我们真的知道,在官方“唱衰”的情况下,上半年仍能够同比上涨7%的GDP是怎么计算出来的吗?社会主义与资本主义之间的差别,真的有教科书上说的那么明显吗?所谓中国特色的社会主义,又到底“特色”在哪里呢?关键的关键,GDP也好,姓“资”姓“社”也好,与我们又有多大的关系呢?</p><p>前两天知乎上有个大开脑洞的问题:某游戏中可以用两把锤子形状的武器合成为一把镰刀形状的武器,而且介绍称其用途是“惩罚政治犯”,这里面是不是在影射什么呀?我的回答是:题主还看漏了一点,它的被动技能叫“Corruption”,翻译成中文叫作“腐败”呀。</p><p>当年新闻发言人吕新华说“你懂的”,背后是一条大老虎横空出世。但无论我们懂不懂,老虎都是在吃肉的,所以,腐败不腐败,又“与我何干”呢?</p><p>也因此,文革的亲历者流沙河选择写下《锯齿啮痕录》这样的“伤痕文学”,而我们则选择“笑”对文革;香港记者张宝华选择跑得越来越快,而我们仍选择“笑”对“印点”。这不过是因为,离得那么远,我们只剩下“笑”的份儿了。</p><p>“伟光正”也好,“矮穷矬”也罢,离得远了看起来都是笑话。只不过某些“伟光正”的笑话恰好触碰了社会的敏感神经,于是给扣上“越过审查”、“叛逆精神”、“指出皇帝的新装”这样的高帽子,但说实话,那实在是高看了无能为力的我们呐。</p><hr><p>(2015.12.12 @ 清华大学清新时报 @ 《来辩 | 一切“狂欢”,不过是“与我何干”》)</p>"},{"title":"在中国征收遗产税的正当性","path":"2015/11/20/heritagetax/","permalink":"http://sx349.github.io/2015/11/20/heritagetax/","content":"<hr><p>所谓正当性,我理解为一件事是否能够得到人们的支持。我认为首先需要论述这样一个观点:任何一种政策或手段本身不具有正当性,而只能从其目的或结果上获得正当性。</p><p>举个例子,杀人是否具有正当性?如果有人说,杀人就是违反malem in se的,所以杀人不具有正当性,这显然是太过武断了。我想这时候,最好还是先问一句,是在什么样的情况下杀人?如果是为了谋财而害命,从目的上来说固然不具有正当性,但如果杀的人恰好是罪大恶极(杀人者未必知道这一点),事后人们又无法推断当时杀人者的动机,那么反而从结果上又得到了些微的正当性;如果是法院的死刑执行者处决死刑犯这样的杀人,本身就是malem in se所允许的,是由民众让渡给国家机关的权利,从目的和结果上都是符合正当性的(这里先搁置关于是否应当有死刑的讨论);如果是像近期新闻中四川宜宾首富被绑匪胁迫杀人,毕竟这里面有被胁迫的成分,我们可以认为有一定的正当性,因为从目的上来说,宜宾首富杀人符合法律规定的紧急避险原则;还有,像美国对广岛、长崎投放原子弹加速二战结束这种事,我们也完全可以辩护说二战的中止避免了更多人的伤亡,所以从目的和结果上来说,使用原子弹杀人都算的上是正当的。</p><p>那么,在衡量目的的正当性和结果的正当性时,又如何进行选择呢?显然,对于一件还没有发生的事,我们既然不知道结果,也就无法讨论其结果的正当性,而只能关注其目的的正当性;而对于一件已经发生的事,这两者都是可以讨论的,其中甚至可以有矛盾冲突。所幸在中国征收遗产税属于前一种,因此这里集中着眼于其目的的正当性。</p><hr><p>由目的的正当性推出决策的正当性,可以归纳为:我想要达到B,故我要做A,又因为B是正当的,所以A是正当的。这里面有两个关键点:其一,做A必须得能达到B;其二,B必须是正当的。这就对B的选择提出了两方面的限制:A必须能达到B,是一种技术上的限制;B必须是是正当的,是一种道德上的限制。</p><p>现在我们来做个代入,A就是在中国征收遗产税,那么B应当是什么呢?</p><p>我们当然可以选择那些显然具有正当性的目的,比如B就是让所有人生活都变得更好。无论持哪种政治哲学观点,都不会觉得让所有人生活得更好是一件不正当的事;然而显然征收遗产税达不到这一点,因为遗产税一般都是高额税,对于富人来说损失是巨大的,而给予他们的弥补可能只是政府由于多拿了钱所以有更多的民生投入,或是由于贫穷民众得到补助因此社会的不稳定因素被减少了,总的来说仍旧得不偿失。</p><p>既然在征收遗产税的情况下富人一定会有损失,那么B能不能是缩小贫富差距呢?如果政府设定较高的遗产税豁免额,使得穷人不需要缴纳遗产税,同时由于政府获得了更多税收用于改善社会的整体环境,比如投入更多在环境保护、医疗卫生、义务教育普及上,穷人是能够受益的,贫富差距看起来也缩小了。但我想,事实上我国的贫富差距会缩小得更厉害,因为富人在这样的激励下,还有另一个选择,就是将财产转移到避税港去,甚至把国籍也转移出去,留在国内的人贫富差距当然就进一步缩小了。</p><p>在“但我想”之前的部分,是我们原来的目的,这个目的是可以获得正当性的;然而通过征收遗产税,我们只能“超额”达成包括了“但我想”之后的部分,新增的这部分则会使得原本的正当性遭到损害,这是因为:缩小贫富差距之所以能够获得正当性,其主要的政治哲学依据要么是功利主义,要么是罗尔斯主义,它们都是支持大政府、支持征税、支持“劫富济贫”的。功利主义的看法是,由于边际效用递减,富人的100美元如果给了穷人,将会创造更大的总效用;罗尔斯主义的看法是,由于我们处在无知之幕背后,无法确定自己处于怎样的社会地位,因此在制定社会规则时,必须保证最差的情况最好,而通过对富人征税对穷人补贴的做法恰好能达到这一目的。然而,当富人转移财产甚至转移国籍时,会对社会造成净损失,他们既没有给政府贡献税收以帮助穷人,同时经济发展也会因为富人及其资本的逃离而受到影响,而经济发展水平可以看作生活水平的一个指数,经济发展受阻也就意味着生活水平下降(或至少是减缓上升)。</p><p>美国的遗产税政策中,有一项条款是,无论公民逝世时其财产是位于国内还是位于国外,均需要(在扣除豁免额之后)缴纳遗产税,此外,放弃美国国籍的代价非常高昂,不仅要缴纳脱籍税,而且未来财产被子女继承时,子女仍要缴纳遗产税,且此时遗产税没有豁免额。反观中国,由于中国不承认双重国籍,所以移民后自动视为放弃中国国籍,并没有“脱籍税”一说,而且中国目前在国际上的影响力尚不足以像美国一样全球追缴遗产税,如果现在开征遗产税,未必能够像我们所期望的那样缩小贫富差距。</p><hr><p>因此,我的观点是,在目前的条件下,中国如果开征遗产税,尽管可以有看似正当的目的,但是由于可行性的不足,使得其正当性遭到了一定的损害。目的上的正当性如果很模糊,就意味着,在开征以后,其结果的正当性将会很大程度上决定事后人们对于征收遗产税的评价,也就是说,结果好,那么遗产税就正当,结果坏,遗产税就不正当。</p><p>为了避免这种民粹主义式的结果,政府不应当在当下贸然开征遗产税,而应当先做好各项准备工作:财产的登记申报(不妨从官员开始),设定脱籍税,加强国籍管控,展开广泛的国际经济合作,这些措施将表明中国政府对于使用遗产税来缩小贫富差距的强有力的态度和决心,使得富人不会只因为遗产税的缘故而“逃离”中国(不过也不应当限制正常的资本流动和人员流动),再开征遗产税时才会使民众相信政府确实有缩小贫富差距的决心,即使最终结果并不如人意,起码能够保证目的上不遭受质疑,而获得部分民众的支持。在政府公信力日渐下降的今日,比起大多数目的上来说是面子工程,结果上来说浪费税收的政策,这已然是好得太多了。</p><hr>"},{"title":"民主精神,妥协精神,合作精神:对经管近几年学生会选举的思考","path":"2015/11/19/studentunion/","permalink":"http://sx349.github.io/2015/11/19/studentunion/","content":"<hr><p>我很认同这样一句话,经管是民主的试验场。曾经和其他院系的同学交流,他们往往会说,你看我们院系的学生会选举,都是愿意参与的人去竞选,如果竞选不上,那就是能力不如对方,一般也会直接留下来做副主席,经管搞成团队竞选,一人得道鸡犬升天,而且候选人在竞选时明枪暗箭,甚至反目成仇,至于吗?我一般会极有涵养的表示,确实不至于,然而有时也不免出于愤懑而回应道:你们的学生会选举,也能叫选举?那只能叫投票。</p><p>说出这样的话,我自然是有底气的。在担任院沟通中心主编期间,我和34届主席向雪银、35届主席杜金岳都约过稿,也和著名网红、32届主席杨奇函有过私下交流,又由于在伯克利交换期间与36届主席陈嘉证是室友,不仅曾彻夜长聊,交换结束之后还为其竞选摇旗呐喊;我还入选了校“饮水思源•服务社会”优秀大学生培养计划,这个计划的一大特色就是出各院系学生会主席,目前本期的同学中有5名院系主席,另有至少5名同学参与了竞选但最终落败,因此对于其他院系学生会选举我也略知一二。</p><hr><p>经管的学生会选举,我认为像议会结派甚于像总统选举。不同的候选人,就是不同的政党,这里面有党魁(候选人),有党鞭(负责筹划票数),有金主(提供必要的宣传资金),还有退居幕后但仍然发挥着影响力的大佬。每年的选举,并不是到了当年的三四月份才开始预热,到了五月份提交材料时才白热化的,而往往是从前一年选举结束之后,为第二年作准备的硝烟就已经悄然升起了。</p><p>一般而言,流程是这样的:譬如去年的选举,两名候选人各自拉起了一个派别(不妨就叫做政党吧),这两个政党无论获胜与否,在竞选甫一结束时,就已经选定了几名培养对象作为今年参选的预备候选人,培养对象首先自身要有意愿,其次在学生会要有资历(最好的当然是从部员做到部长的),最后还要能够在其他大二同学中有威望或者能够树立起威望,对于胜选团队来说,一般会安排其成为唯一一位大二年级的副主席,由于在学生会多一份资历,就会比其他潜在候选人多一些优势;同时所在政党开始新一轮动员:大三同学即将升入大四,进入校级及院级的学生会、团委、科协及各大协会的高层,既是提高影响力,又可以为大一大二同学提供便利;大二同学将提供下一届选举时的候选人及竞选班底,但是由于秋季学期有一学期的交换,所以要到春季学期开始时才有登场的机会;大一同学现在自身定位一般不会太明确,一般依托高年级同学的资源自己争取一些机会。</p><p>春季学期开始,政党正式展开运作。首先是决定候选人,每个政党都会从自己的培养对象中选择一名候选人,那些没有被政党最终推举出来的培养对象一般会服从组织安排,但也可能选择站出来另立山头;当然免不了的还有临时起意的独立候选人。其次是拉团队,独立候选人基本上拉不出团队,另立山头的要看情况,毕竟能作为培养对象其本身资历和能力也是不差的,而政党所推的候选人则基本上是能够拉起一支全方位覆盖的队伍。这里面还有个讲究,在前一年选举结束后,有些培养对象立刻和所在政党的大二同学拉好关系组好团队,相当于自己把自己推上候选人的位置,其它的培养对象没有了政党的团队支持,一般就只能无奈退出了。最后是团队中的新鲜血液,也就是前一年的大一同学,现在的大二同学,这时候就该选择站队了,换一个政党并不是不可以的,甚至有站队站了半个学期,临选举正式报名之前,因为考虑到另一个政党机会更大,而直接跳到另一个政党的。</p><p>选举在五月初报名,但游说在三四月份就开始了,由候选人带着团队轮番和几乎所有同学吃饭,一顿午饭跑三个食堂也是常事,但这一阶段独立代表的名单还未抽签决定,只能是广撒网,固定代表中学生会和团委的高年级同学形势基本明朗,只剩下各班的班长和团支书可供攻略;等到独立代表名单也确定之后,大概离正式竞选还有两周时间,这就是党鞭敦促整个团队集中火力骚扰的时候,可以说大学前几年的人脉关系,到这一天就该拉出来见见真章了。这里面有一个特别的团体,就是体特生;学生会选举大概是体特生与非体特生接触最为密切的一段时间,但是平日里本来就没什么交集,此时想要打交道也是很困难的,最后体特生一般都是看师兄脸色,这时候就看政党中是否有体育方面的大佬撑腰了。</p><p>到了投票这天,形势往往风云变幻,虽然拉票的时候代表们信誓旦旦,几方候选人都声称根据手头掌握的数据对方必败,但毕竟现场是匿名投票,你以为的死忠可能只是不堪其扰装出来的。团队并不会因为大局将定而闲着:那些承诺要投己方候选人的代表有没有来?如果来得太迟了会被拒绝进场,这时候就要靠团队成员去找;此外,如果有多于两名候选人,一般不太可能有一名候选人拿到超过一半的票数,所以必定会进入第二轮进行一对一的决战,这就意味着要争夺已被淘汰候选人的支持者,所以在此之前就要先准备好各项预案。</p><hr><p>以上大概是把竞选的整个流程梳理了一遍,接下来开始进入正题,逐一讨论题目中提到的这三个精神:</p><p>相比于其它院系的学生会选举,乃至于校会主席的选举,我认为经管的学生会选举在民主精神上都更胜一筹。这里面有三方面的因素:政党+团队式的竞选、游说和独立代表。</p><p>政党的存在,使得候选人的理念在各个年级都有其相应的支持者,保证了候选人具有群众基础;通过各年级的成员,候选人可以更好地了解各个年级对于同一政策可能的不同反应,能够更加充分地考虑各个年级的利益;最后,政党的存在意味着政策、理念有更强的可延续性,使得学生会能够培养起自身的特色。</p><p>如果候选人以独立的身份参选,会怎样呢?首先就是缺乏群众基础,有些院系嘲笑经管主席选举搞的鸡犬不宁,全民皆兵,却没注意自己院系的同学们对主席选举的关注度这么低,难道这样还叫民主吗?不夸张地说,在主席选举期间,经管的每个宿舍中至少有一个人是与选举相关的人员,不是团队成员就是独立代表,在这种情况下几乎不可能有人说自己完全不关注选举。</p><p>其次,候选人分身乏术,靠他一个人,怎么来得及了解全院系所有人的诉求呢?据我所知,有些院系的主席候选人,大概在竞选时约谈的对象就是同年级的同学及学生工作组的老师,也就是未来可能的合作对象,这样的做法真的了解了其它年级对学生会的希冀吗?</p><p>此外,就算候选人拼死拼活,一天三顿饭,每顿吃五次,终于赶在候选前聊过了所有同学,这样就能解决问题了吗?我们假定候选人根正苗红,大一进院系学生会体育部搬砖,大二担任部长不仅搞定了马杯还办了系内比赛,大三决定参选学生会主席,请问候选人真的了解体育部之外的工作吗?在竞选的时候,候选人大概只能说,我要好好发展文艺部、学习部等等,具体怎么发展,等我竞选上了找到相应的部长和副主席再说,这样的候选人真的靠谱吗?而团队式的竞选,就意味着这些问题完全可以由团队成员来解答,候选人所要做的就是任人唯贤,掌握大局,各部门的具体事宜在竞选时就可以敲定大纲,不需要许下自己都不知道能否兑现的空头支票。</p><p>说游说是民主精神的又一体现,我想别的院系的同学又要跳脚了。但是,不去游说,候选人又怎么让同学们知道自己的想法呢?虽然说候选人至少会有一张海报,但是且不谈有没有人去看这张海报,就算看了,又能从这个海报上那些假大空里面读出来些什么呢?更有甚者,我知道某院系的候选人直接把另一个院系候选人的理念照抄了一遍,真可谓是把我国自古以来“天下文章一大抄”的精神发挥到了新的境界。相比之下,游说则要有效的多,起码面对面的时候,大家都更愿意掏心窝子说话,而且有什么疑问,当面就能回答,不用在字里行间故弄什么玄虚。</p><p>至于今年竞选时出现的诸如“只给自己做加法不给自己做减法”或是手机录音等等,引起外界对于经管选举的一片嘲讽,我只想说,这有什么不对吗?难道候选人就没有缺点了,没有做错事吗?只要不是刻意造谣,摆事实讲道理地告诉代表们,某某候选人曾经做错了什么事,有哪些缺点,这比代表们从海报上看到的候选人真实得多,代表们知道的更全面难道不是更民主吗?至于手机录音,更是无可非议,我想每个院系都起码有一条要求是代表们可以监督候选人的行为,手机录音不正是体现代表们在行使自己的权利吗?反过来,因为发现对方在录音,就去妄自猜测对方要干什么坏事,于是通过舆论制裁对方,这样的行为恐怕才叫做专制吧?</p><p>最后谈谈独立代表。有些院系人数少,所有人一块儿投票,搞直接民主,这当然是可以的;经管的人数相对较多,所以搞间接民主,也算是符合经管的民情。但是这“间接民主”中的代表如果只是固定代表,就像校会主席选举或是个别院系那样,就只有团委的书记副书记,学生会的主席副主席,加上各班的班长和团支书,我觉得这还不够民主,反倒是更加精英主义的制度。毕竟,班长和团支书虽然是班级和团支部选出来的,但在选拔时,并不是因为他们能够在主席选举上代表整个班的同学,而只是因为他们愿意为同学们服务而且有能力为同学们服务;这和美国的议员不同,议员本身是身兼两职的,既是代表又是公仆。所以,通过抽签选出的独立代表,进一步地保证了学生会选举的民主性。</p><p>再来看妥协精神,我粗一看时,着实没有太理解:学生会选举是要竞争到最后一刻出结果为止的,为什么还要妥协呢?如果是说有多方候选人竞争,这时一方感到自己心有余而力不足或是实在不得民心,因而主动退出或加入另一方,这样的事情在近几年都没有发生过,唯一听闻的新闻反而是某一年竞选时,一名候选人许诺另一名候选人毕业后各大投行正式Offer随便挑,结果还是被后者拒绝了。</p><p>退一步的话,我觉得妥协精神大概就只能理解为,在竞选结束之后,获胜方能够承认自己的不足,并继承落败方的一些观点和做法,但这显然是可以做到,而且一直都被做到的。就比如去年的竞选,落败团队曾争取到海航对经管学院学生的机票优惠,获胜方后来并没有让此事不了了之,而是跟进并接手了这一项目。</p><p>如果妥协精神确实是指我的前一种理解,也即候选者应该审时度势,不要死磕到底,该放手时就放手,我想这只可能发生在事前,比如热能系和法学院的学生会选举都出现过等额选举的状况,这里面也许就有潜在候选人选择了妥协。</p><p>不过,如果把这种妥协精神的适用对象放宽一点,许多候选人团队的成员也许本身也有竞选的实力,但是他们出于种种原因放弃了自己站出来的机会而是成为另一名同学的支持者,尽管在竞选中的付出未必比候选人小,而最后收获的荣誉一定没有候选人大,但是他们还是任劳任怨团结一心,我想这也算是一种妥协精神吧。</p><p>最后看合作精神,我觉得这里面应该区分开政党内的合作和政党间的合作。政党内的合作是经管独有的特色,而且上面也论述得很清楚了,政党内的合作比起候选人独自作战有着诸多方面的优势,这里不再赘述。但是政党间的合作却着实戳到了经管学生会选举的痛处,我觉得这比妥协精神更高一个层次,就是获胜的一方不仅要取长补短,而且要接纳落败一方的成员进入团队,就好像美国总统竞选早期往往会邀请落败方的候选人担任副总统。这一传统在美国早已消逝,但是在清华其它院系的选举中却频频出镜,因为候选人愿意站出来竞选已经说明了他们在为同学们服务上有更高的觉悟,所以让他们来担任副主席确实是最佳人选。但经管在过去几年搞团队式选举时经常是在团队内已经组阁完毕,有时还会用副主席的名额来拉拢颇有影响力但在政党间摇摆的同学,直到今年通过制度改革,要求获胜团队至多只能出五名副主席,而由学生工作组、党委、团委、科协从非该团队成员中推选剩余的副主席名额,才在实质上达成了合作的结果。所以总的来说,过去几年的学生会竞选确实是缺乏这种团队间的合作精神,今年才终于达成了一个重大的突破。</p><p>除了这三种精神以外,我觉得还值得谈谈学生会选举中的现实意义,这就又回到我一开始反驳其它院系同学的那句话,经管的选举才是选举,其他院系那只能叫做投票。现实中的选举,是真的会使两个人反目成仇,老死不相往来的,是真的会有墙头草两面派,如果倒向的一方赢了是投机成功,如果倒向的一方输了只能自认倒霉,是真的会有利益诱惑和使绊子的。对于那些打算未来早早地走出象牙塔插一脚在现实中的人来说,他们会感谢在经管学生会选举中提前打好了预防针,让他们能够早有防备;而对于打算“躲进小楼成一统,管它冬夏与春秋”的人,选举条例中也给了他们放弃参与这场争斗的自由。我在大一初入学时,就对学生会竞选的惨烈有所耳闻,暗自决心绝不站队,结果到了大三交换回国,最终也未能免俗帮了室友一把,但是这段经历也让我重新体会了一回人情冷暖,看似积累已久的人脉关系,实际上异常脆弱。社会就像学生会选举时那样现实,既然你连社会都接受了,经管这样的学生会选举还有什么不好接受的呢?</p><hr>"},{"title":"Cheap Airlines Vs. Poor Services : A Dilemma?","path":"2015/10/27/discussion1026/","permalink":"http://sx349.github.io/2015/10/27/discussion1026/","content":"<p>本文是10月26日《经济学原理》小课的记录,本次小课主要讨论了弹性这一概念如何体现在航空公司的决策制定中。</p><hr><h1 id=\"弹性\"><a href=\"#弹性\" class=\"headerlink\" title=\"弹性\"></a>弹性</h1><p>我询问了一下钟老师,结果钟老师告诉我说课上他明明讲了完全弹性和完全无弹性的……所以为什么大家会告诉我没有讲……</p><p>关于课上提到的完全弹性、完全无弹性和弹性为常数的例子,我来总结一下,不过一般这种例子也只是好玩罢了,毕竟这些都是太过理想的状况:</p><ul><li>完全弹性的例子:竞争市场的供给曲线就是完全弹性的;1块钱的需求曲线,在不考虑外加因素(比如洗钱)的情况下是完全弹性的;</li><li>完全无弹性的例子:房地产市场,由于土地是有限的,当价格足够高的时候,即使商家想要提供更多的房子也会受限于土地有限,趋向于完全无弹性;需求的例子好像没有特别明显的,毕竟价格太贵了总是会让人知难而退的。</li><li>弹性为常数的例子:我在课上说的好像有些问题,实际上弹性为常数的例子有非常多,我所画的那些双曲线其实都是弹性恒为1的情况,当然有可能弹性恒为别的常数;弹性恒为1的例子,比如炫耀性消费,你就打算花300万装个逼,这时候某样东西定价是1万你就买300个,10万你就买30个,总归就是花300万;或者是逃难的时候,把钱都换成硬通货,这时候硬通货的价格和数量对你来说只要相乘等于你的全部财产就可以了,这种情况下你对某种商品的弹性恒为1.</li></ul><p>最后还是要纠正一下大家的观念,就是弹性的高低并不直接决定高低价格,比如考虑P=-Q/2+3/2和P=-2Q+3这样两条需求曲线,前者弹性大,后者弹性小,再考虑这样两种供给曲线:P=Q+1/2和P=Q-1/2,很明显前一条供给曲线导致弹性大的付低价,弹性小的付高价,而后一条供给曲线导致弹性大的付高价,弹性小的付低价。从这个角度说,根据需求质量弹性的高低来说明选择的质量高低同样是有问题的。</p><hr><h1 id=\"长期与短期\"><a href=\"#长期与短期\" class=\"headerlink\" title=\"长期与短期\"></a>长期与短期</h1><p>课上我们讨论了长期与短期的情况下的一些不同,可以说长期与短期也是经济学家热衷但也很难以面对的一个问题。就像我在课上提到的凯恩斯的名言,In the long run we are all dead,显然这样的长期是没有意义的,那么长期到底有多长短期到底有多短该如何决定,确实是值得研究的问题。</p><p>一般来说,我们认为长期弹性更大,短期弹性更小,这是课上也介绍过的内容,理解起来也很就简单:在你已经登上飞机之后,你无法再改变你携带了多少零食和饮料,而这里的长期可以认为是从你开始准备乘飞机开始到飞机落地离开机场结束,这期间你有非常多的机会去提前买好零食和饮料,想办法带到飞机上,弹性就更大。</p><p>但是在宏观部分会学到总供给总需求模型,在模型当中长期总供给曲线是垂直的(由自然失业率决定),而短期总供给曲线是倾斜的,长期的弹性变小,短期的弹性变大,这是为什么呢?</p><p>这时候我们就要重新考虑长期与短期的定义了:短期与长期的唯一差别,其实在于有些短期内变动不了的东西,长期内可以变动。在乘飞机的例子当中,不能改变的是你准备的零食和饮料的量,这导致了你的需求弹性减小,而在总供给总需求模型当中,不能改变的是粘性的工资和价格(所谓粘性,即签了合同之后,工资或价格不能随意变动),这导致了总供给曲线的需求弹性增加,所以并不是长期就弹性大短期就弹性小,而是长期可变量就多,短期可变量就少,而这进一步导致了不同的弹性变动。</p><hr><h1 id=\"马克思主义政治经济学\"><a href=\"#马克思主义政治经济学\" class=\"headerlink\" title=\"马克思主义政治经济学\"></a>马克思主义政治经济学</h1><p>这次memo当中有几位同学提到了“价格由价值决定”等马克思主义政治经济学的内容,所以也谈谈我的看法。</p><p>首先,大家在未来起码还有马克思主义基本原理(所有同学必修)和政治经济学(经济与金融专业必修)这两门课会给大家介绍这一套理论并且有相应的考试,所以大家也不用急着把高中学过的这些内容都扔了,起码先把这两门课给干掉再说。</p><p>但是,在学习经济学原理(以及未来的中微中宏及所有经济学专业课程)的过程当中,还是先把这些内容抛到一边为好,因为这两者之间真的是三观不合。就拿基本的价格理论来说,供求决定价格和价值决定价格是两个完全不同的理论,站在后者的角度上,才出现了诸如“iPhone流水线工人一个月的工资买不起一台iPhone”这种观点,从经济学原理的观点来看,这两个价格(工人工资是劳动力的“价格”)完全是两个市场的供求关系分别决定的,只有政经学者才会觉得这两者能够通过“社会必要劳动时间”联系起来。</p><p>关于一些早期马克斯恩格斯思想的问题,我在马原期中的读书报告(《共产党宣言》和《社会主义:从空想到科学的发展》)中写过(见上一篇),他们的一些认识,比如产业军的出现、小资产阶级的衰落、剥削与剩余价值以及工会的失败等等,放到今天的社会来看都是不成立的。当然现在的马克思主义也有新的发展,但是就高中政治课所学的内容来看,并没有逃出这些早期思想的窠臼。</p><hr><h1 id=\"关于垄断\"><a href=\"#关于垄断\" class=\"headerlink\" title=\"关于垄断\"></a>关于垄断</h1><p>因为垄断是大家在之后要学习的内容,我就不在这里作太多的介绍,简单地说两句。</p><p>垄断之所以与竞争市场不同,最大的区别就在于,垄断商家不担心消费者流失到其他厂家,因此它们能够以高于边际成本价格定价,而竞争市场上只能以边际成本价格定价(为什么呢?这其实是一个Bertrand价格博弈,假设边际成本是C,如果你定价为C+ε,那么我定价为C+ε/2就能掠夺全部消费者,因此最终所有人的定价都固定在C。不过不用担心他们是零利润,因为这时候的边际成本是机会成本,也就是说零利润意味着这些公司没有赚到比其他方式更多的钱,但是还是赚钱的)。</p><p>所以垄断商家应该定价在哪里呢?应该定价在边际收益曲线和边际成本曲线相交处的数量对应在需求曲线上的价格,因为这时候它多卖一单位物品(降低一点价格)的边际收益将会小于边际成本,而少卖一单位(提高一点价格)所减少的边际收益大于所减少的边际成本。更详细的解释可以参考课本~</p><p>最后还有一点要说的,就是现实生活中,其实我们碰到的商品一般都介于垄断和竞争之间,这种状态也叫垄断竞争,一个简单的理解方式就是我们有品牌忠诚度,所以商家可以相比竞争市场多收一点钱,但是又不能像垄断市场那样随便提高价格,否则当价格足够高的时候我们还是会跑到别的商家那里去。</p><hr><h1 id=\"关于画图\"><a href=\"#关于画图\" class=\"headerlink\" title=\"关于画图\"></a>关于画图</h1><p>小课最后我给大家介绍了一些画示意图的工具,这里再列举一下:</p><h2 id=\"几何画板\"><a href=\"#几何画板\" class=\"headerlink\" title=\"几何画板\"></a>几何画板</h2><p>首先推荐的一类是几何画板,主要的优点是几何画板的绘制工具比较多(比如坐标系,或是直接绘制特殊曲线等等),而且可以很方面的作交点,取特殊点等等,还有一点很重要的就是几何画板可以绘制示意图而不需要具体的数据(当然也可以绘制特定的函数或曲线,但是示意图当中一般只要绘制出大概趋势或者相对大小就可以了,往往不涉及具体数据)。</p><h2 id=\"科学计算软件\"><a href=\"#科学计算软件\" class=\"headerlink\" title=\"科学计算软件\"></a>科学计算软件</h2><p>诸如Mathematica、Matlab、Maple之类的科学计算软件,本身当然不是用来画图的,而是用来进行科学数学计算的,但是它们对于一些特定函数的绘制比几何画板更加靠谱,比如Mathematica的一些3D图象绘制、参数方程绘制,还可以实现动画演示。这一类软件对于大家的数学学习也是很有帮助的,反正我在学线代、概统的时候很多计算都是直接扔Mathematica的。</p><h2 id=\"数据统计软件\"><a href=\"#数据统计软件\" class=\"headerlink\" title=\"数据统计软件\"></a>数据统计软件</h2><p>诸如STATA、SPSS、Eviews、R、Origin等等的数据统计软件,这些最好至少掌握一种,因为大家在学计量经济学的时候会用到,普通推荐STATA,有编程基础或者想要挑战自我的强烈推荐R。这些数据统计软件在图表绘制上比较有优势,就是你有许多数据然后做图像这类,而且可以加各种TAG各种水平线竖直线,像STATA可以通过命令直接设置,也可以先作图然后手动调。</p><h2 id=\"其它软件\"><a href=\"#其它软件\" class=\"headerlink\" title=\"其它软件\"></a>其它软件</h2><p>其它软件其实分为两类:一类是更差的,比如画图,或者直接用WORD自带的绘图工具,如果要求并不是很高,使用这些也完全没有问题;另一类是更好的,那种一般就是与特定的功能实现相关的(比如我在某微信群里看到学长在讨论神经网络的绘制软件),我自己接触的也不是很多,如果大家以后有志于参与数学建模比赛,可以参考这方面的资料。</p><hr>"},{"title":"中国房地产市场调控:历史与现实","path":"2015/10/19/discussion1019/","permalink":"http://sx349.github.io/2015/10/19/discussion1019/","content":"<p>本文是10月19日《经济学原理》小课的记录,本次小课主要讨论了房地产价格居高不下的一些可能原因:</p><hr><h1 id=\"国民经济支柱产业\"><a href=\"#国民经济支柱产业\" class=\"headerlink\" title=\"国民经济支柱产业\"></a>国民经济支柱产业</h1><p>房地产价格居高不下的其中一个原因是房地产行业被认为是国民经济支柱产业。在我的理解来说,房地产的这个地位,主要来自于房地产和许多重要工业品之间有着非常密切的联系。直接来看,房地产联系到施工原料(水泥、黄沙、钢筋、混凝土)和装饰材料等,间接地影响了钢铁行业、化工行业进而煤炭、电力等能源行业等。实际上,房价一旦低迷,全国的工业品价格指数也会下跌。</p><p>总的来说,政府的动机是复杂的。尽管我们可以说,在一党制的国家,政府并没有下台的问题,因此也没有改善政策的动机。然而,民意确实是政府需要考虑的重要部分(不过维稳也未尝不是一种应对民意的方法)。在我国(以及许多国家),民众的政治参与可以说是“冷感的”,如果吃饱穿暖,他们对于一些无关的政策,或者一些虽然有关但是短期内不太可能发生在自己身上的政策的态度基本上是冷漠的。而保证民众吃饱穿暖的要素之一就是国家的经济发展。只要国家仍在发展,民众对于房地产高价的呼声也只能是三天打鱼两天晒网,该贵贵该买买。所以限购到了一定时间,“缓和了房价上涨势头”(上涨=一阶导数,势头=二阶导数,缓和=三阶导数,真不容易),很快就要放开。就我个人看来,这是房地产不太可能走衰的主要原因。</p><hr><h1 id=\"投资与投机\"><a href=\"#投资与投机\" class=\"headerlink\" title=\"投资与投机\"></a>投资与投机</h1><p>另一个比较重要的原因(张木助教认为这一项才是主要原因),就是民众使用房地产进行投资与投机的需要。</p><p>在中国,金融市场长期以来都很不完善;相应的,居民的储蓄率非常高,这意味着有大量的资金希望找到理财的渠道。股市的风险,毋庸多说,其它金融产品的投资门槛较高,而且也是近几年才开发出来的,那么居民最佳的投资选择就是房地产。</p><p>实际上,这也算得上是一个“自我实现的信念”,房价越高,人们越愿意通过买房的方式进行投资,房价就更高,不断循环往复,而且,不像股市偶尔会有国家队来收割(比如今年的股灾),房价在预期中始终是保持上涨势头,或者起码是不会下跌的。(这也就是为什么我认为上面那项才是主要原因,因为如果不是政府不能承受房价下跌带来的经济减缓,那么房价可能忽涨忽跌,那么民众也不会希望利用房地产进行投资与投机)</p><hr><h1 id=\"临时性的调控措施\"><a href=\"#临时性的调控措施\" class=\"headerlink\" title=\"临时性的调控措施\"></a>临时性的调控措施</h1><p>还是基于第一项所说,由于房价不能跌,所以调控措施的短期性、临时性在其开始实施时已成定局。这也就意味着,这些调控措施只可能影响那些在短期内有买卖房屋需要的人,比如孩子马上要上学了,再比如要结婚了,还有那些需要转手房子以弥补资金短缺的生意人。(这让我想到房地产行业影响经济发展的可能,一旦房地产价格下跌,注意到许多公司贷款是拿房地产作抵押的,房地产价格下跌,已经贷款的公司有更高的激励欠账不还,而没有贷款的公司则更有可能贷不到款,相当于把更多坏的公司留在市面上,同时阻碍了好的公司发展,但是这个途径有些太过间接。)</p><p>预期对于经济具有很强有力的作用,尤其在宏观(如果大家都预期到要通货膨胀,那么通货膨胀不应该造成任何影响)和金融(如果大家都预期股票明天会涨,那么股票今天就会涨)有许多经典的案例和模型。</p><hr><h1 id=\"社会观念\"><a href=\"#社会观念\" class=\"headerlink\" title=\"社会观念\"></a>社会观念</h1><p>中国人的社会观念涉及房子的有许多,这里我说一条,就是关于买方和租房的偏好。基本上,中国人“家”的观念使得我们在同等条件下偏好买房(比如婚房的需求,比如落户口的需求),这会导致什么结果呢?</p><p>如果没有这种偏好,我们可以假定,在买房市场和租房市场的价格之间应该达到平衡(这么想,买房就意味着省下了未来的租房费用,如果其中一个更贵,会有人从贵的市场转向便宜的市场,最终买房付出的钱应该和租房付出的钱(在折现的情况下)一致),然而由于我们在同等条件下偏好买房,就会使得买房的需求右移,租房的需求左移,买房的价格更高,租房的价格更低。</p><hr><h1 id=\"供需错位\"><a href=\"#供需错位\" class=\"headerlink\" title=\"供需错位\"></a>供需错位</h1><p>在案例中提到,实际上目前中国的空置房数量很多(不过,钟老师认为案例中的空置房估算方式是完全错误的,而且他认为只有开发商没卖出去的空置房有意义,买了而不住的房子不能算空置房,毕竟属于私人财产),如果考虑总共造了多少房子和总共需要多少房子,可能两者的数目相差无几。也就是说,加入现在所有的房子收归政府所有,再由政府分配的话,应该每个家庭都能有房住。</p><p>然而,这种想法是以全国计的,而事实上,当我们说房价高的时候,我们并不是说全国性的房价高,而是我们愿意住的地方房价高。因为大部分人都想要去北上广这样的大城市,所以这样的大城市的需求就大,而小城市和县城的房子需求就小,因此大部分人觉得房价很贵。就好比说,现在北京一间公寓的价钱,可能在偏远地区的县城能够买一条街,但是大部分人都会选择前者而非后者。这里面的问题还是在于,中国目前的地域发展不平衡,人们从西北地区往东部沿海流动,这些地方的房屋供给终究是有限的。前两年信托很热的时候,出现了许多房地产项目甚至是政府合作项目,大多是三四线城市建房,可能也是希望能带动当地经济发展,但我觉得这未免有些本末倒置了,只有经济发展了才会有人来买房才对。<br>不过,区域发展不均衡是比房地产价格居高不下更加宏大的问题,在这里就不进一步讨论了。</p><hr><h1 id=\"房产税\"><a href=\"#房产税\" class=\"headerlink\" title=\"房产税\"></a>房产税</h1><p>房产税其实是控制房价普遍采用的一种手段,但是在国内经常被吐槽,主要就是因为目前我国的土地使用权70年限期使得民众认为这房子都不是我的为什么还要交税(当然这个看法有一定的问题,毕竟房屋所有权还是归业主的)。还有一个问题就是,适用房产税的国家一般也是民众愿意租房生活的国家,而在我国拥有自己的房子的观念一时半会儿还是很难破除的,因此会受到民众很大的阻力。</p><p>站在民众的角度,对于税这种东西,普遍上还是抱着“既要马儿跑又要马儿不吃草”的态度,对于不清楚税制的人,可能觉得国家近几年的一些调控手段还是不错的,比如逐步提高个人所得税起征点等等,但是然并卵,毕竟我国目前最大的税收来源是营业税、增值税等等,你只要买东西,都是在承担纳税义务,所以民众反对增加新的税种也是无可厚非的。</p><p>而且房产税是否能够起效还很难说,加州的房产税实施得也很严格,但挡不住需求高啊,美国是正宗的学区房制度,在伯克利交换时碰到一对清华女北大男的夫妇,就在纠结Santa Barbra的房价太贵,但是又不得不住那边,每天开车上下班,因为那里的公立学校比较好。所以在不收房产税和收房产税但是可能没有抑制房价反而提高房价当中选,倾向前者也是很自然的。</p><hr><h1 id=\"新加坡组屋制度介绍\"><a href=\"#新加坡组屋制度介绍\" class=\"headerlink\" title=\"新加坡组屋制度介绍\"></a>新加坡组屋制度介绍</h1><p>关于新加坡组屋制度,可以参见<a href=\"http://news.163.com/special/00012Q9L/singaporehouse.html,专题报道非常全面。\" target=\"_blank\" rel=\"noopener\">http://news.163.com/special/00012Q9L/singaporehouse.html,专题报道非常全面。</a></p><p>我的看法是:其一,组屋制度说明目前保障房和商品房双轨制本身具有合理性,让商品房的价格随市场波动,但保障房由国家来承担,这是我国未来的发展方向;其二,但是,新加坡能够实施组屋制度,与其国家体制(李氏家族+人民行动党=强有力的控制)以及国家性质(城市国家,总面积不及香港和浦东新区)高度相关,而在中国,中央政府和地方政府本身就存在利益矛盾(尤其是土地税收这一块),其次幅员辽阔,规划很难精细,这意味着我们不可能像新加坡那样做50年的长期规划(连地级市的十年计划都制定不好,这是值得反思的)。</p><hr>"},{"title":"离开中国制造的一年:一个美国家庭的生活历险","path":"2015/10/12/discussion1012/","permalink":"http://sx349.github.io/2015/10/12/discussion1012/","content":"<p>本文是10月12日《经济学原理》小课的记录,本次小课主要讨论了比较优势原理的一些问题。</p><hr><h1 id=\"比较优势原理\"><a href=\"#比较优势原理\" class=\"headerlink\" title=\"比较优势原理\"></a>比较优势原理</h1><p>比较优势原理是一个很简明易懂的原理,它指出,生产者应该各自生产具有比较优势的产品,并且通过交易来使所有人变得更好。但是,当把比较优势原理应用到国家与国家之间的时候,会出现一些与现实不相符的现象,这是由于比较优势原理的一些假设和国家层面的生产与交易不符所导致的。</p><h2 id=\"假设一:代表劳动力\"><a href=\"#假设一:代表劳动力\" class=\"headerlink\" title=\"假设一:代表劳动力\"></a>假设一:代表劳动力</h2><p>可以发现,比较优势原理本来是应用在一个人身上的,这个人在一天8小时可以生产10斤柴或20斤鱼,我们可以很轻易地让一个人决定今天砍多久的柴和捕多久的鱼,来调整他实际生产的柴和鱼的数量。</p><p>然而,当比较优势原理作用在国家上时,我们实际上是把一个国家的劳动力抽象成了一个代表劳动力,而这种类比是有问题的:一个国家不可能突然让所有的劳动力突然都转向一种行业,有些年纪大的工人就不可能再去学校学习然后进入金融领域。那么,一方面培训工人、重新安置工人需要政府财政来支持,另一方面,如果不能培训、不能安置,这些人的失业会冲击国家的政局。退一步说,假定我们让这些员工成功转型了,一个地区的基础设施建设未必就能立刻转型,伴随而来的是该地区的劳动力纷纷外流,那么经济的下滑将是不可避免的,可作为代表的就是美国著名的汽车城底特律。</p><p>有同学在memo中提到了卖馒头和卖衣服的商人的问题,让我回想起我在大一时写memo提到了美国工人和美国农民的说法:如果中国在工业上对美国有比较优势,那么就应该中国人生产工业品,美国人种粮食,但是美国农民在种粮食上相比于美国工人有比较优势,那么就应该美国农民来种粮食,那么美国工人怎么办?实际上,这里的问题就在于,比较优势认为美国工人也会马上去种粮食,而这显然是不太现实的。</p><p>(顺带一提,关于基础设施建设,最近读到一篇论文,研究了旧金山1906年大火之后的经济快速增长现象,作者提出了一个观点,即当时城市的资本(基础设施)都是陈旧的,并不利于生产,但是一场大火烧掉了这些陈旧的设施,使得新的工厂、新的机器得以进入,反而带动了经济的增长,这不禁又让我想到《三体》中章北海选择杀掉那些支持化学工质的老航天人,以争取无工质火箭的新生的场景。)</p><h2 id=\"假设二:同质商品\"><a href=\"#假设二:同质商品\" class=\"headerlink\" title=\"假设二:同质商品\"></a>假设二:同质商品</h2><p>比较优势原理中,没有明说但却至关重要的一点是,所有进行交易的货物都是同质商品;张顺捕的鱼和李逵捕的鱼是同样的鱼,而张顺砍的柴和李逵砍的柴也是一样的柴。否则,如果张顺捕的鱼李逵不爱吃,那么双方就无法进行交易,也无法通过分工使得双方更好。</p><p>因此,即使一个国家对于某种商品的普通型具有比较优势(假如说中国地大物博,牛肉也非常多,生产牛肉具有比较优势),但是这类商品种的“VIP型”(比如日本神户牛肉)就不会因为比较优势原理而停止生产,因为它们实际上并不是同一种商品。(也可以类比之前天猫和淘宝的例子)</p><p>用术语来说,比较优势原理要求所有商品均处于完全竞争市场,否则结论可能并不成立。</p><h2 id=\"假设三:无外部性\"><a href=\"#假设三:无外部性\" class=\"headerlink\" title=\"假设三:无外部性\"></a>假设三:无外部性</h2><p>所谓外部性,就是在交易中没有考虑到的其它影响。正外部性就是好的影响,比如个人接受教育的时候,他考虑的是自己的投入和回报,而不会考虑到因为他接受教育可能让社会变得更好,这就是正外部性;负外部性就是坏的影响,比如企业生产的时候产生污染,如果这是一个只顾自己利益的企业,而且它排污没有成本的话,在决定自己生产多少产品的时候,它就不会把污染对社会造成的影响算进去,这就是负外部性。</p><p>Lawrence Summers曾经宣称,非洲人民需要钱,我们(美国人)需要干净的环境,所以把重污染的产业搬到非洲去无可厚非,各取所需。然而这样的说法很快遭到诟病。从比较优势的角度来说,如果非洲在生产工业品上对美国有比较优势是因为当地可以选择重污染、低成本的生产方式,那么我们就未必应该把产业转移到非洲。接下来的课程中会讲到,面临负外部性的问题,政府可以考虑采用征税的方式,一旦对污染进行征税,可能非洲的生产方式成本更高。</p><h2 id=\"假设四:单一投入\"><a href=\"#假设四:单一投入\" class=\"headerlink\" title=\"假设四:单一投入\"></a>假设四:单一投入</h2><p>在张顺李逵的故事中,两人的投入都是时间;而在国家与国家的贸易中,两个国家的投入都是劳动力;都假设只有一种投入。但是,即使是最基本的生产函数,我们一般也要假设有两个生产要素:资本和劳动力;那么,一旦考虑到多个维度的投入,如何定义比较优势呢?甚至可能连以另一种商品为单位衡量的机会成本都没有明确的定义:假如张顺用一小时时间、花五两银子(买鱼竿)可以抓8条鱼,而用一小时时间、花十两银子(买斧头)可以砍3斤柴,生产函数我表示成f1(x,y)=8(x/1+y/5),f2(x,y)=3(x/1+y/10),那么一条鱼该值几斤柴?</p><h2 id=\"假设五:交易成本\"><a href=\"#假设五:交易成本\" class=\"headerlink\" title=\"假设五:交易成本\"></a>假设五:交易成本</h2><p>农民种的大葱,从地里割了之后,在边上摆摊卖1毛钱,那么运到城里卖多少钱?起码我们可以想见,因为花了交通成本,肯定得卖的贵点。同样道理,当国家与国家之间进行贸易时,有些产品的贸易成本非常昂贵,以至于对于买家来说,出口国已经不再有比较优势了,那么就会选择在本国进行购买。典型的例子就是食品,我在美国买苹果的价钱是一个苹果0.9刀,在国内这个价钱都可以买一斤苹果了,但是显然我们不太可能要求从中国进口苹果(除非是上面假设二提到的特产苹果),因为运苹果过来实在划不来。</p><p>除了这种运输成本外,税收也是一种潜在的贸易成本:一个国家可能对另一个国家征收惩罚性关税(比如美国的“反倾销”),而另一个国家也可以选择出口退税(比如中国的纺织品),这种做法会导致比较优势(在买家看来)发生变化,进而影响双方的行为。</p><h2 id=\"假设六:市场反应\"><a href=\"#假设六:市场反应\" class=\"headerlink\" title=\"假设六:市场反应\"></a>假设六:市场反应</h2><p>虽然我很讨厌政治经济学那一套东西,但是起码这一点它们没说错:市场的反应是有滞后的。就比较优势而言,有些比较优势是基于以往的市场反应给出的,但是未必就是正确的,比如像看天吃饭的农业,也许气候不错的年份有比较优势,气候较差的年份就没有比较优势了,但是目前的气象预测手段还不能让我们知道今年回事怎样的年景,那么这样的比较优势就是不牢靠的。</p><p>还有一种情况,就是根本就没有以往的经验,那么我们也不知道真实的机会成本是怎样的,比如一些新兴行业,这时候国家可能会选择先进行贸易保护,静观其变。</p><h2 id=\"假设七:劳动力流动\"><a href=\"#假设七:劳动力流动\" class=\"headerlink\" title=\"假设七:劳动力流动\"></a>假设七:劳动力流动</h2><p>如果劳动力进行流动会怎样呢?用美国工人和美国农民的例子,如果美国人在种粮食上有比较优势,那么美国工人可以选择符合国家的比较优势,转而学习如何种粮食,也可以选择到中国去制造工业品,关键是看他到底能不能转型成功,以及两种选择的收益如何。如果他确实不太可能转型成功(即种粮食这个选择收益为0),那么他就会选择到中国去制造工业品,长此以往,中国在生产工业品上的比较优势会越来越强。这个结果可能在效率上并没有损失(甚至是有帮助的,因为本来这些人就没有办法生产别的东西了),但是对于国家来说显然是不情愿的,这意味着选择了制造业的国家将会永远被捆在制造业当中,而不能进一步进行产业转型。所以现实中,劳动力的流动一般是有门槛的。</p><hr><h1 id=\"通货膨胀\"><a href=\"#通货膨胀\" class=\"headerlink\" title=\"通货膨胀\"></a>通货膨胀</h1><p>关于通货膨胀,有些同学提到通货膨胀的原因是超发货币,这本身是没错的。但是一般而言,我们可以单纯地认为“通货膨胀”就是表述了一个现象,即物价的整体上涨。在放开贸易之后,中国的低价工业品进入美国,使得美国的整体物价降低了,因此我们说美国的通货膨胀被抑制了,这种说法就理解为美国的整体物价上涨被减缓了,并不需要联系到货币的理论。</p><p>有同学提到了另外一种说法,就是由于生产减少,国家为了刺激经济、降低失业率而选择超发货币刺激经济(比如中国的四万亿),结果导致通货膨胀,这一点是正确的,在宏观部分我们会提到菲利普斯曲线,在短期内政府面临着失业率和通货膨胀的tradeoff,但是从长期来看,失业率会稳定在自然失业率,因此依靠短期内通货膨胀降低失业率的做法最后有可能导致失业率回到自然失业率而通货膨胀程度加剧。根据《经济学原理》的说法,美联储在上世纪就成功地通过适度的通货紧缩与提高失业率,使得美国顺利地在保持失业率基本不变的情况下降低了通货膨胀的程度。</p><hr><h1 id=\"贸易保护\"><a href=\"#贸易保护\" class=\"headerlink\" title=\"贸易保护\"></a>贸易保护</h1><p>如果比较优势是好的,那么为什么美国人要抵制中国货?再比如中国人要抵制日货呢?除了上面说的比较优势本身的问题之外,许多贸易保护无外乎两大理由:</p><p>其一,国家安全:许多行业是事关国家命脉的,比如直接有关的军工行业,不太可能依靠别的国家(尽管我们也有在向外出售武器,但是我们也尽量不买别人的武器,最起码这种武器不能只向别人买,自己也得逐渐能够制造,比如航母);再比如作为战略储备的矿产行业和能源行业,粮食也算搭点边。中国之前一度非常慷慨地出售稀土矿,后来慢慢地收紧了,尽管中国始终是稀土矿蕴含量最大的国家。</p><p>其二,政治诉求:由于上面说的假设一提到的情况,许多时候一些本地工人的利益会受到损害,而当地的议员显然觉得让自己的选民开心比买的裤子便宜10刀更加重要,所以美国许多工会会有这样那样的针对中国(及其他进口货)的诉求;当然,这种诉求平时可能就只是听听罢了,但如果正好面临竞选(比如最近的总统竞选),或是面临中美谈判,那么完全可以作为一个重要的谈判筹码。但中国人抵制日货我认为完全不是这个原因,因为日本生产的东西不太可能导致中国人的下岗(就算有也不会有很多人吧?),这里面更多的是民族感情的原因,当然有人认为这是政府引导民众将注意力从政府本身的不作为导向民族矛盾,这未必不是一种可能的猜测,但于我来说还是不怎么认同的。</p><p>最后还有一点理由,我觉得本身不算太过重要,就是案例中提到说中国有一天会因为东西太便宜而垮了的。这种说法当然是比较可笑的,但是确实把鸡蛋放在一个篮子里是比较不妥的。就好像香蕉在经过人们的改良之后,慢慢地只剩下一个种了,其它的那些野生的香蕉慢慢地都消失了;我们吃的香蕉的种子已经退化了,所以现在的香蕉相当于是克隆出来的,结果某种特定的病害一来,所有的香蕉都遭殃了;所以如果因为比较优势就把某种商品集中到某一个国家、某一个人去生产,可能国家突然政变或是人突然死了,这种商品就再也没了,道理是一样的。</p><hr>"},{"title":"教育的种族差异源自何处?","path":"2015/09/28/discussion0928/","permalink":"http://sx349.github.io/2015/09/28/discussion0928/","content":"<p>本文是9月28日《经济学原理》小课的记录,本次小课主要讨论如何用经济学研究教育中的种族差异。</p><hr><h1 id=\"控制变量\"><a href=\"#控制变量\" class=\"headerlink\" title=\"控制变量\"></a>控制变量</h1><p>一般而言,经济学家所说的控制某些变量,并不是完全地“控制住”变量。比如我们来研究父母教育水平对于孩子教育水平的影响,要控制父母的收入,并不是说把父母收入一样的一群小孩放在一组,然后分别进行比较:一方面所谓“收入一样”如何定义很成问题,可能结果是分成很多组,每组只有很少的样本;而另一方面,这样控制的结果只对每个组有效,相比之下,我们更希望找到一个对整体都有影响的解释。</p><p>再看一个经典的例子,就是研究价格和数量的关系。我们知道竞争市场上供求关系决定均衡价格和均衡数量,也就是说,价格和数量同时收到双重影响:给定供给不变,价格越高数量越高;给定需求不变,价格越高数量越低。假定我们采用完全控制住变量的方法,某几天供给曲线是完全一致的(这首先就很难做到),再观察这几天价格与数量的关系,发现其运动方向相反;然而这个结论是否适用于所有的供给情况下,还需要进一步的检验。</p><p>这就是我们为什么要引入回归分析,它相当于同时考察所有因素的影响,对每个因素进行回归都是控制了其它因素,是一个对整体的解释。</p><p>但是上述价格的例子在进行回归分析时,仍然是有问题的,因为我们不能够把供给作为一个自变量放入回归分析中。解决这种两个因素同时影响的常用手段之一是使用工具变量:有什么东西,它和供给有关,但和需求无关,而且能作为一个因变量使用?经济学家在研究农产品的过程中找到了一个工具变量,也就是天气。某一年天气的好坏(比如降雨天数)直接影响到粮食的供给,但是却不会因此改变人们对于粮食的需求,因此将降雨天数也作为自变量进行回归分析,控制降雨天数就是控制了供给。</p><hr><h1 id=\"回归分析\"><a href=\"#回归分析\" class=\"headerlink\" title=\"回归分析\"></a>回归分析</h1><p>尽管回归分析很强大,但是也不是所有的回归分析都是有效的,有时候回归分析也会出现问题。</p><p>首先是多重共线性的问题,这一点在自变量很多的时候尤其明显。所谓多重共线性,是指某些自变量之间存在非常强的线性关系,比如我们在设置人种的变量的时候,假定有黑人白人两种人,那么我们只能设置一个变量Black[i](黑人为1白人为0),如果同时设置两个变量Black[i]和White[i],两者相加永远是1,就称为完全线性关系;存在多重共线性时,尽管回归得出的参数和非共线性的情况变化不大,但是方差会被扩大,因而有些本来应该是有影响的因素会被认为是没有影响的。</p><p>另一个问题在于回归分析只说明存在相关关系,并不说明两者的因果关系。这里面就有很多文章可以做:</p><p>首先是内生性,就是我们打算研究A对B有多大影响,但是同时B可能也在影响A。我在Berkeley的一篇term paper讨论一个城市的犯罪率对周边城市的房价的影响,显然这样的影响是存在的,但是反过来,周边城市的房价上升,有可能引来更多小偷,使得目标城市的犯罪率反而下降了,这样就有了内生性的问题。为了解决这个问题,我考虑了去年犯罪率对今年房价的影响,因为今年房价不会影响到去年的犯罪率,但是这并没有完全消除去年房价同时影响去年犯罪率和今年房价,也即lurking variable的问题。</p><p>排除了内生性,还要考虑偶然性的因素。比如我在课上演示了两个随机数列的显著相关的例子,显然从数列的生成方式上就说明它们的显著相关只是偶然误差;此外,在时间序列分析上,还有所谓的Spurious Regression的问题,大体意思是两个整体递增的数列很有可能正相关,即使它们完全没有关系,比如中国的GDP增长和我的岁数。为了避免这种偶然性引起的问题,首先要做的就是认识到这个问题的存在,有些研究就是拿了一大堆自变量,一群一群地试过去,试到某一组正好显著了,就拿出来说事儿,这是很难避免偶然性误差的;其次就是要多做稳健性检验(Robustness Check),比如选取不同地区、时间的数据,或者对变量换一种衡量方式(比如本次案例中,对于受欢迎程度如何测量,作者就提出了不同的指标)再重做回归,如果结果相同,或结果虽然不同但是有合理的解释,才能说明这里的相关关系不是偶然性造成的。</p><p>排除这些问题之后,才可以开始谈因果性的问题。用于辅助解释因果性的计量工具有Difference-in-Difference(常用于政策冲击,选择两组人,一组收到冲击一组没有,然后进行比较)、Fixed Effect(加一系列控制变量)、Instrumental Variable(即工具变量)等等,但是归根结底,是要有一个好的故事,而且故事有好的寓意,比如族群认同当中,我们说Sabotage Model就是一个故事,它的寓意是坏学生会拖好学生后腿,但是这个寓意并不正确,因为实验证明好学生也在拖好学生后腿,所以这就不是一个好的故事。</p><p>即使确认了存在因果关系,还存在有Decomposition,也即分解影响的问题。A具体如何影响B,会直接导致决策上的不同,比如我们已经知道,父母的教育水平影响孩子的教育水平,但是到底是因为父母教育水平越高,因此更重视教育,因此孩子的教育水平更高,还是因为父母教育水平越高,收入就越高,因此孩子营养丰富所以教育水平更高,抑或是因为父母教育水平越高,社会地位也越高,甚至存在一些灰色关系而使得孩子获得更高的教育水平,这三者是截然不同的。除此之外,有时Decomposition也能解释无关变量的问题,如果A对B有正向的影响(而且我们经常观察到),但同时A潜在地对B有负向的影响(通常被我们忽略),结果综合起来可能就相互抵销了。</p><p>与回归分析还稍微搭点边的另一个问题在于数据。数据本身是否合理可行?如果拿到的数据是有偏差的,比如存在人为选择数据的问题(注意,为了解释自己的理论刻意忽略一些数据或是伪造删改数据是严重的学术不端),或者是存在幸存者偏差、选择偏差,尽管回归本身没有错,但结论最终还是错的。比如案例中提到,父母打孩子屁股对孩子的教育水平没有影响,这是一个选择偏差的问题,因为父母是否打了孩子屁股,不是暗访得出的结果,而是询问父母的结果,会说出自己打了孩子屁股的家长是怎样的家长?既可能是觉得“棍棒底下出人才”,就是喜欢打孩子屁股的家长(导致教育水平更差),也可能是虽然打过但是并不以此为手段,而且敢于承认的开明家长(导致教育水平更好),两相抵消,可能因此就给出了没有影响的结论。</p><hr><h1 id=\"实验的可行性\"><a href=\"#实验的可行性\" class=\"headerlink\" title=\"实验的可行性\"></a>实验的可行性</h1><p>经济学家很难进行实验,很多时候他们更倾向于利用自然实验的结果,比如政策改变、突发的自然事件或政治事件等等,这是由于经济学家所要研究的对象是人,对于人的研究往往要考虑到可操作性和伦理的问题。</p><p>有些实验是不可操作的,比如有同学在memo中提到建一所学校,选一批黑人白人去上课,并采取不同的政策,这样的学校是不会有人去的;再比如研究超速通货膨胀,经济学家不可能自己跑到哪个国家去主导发生一场超速通货膨胀,只可能是先有某个国家(比如津巴布韦)经济发生问题,或是像索罗斯狙击泰铢那样有大量热钱涌入炒作,经济学家才有机会跟上去进行观察并收集数据做出论证。</p><p>更为严重的是伦理问题:假如上面所说的学校真的成立了,而你的政策被证明确实会对学生产生影响,那么那些收到负面影响的学生该怎么办?他们未来收入的下降由你来负责吗?李宏彬教授讲过一个故事,他们拿到基金会的一笔钱,去乡村小学做双盲实验,随机选一批孩子,资助他们上大学,本来计划做很多年实验,结果第二年就没有做,为什么呢?因为孩子们问他们,为什么成绩最好的那个孩子没拿到钱去读书,而成绩很差家里有钱的那个孩子却拿到了资助?这就是以认为实验对象不可避免的问题。再比如Duflo领导下的J-PAL在贫穷地区搞微观革命,随机选一半人打蛔虫,最后验证打了蛔虫对未来收入是有影响的,我就一直在想剩下那一半人怎么办呢?虽然说起来他们和原来没有J-PAL,没有RCT的时候没有两样,但是很多时候个人的处境不是由绝对水平决定,而是由相对水平决定的,一开始只是收入差距,等到累积起来就可能是社会阶层差距,这并不是没有可能。</p><p>所以经济学家经常会面临这样的两难境地,如果顺应了民意,我们不要随机,可不可以呢?作为国家政策当然是可以的,但是这样做的结果并不是完全真实的,这涉及到经济学上所谓逆向选择和道德风险的问题。用保险来解释比较清楚,逆向选择的意思是,投保的人本身就比不投保的人更有可能发生意外(事前的扭曲),而道德风险的意思是,投保之后人们会更加不小心谨慎(事后的扭曲),存在着两种问题,保险公司就不能以社会上平均的意外发生率为标准设定投保的门槛或费用。在这里也是如此,加入我们要验证资助会使得孩子未来的发展更好,那么如果资助的就是成绩好的学生,我们怎么知道是由于资助所以他们更好,还是因为他们本来就爱学习,即使没有资助一样会过得很好呢?</p><hr><h1 id=\"理论与验证\"><a href=\"#理论与验证\" class=\"headerlink\" title=\"理论与验证\"></a>理论与验证</h1><p>对于大家在memo中提到的一些原因,我稍微做一些点评:</p><p>正面榜样:<br>这个在课上已经说了,实际上已经有人以“The Obama Effect”为题进行了一些研究。</p><p>优越感与自卑心理:<br>白人的优越感和黑人的自卑心理,造成白人更加努力学习,黑人更加自暴自弃,可以通过心理学量表给出相应的优越感或自卑心理的度量,然后对人种进行回归看是否有显著的差异,如果有的话再进一步作为成绩回归分析的自变量。</p><p>基因决定:<br>基因决定人的性格/智商,进一步决定教育水平。基因决定论之前炒得很火,尤其是有这样一个观点:有些环境因素本身也是基因造成的,更进一步扩大了基因对教育水平的影响。关于基因的研究,非常经典的一个做法就是使用双胞胎或多胞胎,尤其是比较同卵双胞胎和异卵双胞胎的不同。</p><p>近朱者赤近墨者黑:<br>这是一个非常trivial的解释,但是我觉得还是值得讨论的,首先是这个“近”、“朱”、“墨”到底如何衡量,指标的设计是很有意思的,不同的设计可能给出不同的结论;其次是到底这里面是如何影响的,是否有隐藏变量或反向因果(父母收入低同时导致子女教育水平低和住较差的小区),是否结合了其它因素,都需要进一步说明。</p><p>历史原因:<br>总的来说就是马太效应,贫者愈贫,富者愈富。如果真的是这样的话,我们是否可以推论,白人之间也应该存在分化?就是说白人如果在一代中有人富有有人贫穷,或是有人掌管大权有人苦逼打工,那么未来白人之间也划分出不同的阶层?</p><p>家庭结构:<br>黑人中有很多单亲家庭或是多子女家庭,因此影响了孩子的教育水平。这个观点在其它小班都有提到,可以用回归分析进行验证,不过我想到的一点是,孩子在多子女家庭中排行第几也是影响成绩的很大因素,值得注意。</p><hr>"},{"title":"阿里巴巴的几个关键年","path":"2015/09/21/discussion0921/","permalink":"http://sx349.github.io/2015/09/21/discussion0921/","content":"<p>本文是9月21日《经济学原理》小课的记录,本次小课讨论了阿里巴巴的发展历程。</p><hr><h1 id=\"市场\"><a href=\"#市场\" class=\"headerlink\" title=\"市场\"></a>市场</h1><p>许多同学在memo中说,马云和阿里巴巴之所以成功,是因为“响应了市场需求”。如果我们把这里的市场理解为服务市场,阿里巴巴是作为渠道或者说信息供给的一方,而中小企业是需求的一方,那么这么说也无可厚非;但是,更加准确地说法是,中小企业是商品的供给方,而外国的采购商则充当了商品的需求方,原先这个市场是不存在的,而马云创立阿里巴巴是“创立市场”,使得交易得以实现。</p><p>前一种说法还有一点瑕疵在于,阿里巴巴做的是一个平台。平台最重要的特点是所谓的网络外部性:简单地说,在平台上的人越多,则每个平台上的人收益越大。平台与平台之间一定是竞争的,大家都想吞并对方成为唯一的平台(这样对社会也是有利的,两个平台合并成一个平台后,本来不存在的平台间合作就变成了可以存在的平台内合作),而且用户也会有这样的期望,这也是为什么Facebook和人人是没法合作的。如果认为阿里巴巴是信息的供给方,会推导出怎样的结果呢?那么根据比较优势,一家美国公司应该对于美国更加熟悉,所以阿里巴巴应该向国外买服务,而不是自己去收集,这和实际是不相符的(阿里巴巴作为平台会希望美国商户也选择阿里巴巴而不是别的平台然后再交换信息)。</p><p>还有一些同学提到了政府政策支持,必须更正一点,支付宝受到政策支持并不是十大原理中说的“政府有时可以改善市场的结果”,正相反,eBay在这场竞争中作为外资受到限制,反而说明政府干预了市场。我猜测大家可能觉得免费就是好的,淘宝的免费是更好的市场结果,这个观点太过朴素了;经济学上所谓的改善市场的结果是指,市场失灵有时降低了社会总福利,政府政策可以纠正这种情况。比如工厂生产高污染产品的时候,做决策并没有考虑污染的成本,这部分污染造成了环境问题,但没有人去负责,因此政府对工厂进行罚款,使得工厂生产该产品的成本上升了,他们会相应的减少产品,尽管交易量减少了,但是污染也减少了,总的来说利大于弊,这才叫改善市场结果。</p><p>再回过来看政府管控外资的政策,在接下来的几章中将会讲到比较优势理论,它认为,如果中国生产轮胎具有比较优势,而美国生产油箱具有比较优势,那么应该各自生产占有比较优势的产品,然后进行交易,这也体现了贸易使双方都更好的原理。然而,比较优势理论有一个问题在于,它把一个国家抽象为一个劳动力,认为这个劳动力可以很快地进行调整自己所擅长的领域,这对于个体成立,但对于国家却未必,轮胎工人很难一下子转而生产油箱,这里面存在摩擦性失业的问题,而作为一个政府,不仅要考虑国家的经济状况,也要考虑政治状况;底特律汽车城的衰落可以说是比较优势理论所预料到的经济结果,但是随后底特律整个城市的发展也陷入了困顿,人烟荒芜、治安状况变差等等,则是随之引发的意料之外而情理之中的连锁反应。除此之外,有一些新兴产业,国家也不知道是否属于自己的比较优势,所以可能采取贸易保护;或者,有些产业涉及到国家安全,比如军火,即使美国人造枪再便宜,中国人也很难放心把这个产业交给别的国家,像稀土矿则是因为属于重要的矿产资源,尽管中国开采稀土矿具有比较优势,但是仍然要逐渐管控对外国的出售。</p><hr><h1 id=\"免费\"><a href=\"#免费\" class=\"headerlink\" title=\"免费\"></a>免费</h1><p>免费也是很多同学都提到的重点话题,我就略过为什么免费吸引了顾客这个问题,以及马云为什么要采取这样一个策略的问题,大家基本上都说得很清楚了。</p><p>我先来解释一下天猫的收费,如果免费真的更好的话,为什么会有商家愿意去一个收费的平台呢?我们来考虑这样两家店,一家店是卖正品毛绒玩具的,卖的比较贵,20块钱一只;另一家店卖同样的毛绒玩具,但是他们的做工或者填充物都要差一些,但是消费者很难分辨出来,因为用料差,他们卖10块钱一只。</p><p>假如现在只有淘宝,这两家店都开始卖毛绒玩具,消费者一看,一个10块一个20块,卖的东西又没什么差别,果断选10块的,这样20块的店的销量就差得多,没办法他们也只能卖10块钱一个,勉强维持生计。然后现在推出了一个新的平台叫作天猫,要求只有卖正品的可以注册,还要交5块钱认证你是正品的费用,那么正品店算了一笔帐,每一只虽然多交5块钱给天猫,但是我能卖20块钱,所以相当于净挣了5块钱,于是就欢天喜地地在天猫开店了;此时,由于用料差的店不能证明自己卖的是正品,大家都不来买了,他们不得不卖8块钱一只才能保证收益。</p><p>这就是所谓的信号理论,上了天猫的店花费一定成本放出了一个信号:我是正品,此时消费者就从本来不能够区分两家店而只愿意付相同的价格(称为混同均衡)到了能够区分两家店而愿意付出不同的价格(称为分离均衡)。这个理论在生活中的应用是,学霸和学渣在简单的课程上无法区分出能力高低,因此用人单位只愿意付一个相同的工资;为了避免这种状况,学霸会选择学渣必定挂科的困难课程,这样用人单位一看你修过这个课就知道你是学霸,给你一个较高的工资,而没修过的就是学渣,工资会比之前更低。</p><p>实际上,混同均衡很有可能同时伴随有逆向选择:用人单位付出的工资一定是取学霸和学渣的平均水准,因此学霸会考虑另谋高就,而接受这个工资的必然是学渣,用人单位就花了冤枉钱了,因此这里潜藏了社会效率的损失;但分离均衡中,学霸为了证明自己是学霸,需要多修一门困难的课程,因此付出了更多的精力才得到了适应自己实际水平的工资,这也造成了社会效率的损失。</p><p>再说一下课上一带而过的关于收费会不会和税收那样降低社会总福利:将收费类比为税收没有太大问题,但关键是怎样的税收。注意到税收楔子是在对每单位物品征税的情况下出现的,如果是征人头税(比如一个商户就交一万块,不管你卖多少商品),或者是对利润征税,在这两种情况下,征税都不会改变商家的供给曲线,因此也不影响价格和社会福利,只是把本来给商家的一部分钱转移到了国家手里。</p><hr><h1 id=\"支付宝\"><a href=\"#支付宝\" class=\"headerlink\" title=\"支付宝\"></a>支付宝</h1><p>关于支付宝是如何解决风险问题的,我在课后和几个同学大致描述了一下,这里我们先假定原来是由买家先付款,卖家货到付款的(不知道大家小学初中有没有订过杂志,订杂志都是先付一年的钱,然后对方一期一期寄过来的)。原来存在一个怎样的信用风险呢?由于是货到付款,卖家可能拿到钱之后决定发次品给买家,当买家发现拿到的是次品的时候,再去维权是很困难的。用博弈论的话说,这是一个拓展式博弈(图我就不画了,大概是买家现决定要不要付钱,如果付钱则卖家决定发什么货);根据反向推理,在卖家发货时,发次品比发正品合算,因此买家付钱拿到的一定是次品,这比买家不付钱要差,所以买家一开始就决定不付钱(也就是不买)。</p><p>支付宝的出现,意味着,如果卖家发了次品给买家,买家可以选择立刻退货,申请支付宝退款。由于钱很容易要回来,卖家可能还要承担运费和机会成本,因此卖家发次品的收益降低了,如果降低到比发正品的收益更低,在买家付钱时收到的一定是正品,这比买家不付钱要好,因此买家决定买东西。只要买东西,淘宝就有赚钱的可能。</p><p>支付宝同时可能充当了给阿里巴巴提供现金流的功能,这也是有同学提到的。但这一点我不太清楚所以也不敢妄下断言,从准备金的角度来说,支付宝中买家已经用于付款给卖家的那部分,应该是被冻结起来不能够动用的(否则用户一旦来取款就要拆东墙补西墙,这还是比较危险的),这样一来只有支付宝中没有用来买东西的钱可以动用,就比总金额少得多了。但如果没有冻结的话,那将会是一笔很大的现金流(或者起码是很容易变现的资产),尤其是支付宝用户如果买了东西到货之后没有立刻评价,那么这部分钱相当于给支付宝发了个几天的无息贷款,这部分息差累积起来还是很惊人的。</p><hr><h1 id=\"企业家精神\"><a href=\"#企业家精神\" class=\"headerlink\" title=\"企业家精神\"></a>企业家精神</h1><p>这里就不谈泛泛的比如企业家需要魄力、胆大心细等等之类,大家也都说过了。投资领域有一条准则叫作高风险高收益,其实对应到股票等理财产品中,这是可以根据一些假设根据CAPM理论计算出来的。然而这里面有一个问题是,高风险高收益这句话,就等于是十大原理的第一条,人面临权衡取舍,但是这并没有说明人们该怎么进行权衡取舍。这里我们权衡取舍的就是风险和收益,作为一个普通人,我们往往是风险厌恶的,也就是说,同样的收益,风险越小越好,结果是我们会更多地考虑,说不定就是差的那个结果发生了(有点像所谓的墨菲定律)。而我觉得,很多企业家有时候可能是风险偏好的,同样的收益,风险越大越好,结果是他们会更多地考虑,说不定就是好的那个结果发生了(注意这里的说法不严谨,只是一个通俗的解读),因此他们会更敢于去拼一把。</p><hr>"},{"title":"谈谈版权的经济学","path":"2015/09/17/copyright/","permalink":"http://sx349.github.io/2015/09/17/copyright/","content":"<p>Q:写这篇文章有怎样的缘起?<br>A:昨天经原小课集体备课,与其它助教说起,现在很多人抱有“朴素的经济学观点”。所谓“朴素的经济学观点”是我自己总结的一个名词,凡是依靠一己得失、道德规范亦或是舆论风议为基础而得出的经济学观点,大抵可以归入这个范畴。比如有人说,小贩看人出价,着实可恶:明明同一个钱包,看你穿得像学生就五块十块,倘若你衣着光鲜就一下狮子大开口涨到五十,真是有违“价格公平”。恕我孤陋寡闻,“价格公平”一词从未见于任何我所读过的经济学书籍中,倒是听说过一个经济学术语叫“价格歧视”,而小贩此举恰恰是第三类价格歧视的一个经典例子。当然,“朴素的经济学观点”未必是错的,只是经济学的结果有时是非道德、反常理、与舆论无关的,“朴素”与正确之间有时难免会差之毫厘。</p><p>Q:那么,王源宗VS央视一事,莫非也有一些“朴素的经济学观点”?<br>A:在我看来,可能有两类观点是颇为典型的“朴素的经济学观点”。<br>其一,可以称之为“种瓜得瓜,种豆得豆”论。我种下了瓜,花了多少成本多少心血,我就要把这些都从卖瓜的价钱里收回来,这逻辑是不错的;既然如此,我花这么大功夫,跑到西藏,摸爬滚打了一年才拍出这个作品,那么最后使用我这个材料的人,就应该按照我的辛苦程度付账才对。<br>其二,可以称之为“苹果与思想”论。你有一个苹果,给我你就没有了;你有一个思想,你告诉我之后,我们俩就都有这个思想。那么,你们大公司的软件,Ctrl+C、Ctrl+V就能复制一份,也不费什么功夫,而且你们自己的原件还在,凭什么卖这么贵?同样的,你的视频放到网上,反正是给别人用的,我们央视用你的视频,给点钱意思意思就算了嘛。</p><p>Q:我想大部分人都会认可前者而否定后者吧?<br>A:我们先来看第二个观点,用经济学的语言来解读的话,它的意思是:物品的价格应该定在供给曲线和需求曲线相交的位置,而软件或是视频,一旦制作完成,再复制一份的成本几乎为0,也就是说它们的供应曲线是一条接近于0的水平直线,那么均衡价格就应该是接近于0。<br>在竞争市场上,这个观点没错,但是它忽略了一个前提,就是在这个价格下,市场将不复存在:实际上,由于版权方的固定成本很高而边际成本很低,边际成本一定低于平均总成本,如果定价等于边际成本,那么他一定会亏钱,版权方宁可退出市场。<br>然而,这个市场的存在是有益的,我们需要更多更好的软件,我们也需要构思巧妙、拍摄精良的视频和照片。那么如何让这些在竞争市场上必定会亏钱的东西留在市场上呢?就要靠版权。为什么版权有用?因为它把竞争市场变成了垄断市场,生产者可以自由决定产品的价格,进而提高生产者的收益。这样一来,平均价格就可以超过平均总成本,生产者有钱可赚,就会继续留在市场上。<br>不过,这个定价也是有讲究的。如果生产者不能区分不同的购买者,只能定一个价格,那么应该提供边际收益曲线和供应曲线交点处的数量,这时能够达到最高收益;如果可以不同的人出不同的价格,那么理论上可以完全根据需求曲线定价,攫取所有的剩余,这就称为完全价格歧视。<br>读到这里,我们就会发现第一个观点的问题了。我们的讨论其实一直都围绕着供应曲线和需求曲线,而供应曲线刻画的只有边际成本(产品多生产一份的成本,这里视频或软件就是复制一份的成本)而没有固定成本(产品引发的成本,这里视频或软件就是拍摄和开发的成本)。因此,即使有版权保护,即使是垄断,最后定价的时候,和付出的固定成本应该是完全无关的,如果你拍得好,大可以多收一些钱,要是你拍得不好,那么别人不愿意出高价,你也没办法。</p><p>Q:原来如此。那么最后一个问题,我们平时也都在用盗版软件,看盗版电影,听盗版音乐,用盗版的字体和扣掉水印的图片做海报,岂不是会让这些版权方都退出市场?<br>A:首先,有些版权方可以依靠非经济的效益来弥补低价和高平均总成本之间的差距。我们平时看剧,都接触过字幕组,字幕组的工作不也是一个创作的过程吗?但是,这些工作人员可以获得别的方面的收益,比如有人觉得分享是很有乐趣的,有人想要在这个过程中锻炼自己的英语能力,有人觉得赶在别人之前完成翻译很有成就感。所以,尽管我们没有啥金钱贡献,暂时也还不用担心字幕组集体罢工。(当然,我们也应当给这些工作人员以足够的尊重和钦佩)<br>其次,只要生产者目前从那些愿意付钱的人手里拿到足够的钱,弥补了平均总成本,那么我们的盗版应该还不至于危及他们的生存。其实,如果我们觉得软件价格过于昂贵,就意味着我们处于生产者设定价格时的均衡数量右侧,也就是说我们本来就不是他们期待的购买软件的对象;对于他们来说,关键是那些本来有意愿而且买得起软件的买家不要去用盗版软件,而这些买家恐怕就不是普通人了,主要还是各种公司或者机构,相比之下受到的监管更加严格,维权的收益成本比也更高。<br>最后,作为普通人,无论是版权方还是使用方,都难免觉得自己是弱势群体。作为版权方,比如此次事件中的自由摄影师王源宗,面对大公司侵权,往往是维权收益远低于维权成本;作为使用方,又觉得这样的高价承受不起。但这种情况并非没有解决的办法:版权方可以自己进行结盟(比如2011年百度文库侵权事件作家集体声讨百度文库),或是依靠已有的专业版权机构(比如音乐领域有环球音乐及其在中国的授权机构松巴音乐);使用方也可以考虑众筹再分享(比如两人约定各买一本书然后交换着读),或是依靠自己所在的组织机构(比如清华向学生提供正版软件以及众多数据库;再比如政府用财政收入补贴相关企业),此外有些非商业目的的使用,我们又作为学生,可以尝试与版权方进行交涉;如果实在没有合法使用的方式,在选择盗版之前,最后一招是先找找有没有公有领域的替代品,比如遵循知识共享协议的文学艺术作品,或者用开源项目以及共享软件替代商业软件。版权的设置毕竟是有利于全社会的,我们应当尊重版权;但是,为了使版权所保护的作品产生更大的价值,就如何尊重版权这个问题,理应发展出更加灵活有效的解决方式。</p><p>最后的最后,应景的罗嗦一句:本作品采用知识共享署名-非商业使用-相同方式共享4.0国际许可协议进行许可。</p>"},{"title":"思源新马实践感想","path":"2015/09/14/siyuan-2015summer/","permalink":"http://sx349.github.io/2015/09/14/siyuan-2015summer/","content":"<p>我在这次思源实践中留下最深的两点感受,大抵都是和经济学相关的。</p><hr><p>第一点,自然是在新加坡的实践过程中,深刻地感受到为何新加坡被称为经济学家最喜欢的一个国家。在新加坡,许多制度细节都体现了曼昆所谓十大原理之一“人们会对激励作出反应”,而且几乎每天我们都会有新发现。</p><p>在乘坐地铁时,我就注意到,首先使用地铁卡(EZLink)的票价比同样路线的纸质车票要便宜得多,这激励人们多使用地铁卡而少使用纸质车票;其次,如果你已经购买了纸质车票,每张纸质车票会多收取0.1元,在你第三次使用该纸质车票时返还这0.1元,在第六次使用时更会多返还0.1元,而如果你已经购买了地铁卡,在地铁卡中的余额低于3元时就会要求你去充值才能使用,显然大多数人都会选择充值以免浪费了卡里剩下的钱,这样变相地激励了人们重复利用车票和地铁卡,减少污染。</p><p>新加坡的组屋制度更是如此:组屋的申请条件是年满21岁、组成核心家庭的新加坡公民,而单身者必须到35岁才能申请,因此有人戏称新加坡女生想要暗示结婚,只需给男生多看一些组屋的广告单即可。不仅如此,在组屋发展的沿革中,新加坡更是将大量社会政策融入了组屋的制度设计中,比如生育三个孩子的家庭购买组屋可以优惠,子女购买父母附近的住房或是父母购买子女附近的住房也可以享受优惠,家庭有两次购买组屋的机会,但如果第二次是在二手市场上购买,不仅不需要交纳转手费,还可以获得政府的额外补贴。</p><p>以上这两例虽只是管中窥豹,然而我们业已可见一斑,新加坡看似违背了著名的“看不见的手”配置资源的规律,然而政府却依靠强有力的控制和有效的激励创造了一个经济高速发展,人民生活安康的社会,这也再次印证了十大原理中的另一条“政府有时可以改进市场结果”:一方面,当社会函数除了注重效率也注重公平时,政府的干预能改善社会的公平,因此提升了社会效用;另一方面,有些市场行为具有外部性,比如我们重复利用卡片,本身是对环境的保护,具有正外部性,政府的激励措施对于重复利用有奖励(变相补贴),对不重复利用则相应地罚款(变相征税),也提高了总效率。</p><hr><p>第二点,则是从民族融合引发的一些思考。</p><p>在经济学上有个名词叫价格歧视,它和我们日常所说的歧视含义大相径庭,但是却并非没有关系,因为曾经有过这样一个问题:黑人比白人的工资低是不是一种歧视?头脑简单的政治活动家们可能立刻跳起来表示这是种族歧视,然而经济学家却提出,黑人和白人的肤色可以有效地把这两类人区分开来,在劳动力市场上,就可以划分出白人劳动力市场和黑人劳动力市场,在一定的历史阶段中(甚至可能包括现在),白人的生产率都是要比黑人更高的,那么在雇佣劳动力时,通过肤色给白人付更高的工资是一种有效率的做法,这是第三类价格歧视的一个典型例子。</p><p>如果把同样的做法从价格中跳出来,在不完全信息动态博弈中,称之为分离均衡(与之相对的是混同均衡),分离均衡就是通过一定的措施分辨出不同类型的参与者,分别采取不同的措施,显然这是比混同对待效率更高的一种做法。</p><p>新加坡和马来西亚都是多民族国家,官方总是在推动民族融合。然而,无论民族融合做得多好,就像新加坡这样,各族人民齐聚一堂其乐融融都说英文,各民族的客观差异仍然是存在的。比如,我们在新加坡最大的反对党——工人党组织的集会上,遇到这样一种情况:一名选区候选人在进行演讲时,使用了两种不同的语言,先是用普通话说了一些内容,然而又用英文重复了一些内容,然而我注意到,他所说的话前后有所差异,普通话专门提到了老年人的养老问题,还说到了自己小时候生活的艰辛,而使用英文时却用年轻人的教育问题替代了养老问题,而且也未提到过去生活困难云云。后来我们讨论的结论是:使用普通话的主要是老年人(新加坡成为乐龄人士),他们更关心自己的养老问题,而且也曾经历过新加坡建国时的艰难;而使用英文的则更多是新一代的新加坡人,教育是他们更关心的话题。除此之外,有些候选人使用闽南话或是马来语进行宣讲,同样都是有着其目标群体的。</p><p>我们也听到这样一种说法:由于新加坡生育率较低,除了积极引进移民外,政府还希望鼓励公民多生几个孩子;而李光耀作为华人,自然更希望华人能够多生,因此他采取的做法是,对于高学历的公民,多生育孩子可以获得奖励,而华人的学历很高,因此这个政策相当于鼓励华人生育。</p><p>那么,我们为什么不认同直接的种族歧视,却认同或至少默许依靠语言或者学历来区分对待不同民族呢?</p><p>一个类似的例子是在我们和马来西亚《东方日报》执行主编陈利良先生交谈时所联想到的:陈先生提到,他不提倡在报社里搞小团体,因此他也不会特地照顾某些朋友推荐的人,省得别人觉得受了他的恩惠。我当时就想到,实际上,这种较为隐蔽的小团体是不可避免的,我们每个人自己也会有这样的经历:比如我曾经在志愿中心任职,工作中结识了一位学弟,后来团委书记希望我推荐担任志愿副书记的人选,我自然就推荐了他;这种小团体内部的相互“照顾”本身,我们乍看之下不会觉得有什么问题。然而,如果当我们推荐的人是我们的亲戚,就难免会被扣上“任人唯亲”的帽子。这个例子同样是一个歧视,不过是在我熟悉的人和我不熟悉的人之间有了区别对待,但为什么我们能够接受推荐一个熟悉的学弟学妹,而不能接受推荐一个熟悉的亲戚呢?</p><p>这个问题是我在最后一天的交流会上提出的,从那时到现在又过去了三天,我也有了一些新的思考。最主要的有两点:首先,这种区别对待是否是有着更有意义的目的,比如我们比较一下高学历生育奖励和直接对华人生育进行奖励,前者追求的是更有意义的目标,就是提高新加坡的劳动力数量和质量,提高华人数目只是一个附赠品;其次,这种区别对待是否能够为被区别对待者所发觉,直接的种族歧视是非常明显的,因此也会招致非常激烈的反对,然而如果依靠语言,听不懂马来语的人未必知道候选人对马来人做怎样的许诺;同样的,亲戚关系是比较容易被发现的,而和某人共事过这种关系却比较难以获知。</p><p>再回过头来看,其实我们就会发现,在新加坡的种族融合的大环境下有很多这种经验的运用,比如组屋中的民族比例政策,实际上就是一个扣上了“民族融合”大帽子的区别对待,如果一个社区中已经住了足够多的华人,即使有很多华人在等着住,政府也会把剩下的房子分给马来人和印度人。</p><hr><p>最后,我想再来表达一下对如下观点的看法:“政治问题归根结底都是经济问题。”</p><p>有人认为这句话的意思大致等同于“经济基础决定上层建筑”,我是不赞同的。我的理解是,人民对于政治总是冷感的,政府能发展好经济,维持就业率,那么人民就会支持政府,这一点在中国是如此,在新加坡是如此,在早期的苏联也是如此;而一旦政府无法维持经济增长,也就无法维持民众的生活水平,民众的政治热情就会瞬间高涨。不过,从这个角度来说,也就意味着所有在经济快速发展期间的政治问题,都不是经济问题,比如不同政治派别的相互倾轧,就是纯粹的政治问题。</p>"},{"title":"与基督徒论辩","path":"2015/08/25/christianity/","permalink":"http://sx349.github.io/2015/08/25/christianity/","content":"<p>在伯克利交换期间,与基督徒论辩多次,将其中部分内容节录如下。</p><hr><h2 id=\"关于上帝是否存在与耶稣的复活\"><a href=\"#关于上帝是否存在与耶稣的复活\" class=\"headerlink\" title=\"关于上帝是否存在与耶稣的复活\"></a>关于上帝是否存在与耶稣的复活</h2><p>就我所认识的、与我曾经交流过的基督徒而言,他们大抵承认,上帝的存在是基督教义中最核心的概念。而耶稣基督的复活(Resurrection),则是这一概念最关键的表现,因为其存在两重意义:其一,耶稣能够复活,说明上帝作为一个拥有超越人类能力的人是存在的;其二,耶稣的死是代替人类去死的,他背负着人类的原罪,而他的复活则意味着人类的原罪被上帝所宽恕。</p><p>反之,耶稣的复活如果是假的,那么一方面上帝的存在将失去一个极为重要的证据,更重要的是,这意味着耶稣没有能洗刷人类的原罪,因此人类在审判日终究无法得到救赎,对基督教的信仰也将毫无意义。在《圣经》中,耶稣的门徒也曾说,如果上帝真的不存在,那么将会是信仰者的灾难。</p><p>所以争议的关键转到耶稣复活这一事件的真假上来。显然,我们不可能回到当时当地,重新复现一遍当时的历史真相,因此基督徒提出了以下理论,来论证耶稣复活的真实性:</p><ul><li>首先,关于耶稣复活存在许多证据,之所以这些证据是真实的,是因为它们同时被记录在拥护基督和反对基督双方的史籍上,且记录的时间和耶稣复活事件发生的时间相去不远;</li><li>这些证据中,有四条如下所述:</li></ul><ol><li>耶稣在十字架上死去,流出的血液为死后的暗红色;</li><li>耶稣死后入殓,其墓有卫兵把守;</li><li>在有人把守的情况下,耶稣的墓空了;</li><li>耶稣重新出现在门徒和敌人面前;</li></ol><ul><li>然后我们来考察各种可能的解释,比如有人认为,耶稣确实死了,人们看见的是幻象,这没有违背1和2,然而不能解释3,解释4时也很牵强;如此这般对于所有的解释都考察一遍,只有上帝使耶稣复活恰好符合所有证据。</li></ul><p>(顺带一提,《来自地球的男人》 (The Man From Earth) 中的解释,即主角就是耶稣,其身体机能永不老化,而且曾经和佛祖学习过一些屏息之术,因此在十字架上假死并复活,不能解释1和3。)</p><p>我的反驳有两个方面:</p><ol><li>这里所用的方法,在概率学上其实就是Maximon Likelyhood Estimation,然而MLE毕竟只是估计而不是准确值,举个例子,抛掷一枚标准硬币2次,出现2次正面,MLE将会估计硬币出现证明的概率是1,然而这个估计并不改变这枚硬币是标准硬币——也即正面反面出现概率各1/2——的事实。换句话说,上帝使耶稣复活这一解释确实更好的解释了这些证据(也即这些证据出现的概率更高),但是如果事实并非如此,那依然并非如此,不会因为“更能解释”就改变事实。</li><li>上面这个反驳比较和稀泥,而且美国的刑事体系其实遵从的就是这样的逻辑体系(用“被告确实犯罪”所能解释的证据越多,“被告确实犯罪”的可能性——或起码陪审团所认为的可能性——就越高)。另一种反驳的立论是,隐含的前提是否成立?除了上面这个反驳中逻辑体系成立的前提外,至少还有两个前提值得商榷:<br>其一,记录在正反双方史籍上且记录时间相近,是否就意味是真实证据?是否有可能是选择性偏差,即基督教后来逐渐兴起,因此销毁或是篡改了并没有记录这一事件的文献?毕竟古人的记录手段是有限的。(我们暂且不纠结“记录时间相近”这件事的真实性,因为我不是专业人士,不清楚探测古籍年代能够精确到什么单位。)<br>其二,基督徒的理论其实是将“超自然力量存在”隐含地作为前提的,以此证明“上帝使耶稣复活”,然后推导出“上帝存在”,这构成了一个循环论证。无神论者根本就不承认“超自然力量”存在,在这一基础上,“上帝使耶稣复活”这一解释首先就会被排除。试想,倘若我们认同“超自然力量存在”这一前提,那么犯罪案件还怎么侦查?说不定是超自然力量作案,比如密室杀人是高维嵌入,跳楼自杀是心灵感应,等等。所以刑警侦查时,首先就会认定,不管现场再怎么神奇,都一定是人作案,在这个基础上才可能进行侦查;如果刑警是一个无比虔诚的基督徒,认为这个人之所以死是上帝派天使带他去天堂,或是撒旦派魔鬼害死了他,那还侦查个毛线啊。</li></ol><p>(关于上帝存在,还有一个小插曲,也就是为什么上帝不引发神迹向世人证明他存在,毕竟这是最容易使人类相信上帝存在的做法;然而截至目前,唯一创造过“神迹”的大概就是魔术师了,比如穿越长城、偷走自由女神像的大卫·科波菲尔。)</p><hr><h2 id=\"关于上帝创造人类\"><a href=\"#关于上帝创造人类\" class=\"headerlink\" title=\"关于上帝创造人类\"></a>关于上帝创造人类</h2><p>我周围的基督徒都逐渐承认,《创世纪》中的内容并不是真实的,而是作为比喻或暗示的存在。然而,他们没有放弃上帝造人的说法,这个说法我接触到有两个层次的表述:</p><ol><li>上帝创造了宇宙,认为宇宙不可能无中生有,在宇宙诞生之前上帝即存在。</li><li>上帝创造了生命,认为地球能够适合于生命存在,尤其是人类存在,是上帝所设计的。</li></ol><p>对于前者的反驳如下:首先,宇宙并不是无中生有,是由奇点大爆炸而形成的,奇点本身有质量和能量;其次,“时间”是在大爆炸之后才出现的,“宇宙诞生之前”是一个伪命题。这两点是物理学理论中最符合一切证据的理论结果,因此仿照基督徒对于耶稣复活的理论,宇宙在大爆炸中诞生,比起宇宙由上帝智能设计出来,是更加合理的理论。</p><p>但是,有些基督徒会提出,这个宇宙的一切参数都是合理的,只要稍作改变人类就不可能生存,地球的环境也是如此适宜人类生存,这绝不可能是偶然的,应当是上帝设计的结果。这显然是无稽之谈,我举了个例子:用计算机生成两个完全随机的数列,在生成足够多次后,极有可能存在两个数列显著正相关,这时如果我们问为什么这两个数列显著正相关,显然不是因为我们故意安排了这两个数列。把这个道理放在人类身上,就是所谓的人择定理:为什么这个世界适合人类生存?因为不适合人类生存的世界(物理常数不适合的宇宙,或是生态系统不发达的地球)中人类不曾存在或已经灭亡,只有人类生存的地方能够有人问这样的问题。</p><p>我试图用更通俗的例子表达这一理念,例如:对于某个特定的人类,并不一定能够留下后代,而后代再留下后代概率更低,尤其是考虑到饥荒、战乱等等,一个家族能流传十代也许都是件幸事。然而,每个现在活着的人,比如我,我的父亲成功留下了后代,我的父亲的父亲也成功留下了后代,我的父亲的父亲的父亲……,每一代人竟然都留下了后代,倘若我们计算一下这个概率,一定是惊人的低,只要稍有偏差我就不可能被生下来。但是,我已经确定被生下来了,因此这一路上每一个本来很低的概率,现在的条件概率都必然是(也必须是)100%。<br>还有一次,我问一名基督徒朋友,他在晚饭中吃了几片橙子;他说吃了五片,于是我又问他,为什么你吃了五片?你是特别挑选了5这个数字的吗?实际上你只是拿了几片,你并没有在意它是三片、四片还是五片,你在挑完之后,甚至吃完之后,甚至我问了这个问题之后,才发现是五片。当然,特地挑选吃几片橙子还是有可能的,那我再问,你的芝麻面包上有多少颗芝麻?比如78颗,那为什么是78颗?这些问题的涵义是,有些事情没有为什么,它是随机结果的一个实现,它表现出这种实现只是因为存在表现出这种实现的可能性,而不一定是因为有人故意要表现出这种实现,毕竟你就算可以选晚饭吃几片橙子,你也不能选择你的芝麻面包上有多少颗芝麻。</p><hr><h2 id=\"关于Pascal’s-Wager\"><a href=\"#关于Pascal’s-Wager\" class=\"headerlink\" title=\"关于Pascal’s Wager\"></a>关于Pascal’s Wager</h2><p>Pascal’s Wager大致的含义是:世界可能有两种状态,要么有上帝,要么没有上帝;而每个人有两种选择,要么信仰基督教,要么不信仰基督教。如果有上帝,那么信仰基督教比不信仰基督教要好;如果没有上帝,那么信仰基督教也不会比不信仰基督教更差。</p><p>然而我们很容易注意到,Pascal’s Wager适用于所有宗教、信仰,比如伊斯兰教只需要把上帝换成真主安拉,佛教则换成佛祖,仍然讲得通。所以问题就出在,如果没有上帝,是不是信仰基督教就没有风险呢?如果实际上控制世界的是撒旦,在审判日所有信仰基督教的人都要被处死呢?你可以说上帝可能存在,我也可以说撒旦也可能存在,这样一来信仰基督教就并不总是最好的选择了,Pascal’s Wager也不攻自破了。(当然,Pascal’s Wager作为概率论和决策论的一个例子还是有一定意义的。)</p>"},{"title":"数据深扒:纪录片《穹顶之下》(无图版)","path":"2015/08/21/underthedome-np/","permalink":"http://sx349.github.io/2015/08/21/underthedome-np/","content":"<center><span style=\"font-style:bold;font-size:28px\">本文为无图版,<a href=\"/2015/08/21/underthedome\">有图版</a>请在Wi-Fi环境下查看</span></center><hr><p>本文片源:<a href=\"http://t.cn/Rwjlj0M\" target=\"_blank\" rel=\"noopener\">Youtube</a><br>参考片源:国内片源全部已跪,请科学上网。<br>初稿完成:2015-03-12 @ 有道云笔记<br>最近修改:2015-08-21 @ 盗泉<br>欢迎转载,转载请注明原作者信息。</p><hr><p>视频时间:00:03:19</p><p>文字引用:</p><blockquote><p>山西全境<br>88.4%的河流断面遭到污染<br>62%的断面失去功能</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中华人民共和国环境保护部:<a href=\"http://t.cn/Rw3tiDw\" target=\"_blank\" rel=\"noopener\">山西从污染困境突围</a><blockquote><p>全省26条主要河流中,有<strong>88.4%的断面受到不同程度的污染,有62%的断面失去了使用功能</strong>,主要城市集中式生活饮用水源地水质有17.2%不达标。</p></blockquote></li></ul><hr><p>视频时间:00:04:23</p><p>文字引用:</p><blockquote><p>(北京市2014年)污染天数175天</p></blockquote><p>数据来源:北京市环保局</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>北京市环保局:<a href=\"http://t.cn/RwYnOTH\" target=\"_blank\" rel=\"noopener\">2014年北京市污染减排力度大空气质量持续改善</a><blockquote><p>2014年PM2.5一级优天数达到93天,较2013年的71天大幅度增加22天,增幅31%;PM2.5导致的重污染天数也明显减少,从2013年的58天下降为45天,减少13天,降幅22%。</p></blockquote></li><li>国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a><br>根据全国城市空气质量日报的数据进行统计,北京市2014年的天气质量分别是:优28天 良139天 <strong>轻度污染85天 中度污染65天 重度污染30天 严重污染15天</strong> (缺3天数据)</li></ul><hr><p>视频时间:00:04:51</p><p>文字引用:</p><blockquote><p>天津2014年空气污染天数197天<br>沈阳2014年空气污染天数152天<br>兰州2014年空气污染天数112天<br>石家庄2014年空气污染天数264天</p></blockquote><p>数据来源:国家环境保护部</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a><br>数据基本吻合</li></ul><hr><p>视频时间:00:07:05</p><p>旁白引用:</p><blockquote><p>“这是世界卫生组织的标准(25),这个是美国的(37.5),这个是欧洲(50),这个是中国(75);那么我们在当天呼吸进去的是多少呢?几乎是中国标准值的五倍。”</p></blockquote><p>数据来源:《环境空气质量标准》GB 3095-2012</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li><a href=\"http://t.cn/Rw3YVjf\" target=\"_blank\" rel=\"noopener\">《环境空气质量标准》GB 3095-2012</a></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>其它标准</li></ul><table><thead><tr><th align=\"center\">组织或国家</th><th align=\"center\">标准</th><th align=\"center\">数值</th></tr></thead><tbody><tr><td align=\"center\">世界卫生组织</td><td align=\"center\"><a href=\"http://t.cn/hxqwX\" target=\"_blank\" rel=\"noopener\">《空气质量准则》</a></td><td align=\"center\">25</td></tr><tr><td align=\"center\">欧盟</td><td align=\"center\"><a href=\"http://t.cn/S7SVpf\" target=\"_blank\" rel=\"noopener\">Air Quality Standard</a></td><td align=\"center\"><strong>25(年均)</strong></td></tr><tr><td align=\"center\">美国</td><td align=\"center\"><a href=\"http://t.cn/Rw3YDMm\" target=\"_blank\" rel=\"noopener\">National Ambient Air Quality Standard</a></td><td align=\"center\"><strong>35</strong></td></tr></tbody></table><ul><li>欧盟的<strong>50是PM10的24小时平均浓度限值</strong>,PM2.5并未规定24小时平均浓度限值</li><li>美国的<strong>37.5</strong>可能是误用了<strong>世界卫生组织的过渡时期目标-3</strong></li><li>305.91/75 ≈ 4.0788,因此检测数据是中国标准值的<strong>4倍左右</strong>,并未达到5倍</li></ul><hr><p>视频时间:00:07:34</p><p>文字引用:</p><blockquote><p>苯并[a]芘benzo[a]pyrene 176<br>超标14倍</p></blockquote><p>旁白引用:</p><blockquote><p>“其中,世界上最强的致癌物(之一)苯并[a]芘,它是国家标准值的多少倍呢?14倍。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>苯并[a]芘的<strong>相对分子量为252</strong>,因此176不是指相对分子质量</li><li>根据<a href=\"http://t.cn/Rw3YVjf\" target=\"_blank\" rel=\"noopener\">《环境空气质量标准》GB 3095-2012</a>,苯并[a]芘24小时<strong>平均浓度限值为0.0025 μg/m<sup>3</sup></strong>;因此176也不是检测数据</li></ul><hr><p>视频时间:00:08:20</p><p>旁白引用:</p><blockquote><p>“所以如果它打开的话,两克黑碳能有多大?能有整个篮球场那么大;所以它可以吸附很多的致癌物和重金属。”<br>数据来源:中科院大气物理研究所</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>“Characterisation and evaluation of reference materials for black carbon analysis using elemental composition, colour, BET surface area and 13C NMR spectroscopy” <a href=\"http://t.cn/Rw3TmSx\" target=\"_blank\" rel=\"noopener\">(Hammes et al., 2008)</a><br>标准篮球场面积为420m<sup>2</sup>,而上表中列出的富黑碳材料比表面积最高值<strong>仅为73.9m<sup>2</sup>/g</strong>。</li><li>“Surface Area and Properties of Carbon Black”<a href=\"http://t.cn/Rw3Tx7x\" target=\"_blank\" rel=\"noopener\">(Smith et al., 1941)</a><br>另外一种称为炭黑(Carbon Black)的物质,其比表面积<strong>最高可以达到990m<sup>2</sup>/g</strong>,尚不清楚该物质是否等同于黑碳。</li></ul><hr><p>视频时间:00:08:33</p><p>旁白引用:</p><blockquote><p>“在中国这样的黑碳有多少呢?这是2009年NASA(美国航空航天局)做的一个测算,那个紫得发白发亮的地方是中国,(黑碳)它像幽灵一样在我们的上空飘荡。”</p></blockquote><p>数据来源:美国国家航空航天局(NASA)</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/RwQGtLm\" target=\"_blank\" rel=\"noopener\">Jun</a>原答4(回答已删除)<br>提供来源 <a href=\"http://t.cn/RwQoCGR\" target=\"_blank\" rel=\"noopener\">Scientific Visualization Studio: Atmospheric Black Carbon Density</a></li><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第一处<blockquote><p>《穹顶之下》的这幅图取自2009年8月1日到11月19日的全球黑炭或烟尘的运动轨迹。</p></blockquote></li></ul><hr><p>视频时间:00:12:31</p><p>旁白引用:</p><blockquote><p>“这个是中科院给我们提供的测算,这个复杂的图表显示的是当PM2.5的值升高的时候,人群的死亡率是随之上升的。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/Rw8OlAJ\" target=\"_blank\" rel=\"noopener\">宋宁世</a></li></ul><ol><li>论文来源 “Time-series analysis of mortality effects from airborne particulate matter size fractions in Beijing”<a href=\"http://t.cn/Rw8Ea1x\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a></li><li>红色数据点没有进行实质的拟合。</li><li>论文中的原图表中标注了PM2.5的数值,但2006~2009年北京市年均PM2.5的数值是减少的,与纪录片的结论相悖。</li><li>环境卫生学中评定污染物健康危害更常用的是Dose response curve,根据论文中提供的图表,PM2.5值超过一定数值后,相对健康风险的数值反而下降了。</li><li>根据论文提供的结论:在不同国家所进行的环境卫生学调查的结果显示出了不一样的现象,在北京,PM2.5对人的健康带来的负面影响反而比发达国家的结果要小。</li><li>综上所述,纪录片制作团队或者并没有去查证过自己数据的来源及可靠性,或者没有具备验证数据可靠性的知识水平,又或者刻意对自己的数据进行了处理/刻意隐藏不利于自己的数据</li></ol><hr><p>视频时间:00:12:51</p><p>旁白引用:</p><blockquote><p>“我看到中国的前卫生部部长陈竺的测算结果是,在中国,每年因为大气污染过早死亡的人数是50万人。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>陈竺发表在《柳叶刀》The Lancet上的文章<br>“China tackles the health effects of air pollution”<a href=\"http://t.cn/Rw3Hi6E\" target=\"_blank\" rel=\"noopener\">(Chen et al., 2013)</a><blockquote><p>“Studies by the World Bank, WHO, and the Chinese Academy for Environmental Planning on the effect of air pollution on health concluded that between <strong>350 000 and 500 000 people die prematurely each year as a result of outdoor air pollution in China</strong>.”</p></blockquote></li><li>21世纪经济报道:<a href=\"http://t.cn/8Fv0WTR\" target=\"_blank\" rel=\"noopener\">陈竺:中国每年因空气污染导致早死35万-50万人</a><br>总结:陈竺的文章发表于The Lancet 《柳叶刀》,其结果远低于《柳叶刀》于2012年12月14日发表的《全球疾病负担2010年报告》的评估,后者提出2010年中国因室外PM2.5污染导致120万人早死以及2500万伤残调整寿命年损失;前者认为《报告》是目前国际上对中国室外空气污染健康影响最严重的估计,但可能高估了中国室外空气污染的健康影响;而《报告》的作者之一、中国疾病预防控制中心原副主任杨功焕认为前者所引用的报告主要考虑了PM10及部分PM2.5数据,存在不确定性。</li></ul><hr><p>视频时间:00:13:33</p><p>文字引用:</p><blockquote><p>2013年1月重雾霾期间,27个城市出现急诊的人数爆发增加10%~150%不等,以儿童和老人为主。</p></blockquote><p>数据来源:北京大学环境科学与工程学院环境与经济研究所</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《2013年1月中国大面积雾霾事件直接社会经济损失评估》<a href=\"http://t.cn/Rw3HNng\" target=\"_blank\" rel=\"noopener\">(穆泉,张世秋,2013)</a><blockquote><p>“根据对2013年1月份期间发生过严重雾霾事件的20个省市雾霾造成医院门/急诊相关的公开报道数据的统计,全国14个省市所属的27个城市有出现因雾霾导致医院门/急诊人数激增的报道,门/急症看病人数增加比例<strong>从10%到150%不等,并且就诊人群主要为儿童和老人</strong>。”</p></blockquote></li></ul><hr><p>视频时间:00:14:21</p><p>论文引用:</p><blockquote><p>《Effects of Airborne Metals on Lung Function in Inner Mongolian Schoolchildren》<br>《沙棘油对SO2染毒运动训练大鼠心肌脂质过氧化的影响》<br>《中韩儿童肺活量与影响因素的比较研究》<br>《非沙尘天气大气颗粒物中金属元素浓度与儿童肺功能相关性的初步研究》</p></blockquote><p>旁白引用:</p><blockquote><p>英文:“It has been shown in several studies around the world now that children in cleaner places grow faster than children grown up in a more polluted area. Poor lung function early in life is predictive of lung problems later on in life.”<br>中文字幕:“目前世界各国的研究表明,在更干净的地方长大的孩子的肺生长速度比污染地区长大的孩子要快。年幼时发育不全的肺功能,很可能导致成长后的肺部疾病。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>这4篇论文的相关信息分列如下:</li></ul><ol><li>“Effects of Airborne Metals on Lung Function in Inner Mongolian Schoolchildren”<a href=\"http://t.cn/Rwd9LEh\" target=\"_blank\" rel=\"noopener\">(Madaniyazi et al., 2013)</a><br>(见第四篇论文分析)</li><li>《沙棘油对SO2染毒运动训练大鼠心肌脂质过氧化的影响》<a href=\"http://t.cn/Rw8Y9lb\" target=\"_blank\" rel=\"noopener\">(谢奇,沈一岚,2011)</a><br>本文与雾霾<strong>完全无关</strong>。其结论为:“SO2污染的运动环境下服用沙棘油能改善SO2对大鼠心肌细胞的损伤,提高心肌组织抗自由基氧化的能力。”</li><li>《中韩儿童肺活量与影响因素的比较研究》<a href=\"http://t.cn/Rw8ThL3\" target=\"_blank\" rel=\"noopener\">(金大善等,2011)</a><br>本文结论是:“在潜在影响因素中,<strong>室内是否饲养宠物及家庭年平均收入</strong>对女生肺功能指标的影响有统计学意义(P<0.05),其他指标的影响均无统计学意义;在控制身高、体重、年龄后,中国学生FVC和FEV1与韩国学生相比,男生高0.015L和0.026L,女生高0.128L和0.136L。”</li><li>《非沙尘天气大气颗粒物中金属元素浓度与儿童肺功能相关性的初步研究》<a href=\"http://t.cn/Rw8T5Iu\" target=\"_blank\" rel=\"noopener\">(丽娜·马达尼亚孜等,2011)</a><br>本文同第一篇论文采用的数据部分重合,实验设计和计量方法完全一致,结论相近。其结论为:“学龄儿童日均PEFR值的下降与大气颗粒物中金属元素的含量的升高有关。”“本次研究结果提示<strong>大气颗粒物中的金属元素可能与学龄儿童PEFR之间存在负相关性</strong>,并且滞后效应更为明显。本次研究还发现细颗粒物中的金属元素对学龄儿童肺功能的影响更明显。”其中PEFR(最大呼气流速)是肺功能的一项指标。</li></ol><ul><li>这四篇论文中,第二篇论文完全是因为<strong>恰好与第三篇论文的结尾段共用一页而被误引用</strong></li><li>另外三篇均为北京大学公共卫生学院劳动卫生与环境卫生学系潘小川教授所带研究生撰写的论文,其中第三篇<strong>与空气污染基本无关</strong>,第一篇、第四篇<strong>基本为相同内容</strong>,虽然论文结论和字幕内容相关,但并不完全相符</li><li>此外,在第一篇选择英文论文有<strong>故意误导观众这些研究是来自“世界各国”的嫌疑</strong></li></ul><hr><p>视频时间:00:17:00</p><p>文字引用:</p><blockquote><p>随着颗粒物和空气污染暴露的不断提高,将增加人类患肺癌的风险,颗粒物被列为人类一级致癌物。</p></blockquote><p>旁白引用:</p><blockquote><p>“世卫组织告诉我,颗粒物是被列为人类的一级致癌物。”</p></blockquote><p>数据来源:IARC Scientific Publication NO.161</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/zRVOC5w\" target=\"_blank\" rel=\"noopener\">IARC Scientific Publication NO.161</a> 中<strong>完全没有与纪录片所引文字一致或相近的论述</strong>。</li><li>纪录片所引用的可能来自将颗粒物列为人类一级致癌物的新闻报道<a href=\"http://t.cn/zRV0nGx\" target=\"_blank\" rel=\"noopener\">IARC: Outdoor air pollution a leading environmental cause of cancer deaths</a><blockquote><p>“Particulate matter, a major component of outdoor air pollution, was evaluated separately and was also <strong>classified as carcinogenic to humans (Group 1)</strong>. ”<br>“The IARC evaluation showed <strong>an increasing risk of lung cancer with increasing levels of exposure to particulate matter and air pollution</strong>.”</p></blockquote></li></ul><hr><p>视频时间:00:17:17</p><p>旁白引用:</p><blockquote><p>“我就向NASA(美国国家航天局)提出了一个申请,能不能你们把这十年来华北上空的卫星图片发给我,这是我收到的:2012年,北京的上空;2011年,12月的北京;2010年,华北上空;2009年,华北上空;再看看十年前,2005年的华北。”</p></blockquote><p>数据来源:美国航天航空局(NASA)</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第二处</li></ul><ol><li>提供了一篇参考文章:NASA中文微信公众号:<a href=\"http://t.cn/RwRmg3x\" target=\"_blank\" rel=\"noopener\">《这些年,我们遇到的雾霾》</a></li><li>提供了2001年至2006年全球PM2.5的平均值地图,对于发展中国家(包括中国)来说,是第一张PM2.5卫星估计值图,这也可以解释雾霾不是最近出现的。</li></ol><ul><li>知乎:<a href=\"http://t.cn/RwQg0Bt\" target=\"_blank\" rel=\"noopener\">唐文韬</a> 第2点</li></ul><ol><li>目前,就答主所知,关于雾霾历史记录的研究最可靠的是能见度加上相对湿度的记录;其次Shortwave irradiance也可以作为气溶胶研究的侧面证据,但是不能用于直接研究雾霾。</li><li>视频中使用的这种图更是不能用于确认雾霾的:雾和霾都是根据能见度和相对湿度进行定义的,这里看不出任何相关信息,也分不出是云还是雾霾。<br>(但上述微信号文章中提到“亮白色的是云雾,暗灰色的是遭到污染的大气”,结论相悖)</li></ol><hr><p>视频时间:00:17:56</p><p>旁白引用:</p><blockquote><p>“这是2004年,十一年前,那时候我们还没有PM2.5的检测,但是我们有PM10。她说根据当时的估算,那个时候,在污染期的话,PM2.5可以达到三百到四百,属于今天的严重污染。”</p></blockquote><p>数据来源:北京大学环境科学系</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据,但根据《北京市大气PM10环境背景值的计算方法探讨》<a href=\"http://t.cn/Rw3exWR\" target=\"_blank\" rel=\"noopener\">(李金香等,2007)</a>,可知当时北京市北部和南部共有5个清洁站点确实已经可以监测PM10数据。</li></ul><hr><p>视频时间:00:20:16</p><p>旁白引用:</p><blockquote><p>“从科研机构的PM2.5站点数据来看,2006年年均PM2.5浓度要高于今天。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>根据中科院论文”Time-series analysis of mortality effects from airborne particulate matter size fractions in Beijing”<a href=\"http://t.cn/Rw8Ea1x\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a><br>2006年年均PM2.5浓度为84μg/m<sup>3</sup></li><li>人民网:<a href=\"http://t.cn/RZtZzMi\" target=\"_blank\" rel=\"noopener\">北京2014年PM2.5年均浓度同比下降4.0%</a><br>2014年年均PM2.5浓度为85.9μg/m<sup>3</sup></li></ul><hr><p>视频时间:00:20:26</p><p>旁白引用:</p><blockquote><p>“此后,主要大气污染物开始整体下降,但是,对能见度能够产生影响的细颗粒物增加了,极端重度污染天数增加了,人们对于污染感受强烈。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>上表:中国环境与发展国际合作委员会2012年年会<a href=\"http://t.cn/Rw1vDS2\" target=\"_blank\" rel=\"noopener\">《区域空气质量综合控制体系研究》研究报告</a><blockquote><p>“为进一步了解北京PM2.5化学物种的长期变化,采用相同方法重构了1999年7月至2008年7月清华观测点PM2.5可鉴别物种的季均浓度(图 1-13),其中最为突出的特征是PM2.5中二次成分比例持续增长。”</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>下表:未找到统计数据,或可根据国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a>进行统计而得。</li></ul><hr><p>视频时间:00:20:44</p><p>旁白引用:</p><blockquote><p>“过去30年内,我国的肺癌死亡率上升了465%,虽然吸烟和老龄化仍然是这个数字的主要因素,但细颗粒物明确的致癌风险越来越得到重视。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>华商报:<a href=\"http://t.cn/8kPZDNm\" target=\"_blank\" rel=\"noopener\">中国或成肺癌第一大国 过去30年发病率上升465%</a><br>虽然数据无误,但本文<strong>完全没有提到细颗粒物的致癌风险</strong>,只笼统地在建议中说“对于没有吸烟史的非高危人群,要想科学预防肺癌,除了要做到拒绝被动吸烟、重视并积极治疗肺部慢性疾病、减少室内外空气污染、家庭装修后保持通风换气之外,也应每年进行定期体检筛查”。</li></ul><hr><p>视频时间:00:22:52</p><p>旁白引用:</p><blockquote><p>“中科院告诉我它的来源是什么:燃煤、燃油、生物质燃烧、工业、养殖、化肥、扬尘。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据来源;</li><li>中科院大气物理研究所王跃思研究员的<a href=\"http://t.cn/Rw14Pf5\" target=\"_blank\" rel=\"noopener\">相关结果</a></li><li>《科学世界》对王跃思研究员访谈稿中另一<a href=\"http://t.cn/Rw1Guv0\" target=\"_blank\" rel=\"noopener\">相关结果</a></li></ul><hr><p>视频时间:00:23:06</p><p>旁白引用:</p><blockquote><p>“这就是结论,中国的PM2.5百分之六十来自于燃煤和燃油,也就是化石能源的燃烧。”</p></blockquote><p>数据来源:中科院大气物理研究所、清华环境科学与工程系、北大环境科学系</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据;根据上一条的数据,若认为燃煤和燃油大致即为燃煤和机动车,则比例大约为<strong>46%~68%不等</strong>。</li></ul><hr><p>视频时间:00:23:58</p><p>旁白引用:</p><blockquote><p>“这种燃烧的强度有多大呢?我们可以来看看,在2014年,这个数值可以代表全球化石能源燃烧的燃烧强度,你可以看看中国有多亮,那个烧得发红发亮发白的地方就是我们国家。从图表来看的话,它比欧洲高出三到四倍。”</p></blockquote><p>数据来源:美国航空航天局(NASA)</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/Rw8Lp8k\" target=\"_blank\" rel=\"noopener\">Luke liking</a><br>提供原始视频连接:<a href=\"http://t.cn/RwRmJaK\" target=\"_blank\" rel=\"noopener\">Nitrogen Dioxide from Aura/OMI, 2013-2014</a>,答主已搬运至<a href=\"http://t.cn/RwRmXqk\" target=\"_blank\" rel=\"noopener\">优酷</a></li><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第三处<blockquote><p>这段视频取自2013年7月至2014年7月AURA/OMI卫星上对流层二氧化氮的数据。</p></blockquote></li></ul><hr><p>视频时间:00:25:07</p><p>旁白引用:</p><blockquote><p>“你们知道中国烧了多少煤吗?2013年的时候,我们已经是三十六亿吨了,但你知道全世界其它国家烧了多少煤吗?我们比全世界所有其他国家加起来烧的煤还要多。上一个达到这样消费量的是1860年的英国,但在之后他们为此付出了沉重的代价。”</p></blockquote><p>数据来源:中国煤炭工业协会</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>环球时报:<a href=\"http://t.cn/8FdNPc4\" target=\"_blank\" rel=\"noopener\">雾霾逼中国加速摆脱“煤依赖”</a><blockquote><p>2013年中国煤炭消费量为36.1亿吨,<strong>占世界煤炭消费总量的49.33%</strong>,而中国煤炭消费增量也占世界增量的30.43%。</p></blockquote></li></ul><hr><p>视频时间:00:27:19</p><p>旁白引用:</p><blockquote><p>“清洁空气联盟的测算显示,伦敦当时空气污染的程度可能远超中国当下,虽然当时并没有PM2.5的检测,但是大烟雾事件发生时,二氧化硫的浓度超过世界卫生组织标准的190倍,比中国PM2.5超标的倍数高出一个量级。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>在清洁空气联盟(CCAC)网站上未找到相关数据</li><li>实际来源可能是由伦敦市出版的伦敦大烟雾事件纪念册:<a href=\"http://t.cn/zYt0WAt\" target=\"_blank\" rel=\"noopener\">“50 years on The struggle for air quality in London since the great smog of December 1952”</a><blockquote><p>“Sulphur dioxide (SO2) followed a similar pattern with concentrations rising from 0.41 mg/m<sup>3</sup> on 4 December to 2.15 mg/m3 on the 5th and to <strong>3.83 mg/m<sup>3</sup></strong> on both the 7 and 8 December.”</p></blockquote></li><li>世界卫生组织<a href=\"http://t.cn/hxqwX\" target=\"_blank\" rel=\"noopener\">《空气质量准则》</a>规定SO2的24小时平均浓度限值为20 μg/m<sup>3</sup>,3.83mg=3830μg,即<strong>190倍左右</strong>。</li></ul><hr><p>视频时间:00:27:45</p><p>旁白引用:</p><blockquote><p>“所以在1960年代,在这个大烟雾事件发生完之后,其他的国家纷纷开始减少和控制自己的煤炭用量,但当时恰逢中国改革开放的开始,这个已经封闭了多年和落后了多年的国家,迫切地需要一种巨大的能量来让自己起飞,它的选择是煤炭,这是它增长的曲线。”<br>数据来源:美国自然资源保护协会(NRDC)</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>在NRDC官网以关键词”coal”搜索未找到任何相关数据</li><li>未找到“煤炭占全社会排量百分比”曲线</li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 9-2 能源消费总量及构成<br>煤炭<strong>占我国能源消费总量</strong>的比重自1978年至今始终处于70%上下,最低为66.0(2013年),最高为76.2(1990年),<strong>占全社会排量</strong>比重不应当出现不降反升的情况。</li></ul><hr><p>视频时间:00:28:15</p><p>旁白引用:</p><blockquote><p>“那中国这么多煤用在了哪儿呢?2013年的这三十六亿吨,我们可以看一下,它其中三亿八千万吨烧在了京津冀,而这三亿八千万吨当中,有三个亿烧在河北。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>未找到2013年相关数据,2012年数据参见:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a><br>2012年煤炭消费量为:北京2270万吨,天津5298万吨,河北31359万吨,合计38927万吨,基本无误。</li></ul><hr><p>视频时间:00:32:43</p><p>文字引用:</p><blockquote><p>北京强霾天气外来污染源输送占1/5</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国气象局:<a href=\"http://t.cn/Rwrkg4x\" target=\"_blank\" rel=\"noopener\">北京强霾天气外来污染源输送占1/5</a></li></ul><hr><p>视频时间:00:33:23</p><p>旁白引用:</p><blockquote><p>“在你们江苏,每三十公里就有一个电厂。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>冰冷雨天:<a href=\"http://t.cn/RwdxVk2\" target=\"_blank\" rel=\"noopener\">柴静老师,请注意一点细节</a><blockquote><p>实际上这个“三十公里”的说法出于去年两会时全国政协委员、国家电网公司董事长、党组书记刘振亚的一段发言。<br>刘振亚的发言原文是这样的:“特别是近年来东中部地区雾霾严重,给人民群众生产生活、身心健康造成严重危害。目前京津冀鲁、长三角地区单位国土面积煤电装机分别是西北地区的13倍、26倍,<strong>沿长江平均30公里一个电厂</strong>,从南京到镇江平均10公里一个电厂,严重超出环境承载能力”。<br>刘振亚的这段话就不太有说服力,比如从武汉到南京600公里长江,有20个发电厂吗?而柴静老师再一次加以改造,成了“江苏省内”这就不应该了。而且从“长江沿岸”到“江苏省内”就是从直线概念升级到了平面概念,这时只能说“每平方公里多少”,而不能还说“隔多少公里一个”了。</p></blockquote></li></ul><hr><p>视频时间:00:33:43</p><p>旁白引用:<br>“而被夹在中间的上海,它自己烧的煤也不少,它是单位燃煤量最高的国际性大都市,每一平方米的土地上要烧掉十公斤煤炭。”</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到关于“单位燃煤量最高的国际性大都市”的相关数据</li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>上海的单位城区面积燃煤量参见:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a><blockquote><p>城区面积:6341平方公里<br>2012年煤炭消费量:5703万吨<br>单位面积煤炭消费量:8.99公斤/平方米</p></blockquote></li></ul><hr><p>视频时间:00:34:29</p><p>旁白引用:</p><blockquote><p>“那这样的褐煤我们用了多少呢?这是这些年来我们生产它的曲线。”</p></blockquote><p>数据来源:中国煤炭工业协会</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>未找到中国煤炭工业协会的数据</li><li><a href=\"http://t.cn/Rw1umEf\" target=\"_blank\" rel=\"noopener\">KNOEMA世界数据图册</a></li></ul><hr><p>视频时间:00:35:34</p><p>旁白引用:</p><blockquote><p>“煤是可以弄干净的,英国人是百分之九十五以上都给它洗干净,那么在中国我们洗了多少呢?我们才洗了不到一半的煤,剩下的另外一半当中基本上有几亿吨是没有任何用,而且又污染又浪费的叫灰分的东西。我们就这么把它拉着,用汽车拉用火车拉,满世界地跑,再放到煤炉子把它烧出来,毫无作用。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>中华商务网:<a href=\"http://t.cn/Rw13v4v\" target=\"_blank\" rel=\"noopener\">我国煤炭入洗发展空间巨大</a><blockquote><p>美国原煤入洗率在60%左右,南非和俄罗斯在60%以上,澳大利亚和<strong>英国入洗率在75%以上</strong>,加拿大和德国为95%,而<strong>我国(2002年)只有34%</strong>。</p></blockquote></li><li>中国能源报:<a href=\"http://t.cn/Rw13vDf\" target=\"_blank\" rel=\"noopener\">国家将不再新增煤制气项目</a><blockquote><p>中国煤炭工业协会的数据显示,2013年<strong>我国原煤入洗率仅为59.8%</strong>。</p></blockquote></li><li>百度知道:<a href=\"http://t.cn/Rw13PTP\" target=\"_blank\" rel=\"noopener\">燃煤脱硫的简史及其发展</a><blockquote><p><strong>我国当前的煤炭入洗率较低,大约在20%左右</strong>,而美国为42%,<strong>英国为94.9%</strong>,法国为88.7%,日本为98.2%。</p></blockquote></li></ul><hr><p>视频时间:00:36:03</p><p>旁白引用:</p><blockquote><p>“即使在中国城市里面,我们现在都有四分之一的人是在烧散煤,他们烧的量不算大,只有百分之二十,但是这百分之二十排放出来的二氧化硫跟所有的大电厂加起来一样多,就这么厉害。”</p></blockquote><p>数据来源:环保部环境与经济政策研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到与“四分之一”相关的数据</li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>新华网:<a href=\"http://t.cn/RP5cvS5\" target=\"_blank\" rel=\"noopener\">“煤改气”难成我国雾霾防治“撒手锏”</a><blockquote><p>据环保部环境与经济政策研究中心研究员王汉臣估算,工业锅炉、家庭取暖、餐饮用煤等“散煤”的煤炭消耗量在6亿吨-7亿吨,<strong>占全国煤炭消耗量的20%</strong>,仅次于电力行业,在非电行业最多,排放二氧化硫接近1000万吨,<strong>与电力行业的二氧化硫齐平</strong>,排放氮氧化物是320多万吨,位列第三,仅次于电力和机动车,高于水泥行业排放总量。</p></blockquote></li></ul><hr><p>视频时间:00:36:17</p><p>旁白引用:</p><blockquote><p>“最要命的是,它的毒性还非常大,在北京,冬天PM2.5的毒性远远大于夏天,光致癌物的含量,就是夏天的25倍。它为什么会这么高,就是因为散烧煤炭是主要原因之一,它的致癌物的排放因子是非常高的。”</p></blockquote><p>数据来源:北京大学环境科学和工程学院</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据来源,可参考:</li></ul><ol><li>《北京市大气颗粒物中多环芳烃的组成》<a href=\"http://t.cn/RwdomQn\" target=\"_blank\" rel=\"noopener\">(黄翠玲等,2001)</a></li><li>《北京市部分地区大气气溶胶中多环芳烃污染特征及污染源探讨》<a href=\"http://t.cn/Rwdo3Qb\" target=\"_blank\" rel=\"noopener\">(曾凡刚等,2002)</a></li><li>“Seasonal Variation and Spatial Distribution of Polycyclic Aromatic Hydrocarbons in Atmospheric PM10 of Beijing, People’s Republic of China” <a href=\"http://t.cn/RwdoBlz\" target=\"_blank\" rel=\"noopener\">(Zhou et al., 2005)</a></li><li>“Annual variation of particulate organic compounds in PM2.5 in the urban atmosphere of Beijing”<a href=\"http://t.cn/Rwdodc1\" target=\"_blank\" rel=\"noopener\">(Huang et al., 2006)</a></li><li>“Molecular distribution and seasonal variation of hydrocarbons in PM2.5 from Beijing during 2006”<a href=\"http://t.cn/RwdoD30\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a></li><li>《京津冀地区城市空气颗粒物中多环芳烃的污染特征及来源》<a href=\"http://t.cn/Rwdosbg\" target=\"_blank\" rel=\"noopener\">(王超等,2015)</a></li></ol><ul><li>综合以上文献,北京市不同检测点冬季与夏季PAHs含量的比值在<strong>2至几百之间不等</strong></li></ul><hr><p>视频时间:00:37:42</p><p>旁白引用:</p><blockquote><p>“在中国,煤炭消耗量越多的地方,它的PM2.5的浓度几乎也就是越高。”</p></blockquote><p>数据来源:国家发改委气候战略研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>左图:未找到直接来源;或可根据:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a> 推算。</li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>右图:“Global Estimates of Ambient Fine Particulate Matter Concentrations from Satellite-Based Aerosol Optical Depth: Development and Application”<a href=\"http://t.cn/zjD8Mxs\" target=\"_blank\" rel=\"noopener\">(Aaron van Donkelaar et al., 2010)</a><br>(单位应当是<strong>微克/立方米而非毫克/立方米</strong>)</li></ul><hr><p>视频时间:00:38:24</p><p>文字引用:</p><blockquote><p>2014年<br>74个主要城市<br>PM2.5的年均浓度平均下降了11.1%<br>但能达到国家标准的只有8个城市</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>新京报:<a href=\"http://t.cn/Rw5dmEI\" target=\"_blank\" rel=\"noopener\">去年74个城市PM2.5平均下降11%</a></li></ul><hr><p>视频时间:00:39:09</p><p>旁白引用:</p><blockquote><p>“我们的车在十年之间增加了将近一个亿,光北京2010年一年我们增加的车是多少呢?八十万。”</p></blockquote><p>数据来源:公安部交通管理科学研究所</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 18-25 民用汽车拥有量</li><li>北京晨报:<a href=\"http://t.cn/Rw1dIQK\" target=\"_blank\" rel=\"noopener\">治堵这一年 摇号减少60万新车上路</a><blockquote><p>2010年,北京新增汽车<strong>81万辆</strong>。</p></blockquote></li></ul><hr><p>视频时间:00:39:31</p><p>旁白引用:</p><blockquote><p>“北京本地的污染源当中,最大的就来自机动车。”</p></blockquote><p>数据来源:北京市环保局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>北京市环保局:<a href=\"http://t.cn/8sOWPUL\" target=\"_blank\" rel=\"noopener\">北京市PM2.5来源解析正式发布</a></li><li>知乎:<a href=\"http://t.cn/Rw8gytg\" target=\"_blank\" rel=\"noopener\">大脸撑在小胸</a></li></ul><ol><li>原答主Jun引用<a href=\"http://t.cn/8kBk1xT\" target=\"_blank\" rel=\"noopener\">中国科学院大气物理研究所张仁健研究员的论文</a>,来说明机动车对北京PM2.5的影响并非31.1%,而是4%</li><li>原答主未考虑机动车间接导致的空气污染,除了汽车尾气以外,一半的二次无机气溶胶和一半的土壤尘也由机动车造成,综合考虑后根据原答主提供的论文,也可推算出机动车对PM2.5的贡献<strong>在20-30%左右</strong>。</li><li>柴静的数据图没有问题,问题在于她对“机动车”这一项的理解。</li></ol><hr><p>视频时间:00:39:35</p><p>旁白引用:</p><blockquote><p>“某城市(杭州)一年的雾霾日达到两百天以上。”<br>“因为它(杭州)是全国人均机动车保有量最大的城市,平均两个人一辆车,所以它的机动车污染要占到(接近)百分之四十。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>中国新闻网:<a href=\"http://t.cn/Rw1dyjR\" target=\"_blank\" rel=\"noopener\">杭州机动车保有量破110万 后市场混乱或迎洗牌</a><blockquote><p>截至2013年底,浙江省杭州市机动车保有量突破110万辆,相当于<strong>每三人拥有一辆车</strong>,是国内<strong>人均汽车拥有量最高的城市</strong>。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>钱江晚报:<a href=\"http://t.cn/RwjkfUr\" target=\"_blank\" rel=\"noopener\">2014年,杭州有154个雾霾天</a><blockquote><p>2013年1到10月,据气象部门统计,杭州主城区出现了<strong>209天雾霾天</strong><br>杭州曾做过一份《杭州市大气灰霾成因及关键污染因子预防控制研究》,其中说到机动车尾气、燃煤、扬尘和其他,是杭州PM2.5的主要来源。大致的比例是这样,<strong>机动车尾气占近40%</strong>,燃煤占33%,扬尘占8%,其他则占19%。</p></blockquote></li></ul><hr><p>视频时间:00:40:19</p><p>旁白引用:</p><blockquote><p>“大家经常听到的一个问题是,东京不是比北京车多吗,我们怎么没有看到这么严重的污染呢?北京市交通委给我的解释是这样的:东京百分之九十以上的人在坐轨道交通,他们只有不到百分之六的人在开车。北京有多少呢?北京有百分之三十四的人开车,而且是在中心城区基本这么堵着,一堵的时候一怠速排放是两倍以上,一刹车重金属的排放会更多。”</p></blockquote><p>数据来源:北京交通发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国汽车报:<a href=\"http://t.cn/8sq57xA\" target=\"_blank\" rel=\"noopener\">李春利:治理交通拥堵可借鉴“东京模式”</a><blockquote><p>目前,东京是世界上轨道交通最发达的城市之一,几十条城市铁路和地铁组成的立体公共交通网覆盖整个东京圈,总里程超过2300公里,轨道交通系统每天运送旅客2000多万人次,承担东京全部客运量的86%。<strong>在早高峰时的市中心区, 有91%的人乘坐轨道交通工具,而乘坐小汽车的仅为6%</strong>。<br>出行方式的数据差距更大。<strong>北京总体的小汽车出行比例是34%</strong>,东京是13%,北京是东京的2.5倍。轨道交通出行东京达到了86%,北京是12%,东京是北京的7倍,这两个关键数据看出了北京和东京的差距所在。</p></blockquote></li><li>(虽然数据对上了,但是东京的数据用的是早高峰时,北京用的数据是整体出行比例,为什么不都用整体出行比例?)</li><li>Bureau of Urban development Tokyo Metropolitan Government: <a href=\"http://t.cn/RwdxHQI\" target=\"_blank\" rel=\"noopener\">Basic concept of public transport planning in Tokyo</a></li></ul><hr><p>视频时间:00:40:44</p><p>旁白引用:</p><blockquote><p>“所以在北京每天的高峰期的时候,六环以内每个小时的PM2.5排放量是多少?一吨。”</p></blockquote><p>数据来源:南开大学城市交通污染防治研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据,但可信度较高,参考报道:城市快报:<a href=\"http://t.cn/R7p2aDb\" target=\"_blank\" rel=\"noopener\">两分钟“画”出汽车“污染地图”</a><blockquote><p>日前,南开大学城市交通污染防治研究中心毛洪钧教授课题组研发了一套可完整“追踪”城市每条道路上机动车排放情况,并实时绘制出机动车“污染地图”的智能系统。该系统可为城市空气质量模型、雾霾预报预警系统提供详细的机动车“排放清单”,同时也可为政府限牌、限号、黄标车淘汰、路查路检、低排区限行等调控政策提供效果评估和改进建议。</p></blockquote></li></ul><hr><p>视频时间:00:40:53</p><p>文字引用:</p><blockquote><p>5公里以内44%<br>2公里以内12%<br>1公里以内7%</p></blockquote><p>数据来源:北京交通发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>北京交通发展研究中心:<a href=\"http://t.cn/RwdiLzx\" target=\"_blank\" rel=\"noopener\">中国城镇化机动化进程中的交通问题</a><br>数据应当为:1公里以内7%,<strong>2公里以内19%(7%+12%,而非12%)</strong>,5公里以内44%</li></ul><hr><p>视频时间:00:42:39</p><p>旁白引用:</p><blockquote><p>“在北京每一天在凌晨的时候,都会出现污染的峰值,而且是每天恒定出现,永远比当天下午的时候有机碳的排放大概要高出两倍。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“元素碳和有机碳”一手数据,可参考</li></ul><ol><li>《北京地区夏季PM10污染的数值模拟研究 》<a href=\"http://t.cn/RwdPS1B\" target=\"_blank\" rel=\"noopener\">(王雪松,李金龙,2003)</a><blockquote><p>结果表明,北京市城近郊区夏季的PM10具有明显的时空变化规律。一般在半夜前后和早晨常表现出较高的浓度……有机碳气溶胶以及PM10的浓度水平和变化规律则主要受一次成分的影响。</p></blockquote></li><li>“Estimation of diurnal and seasonal variations of primary and secondary organic carbon concentrations using highly time resolved PM2.5 measurements”<a href=\"http://t.cn/RwdPpLi\" target=\"_blank\" rel=\"noopener\">(Kim et al., 2011)</a><blockquote><p>Diurnal variations of the estimated SOCs show peaks at around 15:00 local time (LT) while POC, TOC and EC concentrations show peaks at morning and midnight.</p></blockquote></li><li>“Elemental and organic carbon in the urban environment of Athens. Seasonal and diurnal variations and estimates of secondary organic carbon”<a href=\"http://t.cn/RwdPHXr\" target=\"_blank\" rel=\"noopener\">(Grivas et al., 2012)</a><blockquote><p>Significant evening OC levels were observed during both periods. Organics from fresh afternoon traffic emissions, combined with persisting secondary particles, were <strong>trapped in the shallow boundary layer</strong>, resulting at high levels around midnight.</p></blockquote></li></ol><ul><li>纪录片在此处试图引导观众建立以下联系:北京凌晨的有机碳污染峰值是由于夜晚柴油车排放的缘故;而上述论文一定程度上反驳了这一观点。</li></ul><hr><p>视频时间:00:44:16</p><p>旁白引用:</p><blockquote><p>“所以柴油车的比例也不算高,它跟散煤差不多,并不多,只占百分之十七,但是它的氮氧化物的排放,要占所有机动车的百分之七十,它的一次性颗粒物的排放,要占到多少呢?99%。”</p></blockquote><p>数据来源:国家环保部《2013年机动车污染防治年报》</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>国家环保部:<a href=\"http://t.cn/RwrgFPs\" target=\"_blank\" rel=\"noopener\">《2013年机动车污染防治年报》</a><blockquote><p>柴油车<strong>占16.1%</strong><br>全国柴油车排放的NOx接近汽车排放总量的70%,<strong>PM超过90%</strong></p></blockquote></li></ul><hr><p>视频时间:00:48:55</p><p>文字引用:</p><blockquote><p>从2004年到2014年,车用柴油标准三次落后于机动车排放标准的实施。</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到视频中原表格,可参考</li></ul><ol><li><a href=\"http://t.cn/RwrkU4F\" target=\"_blank\" rel=\"noopener\">Global Motor Vehicle Emissions Regulations</a><br>欧盟、美国(加州/其它)、日本、阿根廷、巴西、智利、哥伦比亚、澳大利亚、香港、印度、韩国的部分排放标准</li><li>ICCT(The International Council on Clean Transportation): <a href=\"http://t.cn/Rwrk4VG\" target=\"_blank\" rel=\"noopener\">Global passenger vehicle standards</a><br>巴西、加拿大、中国、欧盟、印度、日本、墨西哥、韩国、美国的与乘用车相关的各类标准</li><li>TransportPolicy.net: <a href=\"http://t.cn/Rwrk55X\" target=\"_blank\" rel=\"noopener\">Global Comparison: Fuels</a><br>中国、印度、欧盟、美国的燃油标准对比;点击左侧“Fuels”可查阅澳大利亚 、巴西、加拿大、智利、中国、欧盟、印度、伊朗、日本、墨西哥、俄罗斯、韩国、美国(加州/其它)及国际标准(ISO)的燃油标准</li></ol><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国环境报:<a href=\"http://t.cn/RwrFMbD\" target=\"_blank\" rel=\"noopener\">油品标准制定权该由谁执牛耳?</a><blockquote><p>在柴油方面,<strong>从2004年实施国2机动车排放标准以来,到2014年国4重型柴油车(机)排放标准的实施,车用柴油标准已经3次落后于机动车排放标准的实施</strong>,国2、国3排放标准是在国1柴油标准条件下实施的,国4排放标准是在国3柴油标准条件下实施的。</p></blockquote></li></ul><hr><p>视频时间:00:49:22</p><p>文字引用:</p><blockquote><p>长期比发达国家低两到三个等级,<br>每提高一个等级,排放可减少10%。</p></blockquote><p>旁白引用:</p><blockquote><p>“我们看上去有好油,我们有国IV以及以上的油,但是这部分油只占百分之三的比例。这就是杭州的痛苦,杭州有钱,想买好油,但是2013年以前就是没有,费了九牛二虎之力,你只能烧国III的油。”</p></blockquote><p>数据来源:中国环境科学院车用燃料及添加实验室</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (二)第一点</li></ul><ol><li>现阶段我国油品质量的确总体低于发达国家,不过并不存在量级的差别,并且未来目标上也将与发达国家看齐。</li><li>京V汽柴油的硫含量相比京Ⅳ油品中的硫含量从50ppm降到了10ppm,主要指标符合同阶段欧V油品标准。</li><li>国Ⅳ以及以上标准的成品油占的比例<strong>不止3%</strong>。</li></ol><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到视频中原表格,参见上一条相关数据。</li></ul><hr><p>视频时间:00:50:18</p><p>旁白引用:</p><blockquote><p>“加油站如果不加这个油气回收装置的话,它蒸发出来的量是这么大,每一升大概要蒸发出1.5克到大气空中来。而且我们平常开车的油箱,也会自己蒸发出来,它蒸发强度大到什么程度?就是在北京,我们光蒸发出来的碳氢化合物,就比汽车尾气排放出来的还要多。”</p></blockquote><p>数据来源:北京市环保局机动车排放管理处</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《汽油车加油过程油气蒸发损失试验方法》<a href=\"http://t.cn/RwrdsnO\" target=\"_blank\" rel=\"noopener\">(何仁等,2010)</a><blockquote><p>所有的汽油车都要经过加油后才使用的,如果不采取任何措施,美国环境保护署估计的排放因子为1.32g/L。现在中国使用的一般汽油车只安装活性炭罐,根据文献估计排放因子可减小为1.02g/L。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwrgzEq\" target=\"_blank\" rel=\"noopener\">第七章 废气控制</a><blockquote><p>1.油箱、化油器——挥发之汽油气体,主要为HC,<strong>占HC废气的20%</strong>。<br>2.曲轴箱——经活塞与汽缸壁漏入曲轴箱之吹漏气,主要为HC,占HC废弃的25%。<br>3.排气管——燃烧后排出之气体,主要为HC、CO、NOx,<strong>占HC废气的55%</strong>。</p></blockquote></li><li><a href=\"http://t.cn/Rwrg2hj\" target=\"_blank\" rel=\"noopener\">废气排放检测的必要性</a><blockquote><p>碳氢化合物总称为烃类,是发动机未燃尽的燃料或分解产生的气体。汽车排放污染物中的HC20%~25%来自曲轴箱窜气,<strong>20%来自化油器与燃油箱的蒸发;其他55%由排气管排出</strong>。</p></blockquote></li><li>环境保护部机动车排污监控中心:<a href=\"http://t.cn/RwrgtC5\" target=\"_blank\" rel=\"noopener\">实现城市机动车高效减排的科学管理建议[PPT]</a></li><li>综合上述资料,<strong>油箱蒸发出来的碳氢化合物不会比汽车尾气所排放出来的碳氢化合物更多</strong>。</li></ul><hr><p>视频时间:00:51:09</p><p>旁白引用:</p><blockquote><p>“这个小标委(石油燃料和润滑剂分技术委员会)呢,就是绝大部分,大部分吧,还是这个石化行业的人,大概是百分之六十七左右;大标委(全国石油产品和润滑剂标准化技术委员会)更多,大标委几乎,我记得好像百分之九十以上都是石化行业的人;标委会的秘书处也在石化行业那儿。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>法治周末:<a href=\"http://t.cn/zRiESM8\" target=\"_blank\" rel=\"noopener\">谁在控制油品标准制定</a><blockquote><p>而油品的标准制定,也掌握在全国石油产品和润滑剂标准化技术委员会(以下简称“石化标委会”)手中。根据记者了解,在石化标委会中,<strong>90.9%的委员来自石化系统</strong>、2.3%的委员来自汽车产业系统、4.5%的委员来自环保系统;主任委员来自中石化,且<strong>秘书处设在中石化</strong>。</p></blockquote></li><li><a href=\"http://t.cn/RwrsHbv\" target=\"_blank\" rel=\"noopener\">关于成立全国石油产品和润滑剂标准化技术委员会石油燃料和润滑剂分技术委员会等六个分技术委员会的批复</a><br>根据<a href=\"http://www.cptcstd.org/Attach/20080904113621375.doc\" target=\"_blank\" rel=\"noopener\">附件1:全国石油产品和润滑剂标准化技术委员会第一届石油燃料和润滑剂分技术委员会(SAC/TC280/SC1)委员名单[doc]</a>,委员共57人,其中41人属于石化行业(中国石油化工股份有限公司21人、中国石油天然气股份有限公司16人、中国海洋石油总公司2人、河南省洛阳石油化工工程公司1人、中海油气开发利用公司1人),<strong>占比71.93%</strong></li></ul><hr><p>视频时间:00:57:56</p><p>旁白引用:</p><blockquote><p>“在中国,除了我们的油品标准低之外,我们还有另外一个,在我出发之前远远没有想到的问题就是我们有一半的油是完全失控。全球十大港口现在有七个在中国;远洋货轮带来的污染之大,在靠近海岸线四百米的地方,它排放的污染相当于五十万辆大卡车。所以在深圳,百分之六十的二氧化硫是来自轮船的。”</p></blockquote><p>数据来源:国家环保部机动车排污监控中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>深圳商报:<a href=\"http://t.cn/RhgL9ki\" target=\"_blank\" rel=\"noopener\">深圳率先开展港口船舶污染实质性治理工作 每年2亿元补贴“绿色航运”</a><blockquote><p>目前,大多航船采用含硫率3%左右的重油作为燃料,每年排放二氧化硫约1.6万吨,<strong>占我市总排放量的65.8%左右</strong>;氮氧化物1.9万吨,占总排放量的14.1%左右;排放细颗粒物(PM2.5)1600吨,占总排放量的5.8%左右。研究表明,一艘排水量10万吨的大型集装箱船一次停靠过程中的PM2.5排放量就相当于5.95万辆国Ⅳ柴油货车一天的排放量。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (二)第三点<blockquote><p>根据微信公众号“新兴产业观察者”的数据,轮船用的<strong>重柴油或者渣油占燃料油的比例接近一半(47%)</strong>,而燃料油占总油量消耗量的比例是6.6%,也就是说,<strong>船舶运输占石油消耗的比例是3%</strong>。</p></blockquote></li><li>“四百米”和“五十万辆大卡车”可能是错误引用了此文 GAS2:<a href=\"http://t.cn/SqDZS8\" target=\"_blank\" rel=\"noopener\">One Container Ship Pollutes As Much As 50 Million Cars</a><blockquote><p>But newly released data from Europe suggests that <strong>a single container ship may cause as much pollution as 50 million cars</strong> and release as much as 5,000 tons of sulphur oxide into the air annually, contributing heavily to global warming.<br>Even if the sails were just used to get the ships away from port, that would represent a major reduction in pollution since 70% of ship emissions are <strong>within 400km of land</strong> and fuel consumption.</p></blockquote></li></ul><hr><p>视频时间:00:58:38</p><p>旁白引用:</p><blockquote><p>“也有可能你会生活在机场的边上,飞机一起飞,一吨煤油没有了,一落地,一吨煤油没有了,所以(广州)白云机场(每天)排污量相当于六十万辆出租车的同时排放。”</p></blockquote><p>数据来源:广东省气象局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>广州日报:<a href=\"http://t.cn/RwrBTPf\" target=\"_blank\" rel=\"noopener\">飞机起降成空气污染物新来源</a><blockquote><p>一些以前被忽视的排放源,其排放量也开始逐年增加,吴兑认为最为明显的就是飞机等航空器起降以及巡航带来的空气污染物排放,“比如一架空客A380,<strong>起降一次耗费的燃油量将近2吨</strong>,相当于数千辆汽车的排放。以一天白云机场起降近千架航班计算,<strong>排放量相当于几十万辆汽车</strong>。”</p></blockquote></li></ul><hr><p>视频时间:01:05:05</p><p>图表引用:(粗钢产量与GDP折线图)</p><p>数据来源:国际钢铁协会、中国国家统计局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>国际钢铁协会:<a href=\"http://t.cn/RwrBOwz\" target=\"_blank\" rel=\"noopener\">1980-2013年全球粗钢产量</a></li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 3-1 国内生产总值</li></ul><hr><p>视频时间:01:12:01</p><p>旁白引用:</p><blockquote><p>“在中国,每天要消失八十个自然村,我们房屋的平均寿命只有30年,在中国一共有200多家地级市,其中184家要建成国际大都市。现在中国有13亿人, 但如果我们把目前城市规划当中这些人口都加起来的话它是34个亿。”</p></blockquote><p>数据来源:国家城市和小城镇改革发展中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>都市快报:<a href=\"http://t.cn/zl1OpUO\" target=\"_blank\" rel=\"noopener\">过去10年 全国每天消失80个自然村</a><blockquote><p>中国文联副主席、国务院参事冯骥才10月20日透露,相关部门最新统计数字显示,我国的自然村十年前有360万个,现在则只剩270万个,“比较妥当的说法是<strong>每一天消失80至100个村落</strong>”。</p></blockquote></li><li>华商网:<a href=\"http://t.cn/RwrBIBB\" target=\"_blank\" rel=\"noopener\">中国住房平均寿命仅30年(图)</a><blockquote><p>其实,“国内新住宅平均寿命仅30年”早在2006年举行“全国第七届建筑物改造学术研讨会”上就提出来了。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>人民网:<a href=\"http://t.cn/RwrBx2d\" target=\"_blank\" rel=\"noopener\">183个城市争建国际大都市 全不顾实事求是</a><blockquote><p>在2005年7月21日召开的城市总体规划修编工作座谈会上,建设部部长汪光焘批评有的城市在城市定位上不切实际,盲目追求高速度和高标准。他表示,全国竟有<strong>183个城市</strong>提出建立“现代化国际大都市”的目标,30多个城市提出要建中心商务区,这实际上是透支和浪费日渐缺乏的发展资源。</p></blockquote></li><li>新华网:<a href=\"http://t.cn/8sjnKhq\" target=\"_blank\" rel=\"noopener\">90%地级市争建新城新区 如何避免成“死城”?</a><blockquote><p>“有研究数据显示,全国<strong>新城新区规划人口达到34亿</strong>。这意味着现在中国一倍的人口也装得下。”国家发展改革委城市和小城镇改革发展研究中心副主任乔润令说,由于规划脱离实际,各个新区为招商引资互相压价,结果往往“两败俱伤”。</p></blockquote></li></ul><hr><p>视频时间:01:13:01</p><p>旁白引用:</p><blockquote><p>“国务院发展研究中心做的测算是:十五年之后,我们的耗煤量会达到六十亿吨,我们的用车量会达到四亿辆。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>中国财政部、国务院发展研究中心和世界银行:《2030年的中国:建设现代、和谐、有创造力的高收入社会》<a href=\"http://t.cn/zObF7Rf\" target=\"_blank\" rel=\"noopener\">英文版</a><a href=\"http://t.cn/zOVMypk\" target=\"_blank\" rel=\"noopener\">中文版</a><blockquote><p>“ In the $10 price scenario, China uses 492 million tons less coal compared with business as usual in 2030 (a difference of 15 percent). Under the $20 price scenario, coal use drops by 770 million tons compared with business as usual (a difference of 23 percent). ”<br>(原文未列出Business as usual时的使用量,但根据上文推算,<strong>该值在3280~3350 million tons左右,即三十亿吨</strong>。)</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>中国经济新闻网:<a href=\"http://t.cn/8sn0KOy\" target=\"_blank\" rel=\"noopener\">王青:中国汽车市场将面临需求和制度的双重转变</a><blockquote><p>经过四种预测方法对中国汽车中长期增长情况的预测,到2023年,中国汽车千人拥有量和总保有量将分别达到235辆和<strong>3.4亿辆</strong>,届时新车产销规模约为3800万辆。<br>(作者单位:国务院发展研究中心市场经济研究所)</p></blockquote></li><li>《中国乘用车与商用车保有量预测》<a href=\"http://t.cn/Rwr1rI6\" target=\"_blank\" rel=\"noopener\">(郝瀚等,2011)</a><blockquote><p>中国汽车保有量在2020年和2030年将分别达到2.5—2.9亿辆和<strong>4—5.2亿辆</strong>,其中乘用车保有量增长速度相对较快,且存在一定不确定性,在2020年和2030年将分别达到2.2—2.6亿辆和3.5—4.8亿辆。商用车保有量在2020年和2030年将分别达到0.33和0.47亿辆。</p></blockquote></li><li><a href=\"http://t.cn/Rwr1kRP\" target=\"_blank\" rel=\"noopener\">《中国发展低碳汽车燃料政策建议报告》</a><blockquote><p>据国际能源署 (IEA)预计,中国到2020年汽车保有量将达2.3亿辆,2030年将<strong>超过4亿辆</strong>。(IEA, 交通能源与 CO2,2009)<br>(但在IEA:<a href=\"http://t.cn/Rwr1sCr\" target=\"_blank\" rel=\"noopener\">Transport, Energy and CO2</a>中并未找到相关表述。)</p></blockquote></li><li>清华大学中国车用能源研究中心:<a href=\"http://t.cn/RwreeZ7\" target=\"_blank\" rel=\"noopener\">《中国车用能源展望2012》</a></li></ul><hr><p>视频时间:01:15:43</p><p>文字引用:</p><blockquote><p>要拥有APEC蓝比2013减排<br>47%二氧化硫<br>52%氮氧化物<br>44%PM2.5</p></blockquote><p>数据来源:中科院</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到中科院相关支持数据</li><li>北京市环境保护局:<a href=\"http://t.cn/RwrCWdm\" target=\"_blank\" rel=\"noopener\">APEC蓝的启示</a><blockquote><p>APEC会议空气质量保障措施效果初步评估结果:<br>二氧化硫(SO2)、氮氧化物(NOx)、可吸入颗粒物(PM10)、细颗粒物(PM2.5)、挥发性有机物(VOCs)(以下均用简称)等减排比例分别达到54%、41%、68%、63%和35%左右。<br>PM2.5、PM10、SO2、二氧化氮(NO2)的浓度……比去年同期分别下降55%、44%、57%和31%</p></blockquote></li><li>北京市环境保护局:<a href=\"http://t.cn/RwrC81u\" target=\"_blank\" rel=\"noopener\">减排才是硬道理APEC空气质量保障措施效果评估结果出炉</a><blockquote><p>APEC会议空气质量保障措施效果评估结果:<br>全市分别削减二氧化硫(SO2)、氮氧化物(NOx)、可吸入颗粒物(PM10)、细颗粒物(PM2.5)和挥发性有机物(VOCs)排放约39.2%、49.6%、66.6%、61.6%和33.6%,平均削减50%左右。</p></blockquote></li></ul><hr><p>视频时间:01:16:34</p><p>旁白引用:</p><blockquote><p>“中国的节能环保产业现在可能是一年的产值37000多亿,吸纳就业人口可能是3900多万人,这也是一个产业。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“37000多亿”和“3900多万人”的支持数据。</li><li>国务院:<a href=\"http://t.cn/Rwr9VSK\" target=\"_blank\" rel=\"noopener\">“十二五”节能环保产业发展规划</a><blockquote><p>据测算,2010年,我国节能环保产业<strong>总产值达2万亿元,从业人数2800万人</strong>。<br>到2015年, 节能环保产业<strong>总产值达到4.5万亿元</strong>。</p></blockquote></li></ul><hr><p>视频时间:01:17:39</p><p>旁白引用:</p><blockquote><p>“但是从1970年以来,洛杉矶的车辆还在上升,上升了这么多,三倍;但它们排放呢,降低了百分之七十五。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/RwQGtLm\" target=\"_blank\" rel=\"noopener\">Jun</a> 修订4 原答10(回答已删除)</li></ul><ol><li>VMT的数据多了个0,扩大了10倍。</li><li>汽车和卡车用量(VMT)和排放总量(Cumulative CA vehicle emissions for NOx and HCs)的数据来源:<a href=\"http://t.cn/Rw8hGF5\" target=\"_blank\" rel=\"noopener\">Key Events in the History of Air Quality in California</a>;其中VMT数据吻合但有缺失,但Emissions和左侧Pollution数据不吻合</li></ol><table><thead><tr><th align=\"center\">年份</th><th align=\"center\">VMT(billion)</th><th align=\"center\">Emissions(million tons/year)</th></tr></thead><tbody><tr><td align=\"center\">1970</td><td align=\"center\">110</td><td align=\"center\">1.6</td></tr><tr><td align=\"center\">1980</td><td align=\"center\">155</td><td align=\"center\">1.6</td></tr><tr><td align=\"center\">1990</td><td align=\"center\">242</td><td align=\"center\">1.4</td></tr><tr><td align=\"center\">1995</td><td align=\"center\">271</td><td align=\"center\">1.1</td></tr><tr><td align=\"center\">2000</td><td align=\"center\">280</td><td align=\"center\">1.2</td></tr></tbody></table><hr><p>视频时间:01:18:35</p><p>旁白引用:</p><blockquote><p>“(洛杉矶)一千七百多万人,差不多一千三百多万辆车,都快人手一辆了。整个加州,一天之内,8.5亿的行程,所有的燃料加起来都够来回月球一千六百趟。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>California Department of Transportation: <a href=\"http://t.cn/Rwr6aAK\" target=\"_blank\" rel=\"noopener\">Historical Monthly Vehicle Miles of Travel 1972 -2014 (In Billions)</a> (每天8.5亿公里,折合至英里/每月为15.845billion,基本吻合)</li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/Rwr6PC4\" target=\"_blank\" rel=\"noopener\">City-data</a><blockquote><p>洛杉矶市2013年人口:3,884,307</p></blockquote></li><li>Wikipedia: <a href=\"http://t.cn/zO60Akd\" target=\"_blank\" rel=\"noopener\">Los Angeles County</a><blockquote><p>洛杉矶郡2013年人口:10,017,068</p></blockquote></li><li>Los Angeles Almanac: <a href=\"http://t.cn/Rwr6GzR\" target=\"_blank\" rel=\"noopener\">Vehicle Registrations</a><blockquote><p>洛杉矶郡2008年登记各类车辆共计:7,499,722</p></blockquote></li><li>STREETS BLOG LA: <a href=\"http://t.cn/8kJoEIK\" target=\"_blank\" rel=\"noopener\">Density, Car Ownership, and What It Means for the Future of Los Angeles</a><br>(此处Los Angeles指The greater Los Angeles Region)</li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“一千六百趟”的相关数据支持</li></ul><hr><p>视频时间:01:19:11</p><p>文字引用:</p><blockquote><p>加州柴油车1061745辆<br>每辆车平均耗油量35.2升/车/天</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“1061745辆”和“35.2升/车/天”的支持数据</li><li>Diesel Techonology Forum: <a href=\"http://t.cn/RwrhQG5\" target=\"_blank\" rel=\"noopener\">Total Diesel Passenger Vehicles</a><blockquote><p>California 2013 Diesel Total 609,212</p></blockquote></li><li>Board of Equalization - State of California: <a href=\"http://t.cn/RwrhktQ\" target=\"_blank\" rel=\"noopener\">TAXABLE DIESEL GALLONS 10 YEAR REPORT (NET OF REFUNDS)</a></li></ul><hr><p>视频时间:01:19:24<br>旁白引用:</p><blockquote><p>“根据加州空气资源部的估计,南加州空气中71%的致癌物质都由柴油车产生。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><strong>并非加州空气资源部的估计</strong></li><li>实际来源为南加州空气质量管理局(SCAQMD):<a href=\"http://t.cn/RwBkXap\" target=\"_blank\" rel=\"noopener\">Major Air Toxics Study Finds Vehicles Dominate Cancer Risk November 5, 1999</a><blockquote><p>AQMD found that that <strong>diesel soot accounted for 71% of the cancer risk</strong>, 1,3 butadiene (a byproduct of incomplete combustion in engines) 8% of the risk, benzene (mostly from motor vehicles) 7%, carbonyls (including formaldehyde and acetaldehydes from both mobile and stationary sources) 3%, and other pollutants (primarily from stationary sources) 11%.</p></blockquote></li></ul><hr><p>视频时间:01:21:03</p><p>文字引用:</p><blockquote><p>全年发出重卡柴油车检查传票29913起<br>最终结案3281起<br>罚款1336880美元</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>加州空气资源部(California Air Resource Board): <a href=\"http://t.cn/RwBDrSx\" target=\"_blank\" rel=\"noopener\">2013 ANNUAL ENFORCEMENT REPORT SUMMARY</a><blockquote><p>Heavy-duty Diesel Field Inspections and Citations: conducted 29,913 inspections, issued 3,281 citations, and collected $1,336,880 in penalties.<br>(重型柴油车路检及传票:进行<strong>29913次路检</strong>,发出<strong>3281张传票</strong>,罚款1336880美元。)</p></blockquote></li></ul><hr><p>视频时间:01:23:11</p><p>文字引用:</p><blockquote><p>洁净的大气带来的效益远高于治理大气污染的成本</p></blockquote><p>数据来源:美国南加州空气质量管理局</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><strong>未在南加州空气质量管理局(SCAQMD)网站找到相关数据</strong></li><li>实际来源为美国环境保护局(United States Environment Protection Agency,EPA):<br><a href=\"http://t.cn/RwBDtXi\" target=\"_blank\" rel=\"noopener\">Benefits and Costs of the Clean Air Act - Second Prospective Study - 1990 to 2020</a><br>(且图表数值轴应当以Billion为单位)</li></ul><hr><p>视频时间:01:23:29</p><p>旁白引用:</p><blockquote><p>“我们来看看伦敦的曲线,很多人都跟我将说伦敦要等了四十年、五十年才把污染治好我们也得等这么久,但真的是这样吗?看看这儿(1960~1970段),从他们开始治理污染的前十年,他们就把污染物降低了百分之八十,这是一个极大的改善。”</p></blockquote><p>数据来源:Ren 2000, Hutchingson et al. 2004 and Okubo 2013</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>本处是全片唯一一次给出了论文(或文章)出处的地方,这三篇论文(或文章)分别是:</li></ul><ol><li>“Japanese Approaches to Environmental Management: Structural and Institutional Features.”<a href=\"http://t.cn/RwRZKOg\" target=\"_blank\" rel=\"noopener\">(Ren, 2000)</a></li><li>“Air Quality in London and Tokyo: A Comparison.”<a href=\"http://t.cn/RwRZOlU\" target=\"_blank\" rel=\"noopener\">(Hutchingson et al., 2004)</a></li><li>“Mt. Fuji Lovers Get Bonus with More Days to See It from Tokyo.” <a href=\"http://ajw.asahi.com/article/behind_news/social_affairs/AJ201302250003\" target=\"_blank\" rel=\"noopener\">Okubo, 2013</a></li></ol><ul><li>单从标题就可以看出,第一篇和第三篇论文(文章)都是关于日本东京的,而和伦敦完全没有关系,这个图表其实来自于第二篇论文。</li><li>之所以会引用第一篇和第三篇论文(文章),其实是转引自世界银行和中国国务院发展研究中心:<a href=\"http://t.cn/RPLVQev\" target=\"_blank\" rel=\"noopener\">Urban China : Toward Efficient, Inclusive, and Sustainable Urbanization</a><br>在这本书的第453页Box7.1中,使用了伦敦和东京治理污染的对比图,其中伦敦的图是直接从Hutchinson(2004)中拿来用的,而东京的图则加入了新的数据,因为Hutchinson(2004)关于东京的数据只到1987年左右就结束了;在Box的最后有这样一段文字:Box sources: Ren 2000; Bell, Davis, and Fletcher 2004; Hutchinson and others 2004; Okubo 2013。可以猜测纪录片引用的就是这个图表,但是要么是忘了检查引用是否正确,要么是原本打算放上伦敦与东京的对比数据但后来又把东京的数据撤掉了。</li></ul><hr><p>视频时间:01:23:48</p><p>旁白引用:</p><blockquote><p>“伦敦大烟雾事件发生的时候,英国的能源结构中将近百分之九十是煤炭;1953年,颗粒物的平均浓度超过欧盟标准的十倍。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到直接数据</li><li><a href=\"http://t.cn/RwBmx0c\" target=\"_blank\" rel=\"noopener\">60th Anniversary: Digest of United Kingdom Energy Statistics</a><br>1948年,煤炭<strong>占主要能源的90%</strong>,<strong>占最终能源使用则为77%</strong>。</li><li>“Reassessment of the Lethal London Fog of 1952: Novel Indicators of Acute and Chronic Consequences of Acute Exposure to Air Pollution”<a href=\"http://t.cn/zY1RFwo\" target=\"_blank\" rel=\"noopener\">(Bell and Davis, 2001)</a> (注:TSM应当为TSP)</li></ul><hr><p>视频时间:01:26:46</p><p>旁白引用:</p><blockquote><p>“现在世界平均的天然气占能源结构的比例是很高的,大概百分之二十四,因为天然气是相对清洁的能源。那中国是多少呢?我们只有百分之五。为什么我们这么低?从小到大我听到的标准答案是这样的:因为我们是一个煤多、贫油、少气的国家,这是真的吗?这是真的。但这完整吗?”</p></blockquote><p>数据来源:国务院发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《BP世界能源统计年鉴》天然气在全球一次能源消费中所占份额<br><a href=\"http://t.cn/RwBnqWH\" target=\"_blank\" rel=\"noopener\">2013年</a>: 23.9%<br><a href=\"http://t.cn/R7qKXNQ\" target=\"_blank\" rel=\"noopener\">2014年</a>: 23.7%</li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 9-2 能源消费总量及构成<blockquote><p>天然气占能源消费总量的比重%:2011年5.0;2012年5.2;2013年5.8;</p></blockquote></li><li><a href=\"http://t.cn/RwBm5m4\" target=\"_blank\" rel=\"noopener\">中国煤炭的清洁高效利用——中国煤炭资源与消费</a></li></ul><hr><p>视频时间:01:27:22</p><p>旁白引用:</p><blockquote><p>“我们的天然气探明多少呢?百分之二十二。这意味着我们还有大量的天然气埋藏在中国大地之下黑暗不知名的深处。那我们的石油探明率是多少呢?百分之三十八。在每年探明的这些储量当中,比如说天然气,探明九千亿立方米,但我们开采了多少呢?一千亿立方米。”</p></blockquote><p>数据来源:国土部矿产资源储量评审中心</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>国土资源部:<a href=\"http://t.cn/RZEdWAz\" target=\"_blank\" rel=\"noopener\">《2014中国矿产资源报告》</a><blockquote><p>全国油气资源评价结果显示,截至2013年底,石油地质资源量1037亿吨,<strong>探明率33.8%</strong>,处于勘探中期。天然气地质资源量62万亿立方米,<strong>探明率18.4%</strong>,处于勘探早期。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (三)第四点</li></ul><ol><li>从国土资源部网站可以查到的数据是,2012年我国天然气<strong>新增探明地质储量是9612.2亿立方米</strong>。</li><li>国家发改委的数据是,2012年我国<strong>天然气产量是1077亿立方米</strong>。</li><li>到2013年底,我国天然气可采常规剩余储量是3.3万亿立方米,当年<strong>天然气产量是1171亿立方米</strong>,储采比是28。</li><li>我国油气勘探总体进入低品位资源勘探阶段,勘探成本不断上升。</li><li>虽然中国的天然气生产实现全球第二大增量(9.5%,99亿立方米),但仍有巨大的缺口需要通过增加进口予以解决。</li></ol><hr><p>视频时间:01:27:58</p><p>旁白引用:</p><blockquote><p>“他说拿世界上最大的天然气生产国美国来说,它们有六千三百家天然气石油公司,我们有几家呢?3家;其中百分之七十集中在一家手里——中石油。美国有一百六十家天然气管道公司,我们(主体管道)有几家呢?3家,其中百分之七十集中在一家手里——中石油。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>NaturalGas.org : <a href=\"http://t.cn/RwYIzWY\" target=\"_blank\" rel=\"noopener\">Industry and Market Structure</a><blockquote><p>Producers - There are <strong>over 6,300 producers of natural gas</strong> in the United States.<br>Pipelines - There are <strong>about 160 pipeline companies</strong> in the United States, operating over 300,000 miles of pipe.</p></blockquote></li><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (三)第四点<blockquote><p>目前,中石油已经形成横跨东西、纵贯南北,连接四大气区和主要消费市场,覆盖全国的天然气管网体系,<strong>已投运的油气管网占国内长输管道的80%左右</strong>,其中天然气管网里程超过4.7万公里,覆盖30个省、市、自治区、特别行政区,惠及5亿人口,<strong>天然气销量占国内市场份额的70%左右</strong>。</p></blockquote></li><li><a href=\"http://t.cn/RwRHs10\" target=\"_blank\" rel=\"noopener\">中国石油天然气集团公司2013年度报告</a><blockquote><p>(第23页)公司国内天然气产量占全国<strong>比重75%</strong><br>(第27页)天然气管道总里程占全国<strong>比重80%</strong></p></blockquote></li></ul><hr><p>视频时间:01:32:38</p><p>旁白引用:</p><blockquote><p>“中国有百分之一的环境争议事件进入了司法程序。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据支持</li></ul><hr><p>视频时间:01:35:58</p><p>旁白引用:</p><blockquote><p>“后来我知道,在北京餐饮带来的PM2.5的污染占到至少百分之六。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>经济参考报:<a href=\"http://t.cn/zR4fqwx\" target=\"_blank\" rel=\"noopener\">餐饮油烟成城市PM2.5罪魁 治理遇诸多薄弱环节</a><blockquote><p>王跃思告诉《经济参考报》记者,总的来说,餐饮油烟<strong>在北京市大气中PM2.5中的比例约为13%,最高达15%</strong>,在<strong>京津冀地区所占的比例约为6%</strong>。</p></blockquote></li></ul><hr>"},{"title":"数据深扒:纪录片《穹顶之下》","path":"2015/08/21/underthedome/","permalink":"http://sx349.github.io/2015/08/21/underthedome/","content":"<hr><p>本文片源:<a href=\"http://t.cn/Rwjlj0M\" target=\"_blank\" rel=\"noopener\">Youtube</a><br>参考片源:国内片源全部已跪,请科学上网。<br>初稿完成:2015-03-12 @ 有道云笔记<br>最近修改:2015-08-21 @ 盗泉<br>欢迎转载,转载请注明原作者信息。</p><hr><p>视频时间:00:03:19</p><p>文字引用:</p><blockquote><p>山西全境<br>88.4%的河流断面遭到污染<br>62%的断面失去功能</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中华人民共和国环境保护部:<a href=\"http://t.cn/Rw3tiDw\" target=\"_blank\" rel=\"noopener\">山西从污染困境突围</a><blockquote><p>全省26条主要河流中,有<strong>88.4%的断面受到不同程度的污染,有62%的断面失去了使用功能</strong>,主要城市集中式生活饮用水源地水质有17.2%不达标。</p></blockquote></li></ul><hr><p>视频时间:00:04:23</p><p>文字引用:</p><blockquote><p>(北京市2014年)污染天数175天</p></blockquote><p>数据来源:北京市环保局</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>北京市环保局:<a href=\"http://t.cn/RwYnOTH\" target=\"_blank\" rel=\"noopener\">2014年北京市污染减排力度大空气质量持续改善</a><blockquote><p>2014年PM2.5一级优天数达到93天,较2013年的71天大幅度增加22天,增幅31%;PM2.5导致的重污染天数也明显减少,从2013年的58天下降为45天,减少13天,降幅22%。</p></blockquote></li><li>国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a><br>根据全国城市空气质量日报的数据进行统计,北京市2014年的天气质量分别是:优28天 良139天 <strong>轻度污染85天 中度污染65天 重度污染30天 严重污染15天</strong> (缺3天数据)</li></ul><hr><p>视频时间:00:04:51</p><p>文字引用:</p><blockquote><p>天津2014年空气污染天数197天<br>沈阳2014年空气污染天数152天<br>兰州2014年空气污染天数112天<br>石家庄2014年空气污染天数264天</p></blockquote><p>数据来源:国家环境保护部</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a><br>数据基本吻合</li></ul><hr><p>视频时间:00:07:05</p><p>图表引用:<br><img src=\"/images/UTD/000705.png\" alt=\"000705\"></p><p>旁白引用:</p><blockquote><p>“这是世界卫生组织的标准(25),这个是美国的(37.5),这个是欧洲(50),这个是中国(75);那么我们在当天呼吸进去的是多少呢?几乎是中国标准值的五倍。”</p></blockquote><p>数据来源:《环境空气质量标准》GB 3095-2012</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li><a href=\"http://t.cn/Rw3YVjf\" target=\"_blank\" rel=\"noopener\">《环境空气质量标准》GB 3095-2012</a><br><img src=\"/images/UTD/000705-1.png\" alt=\"000705-1\"></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>其它标准</li></ul><table><thead><tr><th align=\"center\">组织或国家</th><th align=\"center\">标准</th><th align=\"center\">数值</th></tr></thead><tbody><tr><td align=\"center\">世界卫生组织</td><td align=\"center\"><a href=\"http://t.cn/hxqwX\" target=\"_blank\" rel=\"noopener\">《空气质量准则》</a></td><td align=\"center\">25</td></tr><tr><td align=\"center\">欧盟</td><td align=\"center\"><a href=\"http://t.cn/S7SVpf\" target=\"_blank\" rel=\"noopener\">Air Quality Standard</a></td><td align=\"center\"><strong>25(年均)</strong></td></tr><tr><td align=\"center\">美国</td><td align=\"center\"><a href=\"http://t.cn/Rw3YDMm\" target=\"_blank\" rel=\"noopener\">National Ambient Air Quality Standard</a></td><td align=\"center\"><strong>35</strong></td></tr></tbody></table><ul><li>欧盟的<strong>50是PM10的24小时平均浓度限值</strong>,PM2.5并未规定24小时平均浓度限值</li><li>美国的<strong>37.5</strong>可能是误用了<strong>世界卫生组织的过渡时期目标-3</strong></li><li>305.91/75 ≈ 4.0788,因此检测数据是中国标准值的<strong>4倍左右</strong>,并未达到5倍</li></ul><hr><p>视频时间:00:07:34</p><p>文字引用:</p><blockquote><p>苯并[a]芘benzo[a]pyrene 176<br>超标14倍</p></blockquote><p>旁白引用:</p><blockquote><p>“其中,世界上最强的致癌物(之一)苯并[a]芘,它是国家标准值的多少倍呢?14倍。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>苯并[a]芘的<strong>相对分子量为252</strong>,因此176不是指相对分子质量</li><li>根据<a href=\"http://t.cn/Rw3YVjf\" target=\"_blank\" rel=\"noopener\">《环境空气质量标准》GB 3095-2012</a>,苯并[a]芘24小时<strong>平均浓度限值为0.0025 μg/m<sup>3</sup></strong>;因此176也不是检测数据</li></ul><hr><p>视频时间:00:08:20</p><p>旁白引用:</p><blockquote><p>“所以如果它打开的话,两克黑碳能有多大?能有整个篮球场那么大;所以它可以吸附很多的致癌物和重金属。”<br>数据来源:中科院大气物理研究所</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>“Characterisation and evaluation of reference materials for black carbon analysis using elemental composition, colour, BET surface area and 13C NMR spectroscopy” <a href=\"http://t.cn/Rw3TmSx\" target=\"_blank\" rel=\"noopener\">(Hammes et al., 2008)</a><br><img src=\"/images/UTD/000820.png\" alt=\"000820\"><br>标准篮球场面积为420m<sup>2</sup>,而上表中列出的富黑碳材料比表面积最高值<strong>仅为73.9m<sup>2</sup>/g</strong>。</li><li>“Surface Area and Properties of Carbon Black”<a href=\"http://t.cn/Rw3Tx7x\" target=\"_blank\" rel=\"noopener\">(Smith et al., 1941)</a><br><img src=\"/images/UTD/000820-1.png\" alt=\"000820-1\"><br>另外一种称为炭黑(Carbon Black)的物质,其比表面积<strong>最高可以达到990m<sup>2</sup>/g</strong>,尚不清楚该物质是否等同于黑碳。</li></ul><hr><p>视频时间:00:08:33</p><p>图表引用:<br><img src=\"/images/UTD/000833.png\" alt=\"000833\"></p><p>旁白引用:</p><blockquote><p>“在中国这样的黑碳有多少呢?这是2009年NASA(美国航空航天局)做的一个测算,那个紫得发白发亮的地方是中国,(黑碳)它像幽灵一样在我们的上空飘荡。”</p></blockquote><p>数据来源:美国国家航空航天局(NASA)</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/RwQGtLm\" target=\"_blank\" rel=\"noopener\">Jun</a>原答4(回答已删除)<br>提供来源 <a href=\"http://t.cn/RwQoCGR\" target=\"_blank\" rel=\"noopener\">Scientific Visualization Studio: Atmospheric Black Carbon Density</a></li><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第一处<blockquote><p>《穹顶之下》的这幅图取自2009年8月1日到11月19日的全球黑炭或烟尘的运动轨迹。</p></blockquote></li></ul><hr><p>视频时间:00:12:31</p><p>图表引用:<br><img src=\"/images/UTD/001231.png\" alt=\"001231\"></p><p>旁白引用:</p><blockquote><p>“这个是中科院给我们提供的测算,这个复杂的图表显示的是当PM2.5的值升高的时候,人群的死亡率是随之上升的。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/Rw8OlAJ\" target=\"_blank\" rel=\"noopener\">宋宁世</a></li></ul><ol><li>论文来源 “Time-series analysis of mortality effects from airborne particulate matter size fractions in Beijing”<a href=\"http://t.cn/Rw8Ea1x\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a></li><li>红色数据点没有进行实质的拟合。</li><li>论文中的原图表中标注了PM2.5的数值,但2006~2009年北京市年均PM2.5的数值是减少的,与纪录片的结论相悖。</li><li>环境卫生学中评定污染物健康危害更常用的是Dose response curve,根据论文中提供的图表,PM2.5值超过一定数值后,相对健康风险的数值反而下降了。</li><li>根据论文提供的结论:在不同国家所进行的环境卫生学调查的结果显示出了不一样的现象,在北京,PM2.5对人的健康带来的负面影响反而比发达国家的结果要小。</li><li>综上所述,纪录片制作团队或者并没有去查证过自己数据的来源及可靠性,或者没有具备验证数据可靠性的知识水平,又或者刻意对自己的数据进行了处理/刻意隐藏不利于自己的数据</li></ol><hr><p>视频时间:00:12:51</p><p>旁白引用:</p><blockquote><p>“我看到中国的前卫生部部长陈竺的测算结果是,在中国,每年因为大气污染过早死亡的人数是50万人。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>陈竺发表在《柳叶刀》The Lancet上的文章<br>“China tackles the health effects of air pollution”<a href=\"http://t.cn/Rw3Hi6E\" target=\"_blank\" rel=\"noopener\">(Chen et al., 2013)</a><blockquote><p>“Studies by the World Bank, WHO, and the Chinese Academy for Environmental Planning on the effect of air pollution on health concluded that between <strong>350 000 and 500 000 people die prematurely each year as a result of outdoor air pollution in China</strong>.”</p></blockquote></li><li>21世纪经济报道:<a href=\"http://t.cn/8Fv0WTR\" target=\"_blank\" rel=\"noopener\">陈竺:中国每年因空气污染导致早死35万-50万人</a><br>总结:陈竺的文章发表于The Lancet 《柳叶刀》,其结果远低于《柳叶刀》于2012年12月14日发表的《全球疾病负担2010年报告》的评估,后者提出2010年中国因室外PM2.5污染导致120万人早死以及2500万伤残调整寿命年损失;前者认为《报告》是目前国际上对中国室外空气污染健康影响最严重的估计,但可能高估了中国室外空气污染的健康影响;而《报告》的作者之一、中国疾病预防控制中心原副主任杨功焕认为前者所引用的报告主要考虑了PM10及部分PM2.5数据,存在不确定性。</li></ul><hr><p>视频时间:00:13:33</p><p>文字引用:</p><blockquote><p>2013年1月重雾霾期间,27个城市出现急诊的人数爆发增加10%~150%不等,以儿童和老人为主。</p></blockquote><p>数据来源:北京大学环境科学与工程学院环境与经济研究所</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《2013年1月中国大面积雾霾事件直接社会经济损失评估》<a href=\"http://t.cn/Rw3HNng\" target=\"_blank\" rel=\"noopener\">(穆泉,张世秋,2013)</a><blockquote><p>“根据对2013年1月份期间发生过严重雾霾事件的20个省市雾霾造成医院门/急诊相关的公开报道数据的统计,全国14个省市所属的27个城市有出现因雾霾导致医院门/急诊人数激增的报道,门/急症看病人数增加比例<strong>从10%到150%不等,并且就诊人群主要为儿童和老人</strong>。”</p></blockquote></li></ul><hr><p>视频时间:00:14:21</p><p>论文引用:</p><blockquote><p>《Effects of Airborne Metals on Lung Function in Inner Mongolian Schoolchildren》<br>《沙棘油对SO2染毒运动训练大鼠心肌脂质过氧化的影响》<br>《中韩儿童肺活量与影响因素的比较研究》<br>《非沙尘天气大气颗粒物中金属元素浓度与儿童肺功能相关性的初步研究》</p></blockquote><p>旁白引用:</p><blockquote><p>英文:“It has been shown in several studies around the world now that children in cleaner places grow faster than children grown up in a more polluted area. Poor lung function early in life is predictive of lung problems later on in life.”<br>中文字幕:“目前世界各国的研究表明,在更干净的地方长大的孩子的肺生长速度比污染地区长大的孩子要快。年幼时发育不全的肺功能,很可能导致成长后的肺部疾病。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>这4篇论文的相关信息分列如下:</li></ul><ol><li>“Effects of Airborne Metals on Lung Function in Inner Mongolian Schoolchildren”<a href=\"http://t.cn/Rwd9LEh\" target=\"_blank\" rel=\"noopener\">(Madaniyazi et al., 2013)</a><br>(见第四篇论文分析)</li><li>《沙棘油对SO2染毒运动训练大鼠心肌脂质过氧化的影响》<a href=\"http://t.cn/Rw8Y9lb\" target=\"_blank\" rel=\"noopener\">(谢奇,沈一岚,2011)</a><br>本文与雾霾<strong>完全无关</strong>。其结论为:“SO2污染的运动环境下服用沙棘油能改善SO2对大鼠心肌细胞的损伤,提高心肌组织抗自由基氧化的能力。”</li><li>《中韩儿童肺活量与影响因素的比较研究》<a href=\"http://t.cn/Rw8ThL3\" target=\"_blank\" rel=\"noopener\">(金大善等,2011)</a><br>本文结论是:“在潜在影响因素中,<strong>室内是否饲养宠物及家庭年平均收入</strong>对女生肺功能指标的影响有统计学意义(P<0.05),其他指标的影响均无统计学意义;在控制身高、体重、年龄后,中国学生FVC和FEV1与韩国学生相比,男生高0.015L和0.026L,女生高0.128L和0.136L。”</li><li>《非沙尘天气大气颗粒物中金属元素浓度与儿童肺功能相关性的初步研究》<a href=\"http://t.cn/Rw8T5Iu\" target=\"_blank\" rel=\"noopener\">(丽娜·马达尼亚孜等,2011)</a><br>本文同第一篇论文采用的数据部分重合,实验设计和计量方法完全一致,结论相近。其结论为:“学龄儿童日均PEFR值的下降与大气颗粒物中金属元素的含量的升高有关。”“本次研究结果提示<strong>大气颗粒物中的金属元素可能与学龄儿童PEFR之间存在负相关性</strong>,并且滞后效应更为明显。本次研究还发现细颗粒物中的金属元素对学龄儿童肺功能的影响更明显。”其中PEFR(最大呼气流速)是肺功能的一项指标。</li></ol><ul><li>这四篇论文中,第二篇论文完全是因为<strong>恰好与第三篇论文的结尾段共用一页而被误引用</strong></li><li>另外三篇均为北京大学公共卫生学院劳动卫生与环境卫生学系潘小川教授所带研究生撰写的论文,其中第三篇<strong>与空气污染基本无关</strong>,第一篇、第四篇<strong>基本为相同内容</strong>,虽然论文结论和字幕内容相关,但并不完全相符</li><li>此外,在第一篇选择英文论文有<strong>故意误导观众这些研究是来自“世界各国”的嫌疑</strong></li></ul><hr><p>视频时间:00:17:00</p><p>文字引用:</p><blockquote><p>随着颗粒物和空气污染暴露的不断提高,将增加人类患肺癌的风险,颗粒物被列为人类一级致癌物。</p></blockquote><p>旁白引用:</p><blockquote><p>“世卫组织告诉我,颗粒物是被列为人类的一级致癌物。”</p></blockquote><p>数据来源:IARC Scientific Publication NO.161</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/zRVOC5w\" target=\"_blank\" rel=\"noopener\">IARC Scientific Publication NO.161</a> 中<strong>完全没有与纪录片所引文字一致或相近的论述</strong>。</li><li>纪录片所引用的可能来自将颗粒物列为人类一级致癌物的新闻报道<a href=\"http://t.cn/zRV0nGx\" target=\"_blank\" rel=\"noopener\">IARC: Outdoor air pollution a leading environmental cause of cancer deaths</a><blockquote><p>“Particulate matter, a major component of outdoor air pollution, was evaluated separately and was also <strong>classified as carcinogenic to humans (Group 1)</strong>. ”<br>“The IARC evaluation showed <strong>an increasing risk of lung cancer with increasing levels of exposure to particulate matter and air pollution</strong>.”</p></blockquote></li></ul><hr><p>视频时间:00:17:17</p><p>旁白引用:</p><blockquote><p>“我就向NASA(美国国家航天局)提出了一个申请,能不能你们把这十年来华北上空的卫星图片发给我,这是我收到的:2012年,北京的上空;2011年,12月的北京;2010年,华北上空;2009年,华北上空;再看看十年前,2005年的华北。”</p></blockquote><p>数据来源:美国航天航空局(NASA)</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第二处</li></ul><ol><li>提供了一篇参考文章:NASA中文微信公众号:<a href=\"http://t.cn/RwRmg3x\" target=\"_blank\" rel=\"noopener\">《这些年,我们遇到的雾霾》</a></li><li>提供了2001年至2006年全球PM2.5的平均值地图,对于发展中国家(包括中国)来说,是第一张PM2.5卫星估计值图,这也可以解释雾霾不是最近出现的。</li></ol><ul><li>知乎:<a href=\"http://t.cn/RwQg0Bt\" target=\"_blank\" rel=\"noopener\">唐文韬</a> 第2点</li></ul><ol><li>目前,就答主所知,关于雾霾历史记录的研究最可靠的是能见度加上相对湿度的记录;其次Shortwave irradiance也可以作为气溶胶研究的侧面证据,但是不能用于直接研究雾霾。</li><li>视频中使用的这种图更是不能用于确认雾霾的:雾和霾都是根据能见度和相对湿度进行定义的,这里看不出任何相关信息,也分不出是云还是雾霾。<br>(但上述微信号文章中提到“亮白色的是云雾,暗灰色的是遭到污染的大气”,结论相悖)</li></ol><hr><p>视频时间:00:17:56</p><p>图表引用:<br><img src=\"/images/UTD/001756.png\" alt=\"001756\"></p><p>旁白引用:</p><blockquote><p>“这是2004年,十一年前,那时候我们还没有PM2.5的检测,但是我们有PM10。她说根据当时的估算,那个时候,在污染期的话,PM2.5可以达到三百到四百,属于今天的严重污染。”</p></blockquote><p>数据来源:北京大学环境科学系</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据,但根据《北京市大气PM10环境背景值的计算方法探讨》<a href=\"http://t.cn/Rw3exWR\" target=\"_blank\" rel=\"noopener\">(李金香等,2007)</a>,可知当时北京市北部和南部共有5个清洁站点确实已经可以监测PM10数据。</li></ul><hr><p>视频时间:00:20:16</p><p>旁白引用:</p><blockquote><p>“从科研机构的PM2.5站点数据来看,2006年年均PM2.5浓度要高于今天。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>根据中科院论文”Time-series analysis of mortality effects from airborne particulate matter size fractions in Beijing”<a href=\"http://t.cn/Rw8Ea1x\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a><br>2006年年均PM2.5浓度为84μg/m<sup>3</sup></li><li>人民网:<a href=\"http://t.cn/RZtZzMi\" target=\"_blank\" rel=\"noopener\">北京2014年PM2.5年均浓度同比下降4.0%</a><br>2014年年均PM2.5浓度为85.9μg/m<sup>3</sup></li></ul><hr><p>视频时间:00:20:26</p><p>图表引用:<br><img src=\"/images/UTD/002026-1.png\" alt=\"002026-1\"><br><img src=\"/images/UTD/002026-2.png\" alt=\"002026-2\"></p><p>旁白引用:</p><blockquote><p>“此后,主要大气污染物开始整体下降,但是,对能见度能够产生影响的细颗粒物增加了,极端重度污染天数增加了,人们对于污染感受强烈。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>上表:中国环境与发展国际合作委员会2012年年会<a href=\"http://t.cn/Rw1vDS2\" target=\"_blank\" rel=\"noopener\">《区域空气质量综合控制体系研究》研究报告</a><br><img src=\"/images/UTD/002026-3.png\" alt=\"002026-3\"><blockquote><p>“为进一步了解北京PM2.5化学物种的长期变化,采用相同方法重构了1999年7月至2008年7月清华观测点PM2.5可鉴别物种的季均浓度(图 1-13),其中最为突出的特征是PM2.5中二次成分比例持续增长。”</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>下表:未找到统计数据,或可根据国家环境保护部数据中心:<a href=\"http://t.cn/SU4qB9\" target=\"_blank\" rel=\"noopener\">全国城市空气质量日报</a>进行统计而得。</li></ul><hr><p>视频时间:00:20:44</p><p>旁白引用:</p><blockquote><p>“过去30年内,我国的肺癌死亡率上升了465%,虽然吸烟和老龄化仍然是这个数字的主要因素,但细颗粒物明确的致癌风险越来越得到重视。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>华商报:<a href=\"http://t.cn/8kPZDNm\" target=\"_blank\" rel=\"noopener\">中国或成肺癌第一大国 过去30年发病率上升465%</a><br>虽然数据无误,但本文<strong>完全没有提到细颗粒物的致癌风险</strong>,只笼统地在建议中说“对于没有吸烟史的非高危人群,要想科学预防肺癌,除了要做到拒绝被动吸烟、重视并积极治疗肺部慢性疾病、减少室内外空气污染、家庭装修后保持通风换气之外,也应每年进行定期体检筛查”。</li></ul><hr><p>视频时间:00:22:52</p><p>图表引用:<br><img src=\"/images/UTD/002252.png\" alt=\"002252\"></p><p>旁白引用:</p><blockquote><p>“中科院告诉我它的来源是什么:燃煤、燃油、生物质燃烧、工业、养殖、化肥、扬尘。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据来源;</li><li>中科院大气物理研究所王跃思研究员的<a href=\"http://t.cn/Rw14Pf5\" target=\"_blank\" rel=\"noopener\">相关结果</a><br><img src=\"/images/UTD/002252-1.png\" alt=\"002252-1\"></li><li>《科学世界》对王跃思研究员访谈稿中另一<a href=\"http://t.cn/Rw1Guv0\" target=\"_blank\" rel=\"noopener\">相关结果</a><br><img src=\"/images/UTD/002252-2.png\" alt=\"002252-2\"><br><img src=\"/images/UTD/002252-3.png\" alt=\"002252-3\"><br><img src=\"/images/UTD/002252-4.png\" alt=\"002252-4\"></li></ul><hr><p>视频时间:00:23:06</p><p>图表引用:<br><img src=\"/images/UTD/002306.png\" alt=\"002306\"></p><p>旁白引用:</p><blockquote><p>“这就是结论,中国的PM2.5百分之六十来自于燃煤和燃油,也就是化石能源的燃烧。”</p></blockquote><p>数据来源:中科院大气物理研究所、清华环境科学与工程系、北大环境科学系</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据;根据上一条的数据,若认为燃煤和燃油大致即为燃煤和机动车,则比例大约为<strong>46%~68%不等</strong>。</li></ul><hr><p>视频时间:00:23:58</p><p>图表引用:<br><img src=\"/images/UTD/002358.png\" alt=\"002358\"></p><p>旁白引用:</p><blockquote><p>“这种燃烧的强度有多大呢?我们可以来看看,在2014年,这个数值可以代表全球化石能源燃烧的燃烧强度,你可以看看中国有多亮,那个烧得发红发亮发白的地方就是我们国家。从图表来看的话,它比欧洲高出三到四倍。”</p></blockquote><p>数据来源:美国航空航天局(NASA)</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/Rw8Lp8k\" target=\"_blank\" rel=\"noopener\">Luke liking</a><br>提供原始视频连接:<a href=\"http://t.cn/RwRmJaK\" target=\"_blank\" rel=\"noopener\">Nitrogen Dioxide from Aura/OMI, 2013-2014</a>,答主已搬运至<a href=\"http://t.cn/RwRmXqk\" target=\"_blank\" rel=\"noopener\">优酷</a></li><li>微博:<a href=\"http://t.cn/RwRzixS\" target=\"_blank\" rel=\"noopener\">NASA中文</a> 第三处<blockquote><p>这段视频取自2013年7月至2014年7月AURA/OMI卫星上对流层二氧化氮的数据。</p></blockquote></li></ul><hr><p>视频时间:00:25:07</p><p>图表引用:<br><img src=\"/images/UTD/002507.png\" alt=\"002507\"></p><p>旁白引用:</p><blockquote><p>“你们知道中国烧了多少煤吗?2013年的时候,我们已经是三十六亿吨了,但你知道全世界其它国家烧了多少煤吗?我们比全世界所有其他国家加起来烧的煤还要多。上一个达到这样消费量的是1860年的英国,但在之后他们为此付出了沉重的代价。”</p></blockquote><p>数据来源:中国煤炭工业协会</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>环球时报:<a href=\"http://t.cn/8FdNPc4\" target=\"_blank\" rel=\"noopener\">雾霾逼中国加速摆脱“煤依赖”</a><blockquote><p>2013年中国煤炭消费量为36.1亿吨,<strong>占世界煤炭消费总量的49.33%</strong>,而中国煤炭消费增量也占世界增量的30.43%。</p></blockquote></li></ul><hr><p>视频时间:00:27:19</p><p>旁白引用:</p><blockquote><p>“清洁空气联盟的测算显示,伦敦当时空气污染的程度可能远超中国当下,虽然当时并没有PM2.5的检测,但是大烟雾事件发生时,二氧化硫的浓度超过世界卫生组织标准的190倍,比中国PM2.5超标的倍数高出一个量级。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>在清洁空气联盟(CCAC)网站上未找到相关数据</li><li>实际来源可能是由伦敦市出版的伦敦大烟雾事件纪念册:<a href=\"http://t.cn/zYt0WAt\" target=\"_blank\" rel=\"noopener\">“50 years on The struggle for air quality in London since the great smog of December 1952”</a><blockquote><p>“Sulphur dioxide (SO2) followed a similar pattern with concentrations rising from 0.41 mg/m<sup>3</sup> on 4 December to 2.15 mg/m3 on the 5th and to <strong>3.83 mg/m<sup>3</sup></strong> on both the 7 and 8 December.”</p></blockquote></li><li>世界卫生组织<a href=\"http://t.cn/hxqwX\" target=\"_blank\" rel=\"noopener\">《空气质量准则》</a>规定SO2的24小时平均浓度限值为20 μg/m<sup>3</sup>,3.83mg=3830μg,即<strong>190倍左右</strong>。</li></ul><hr><p>视频时间:00:27:45</p><p>图表引用:<br><img src=\"/images/UTD/002745-1.png\" alt=\"002745-1\"><br><img src=\"/images/UTD/002745-2.png\" alt=\"002745-2\"></p><p>旁白引用:</p><blockquote><p>“所以在1960年代,在这个大烟雾事件发生完之后,其他的国家纷纷开始减少和控制自己的煤炭用量,但当时恰逢中国改革开放的开始,这个已经封闭了多年和落后了多年的国家,迫切地需要一种巨大的能量来让自己起飞,它的选择是煤炭,这是它增长的曲线。”<br>数据来源:美国自然资源保护协会(NRDC)</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>在NRDC官网以关键词”coal”搜索未找到任何相关数据</li><li>未找到“煤炭占全社会排量百分比”曲线</li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 9-2 能源消费总量及构成<br>煤炭<strong>占我国能源消费总量</strong>的比重自1978年至今始终处于70%上下,最低为66.0(2013年),最高为76.2(1990年),<strong>占全社会排量</strong>比重不应当出现不降反升的情况。</li></ul><hr><p>视频时间:00:28:15</p><p>旁白引用:</p><blockquote><p>“那中国这么多煤用在了哪儿呢?2013年的这三十六亿吨,我们可以看一下,它其中三亿八千万吨烧在了京津冀,而这三亿八千万吨当中,有三个亿烧在河北。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>未找到2013年相关数据,2012年数据参见:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a><br>2012年煤炭消费量为:北京2270万吨,天津5298万吨,河北31359万吨,合计38927万吨,基本无误。</li></ul><hr><p>视频时间:00:32:43</p><p>文字引用:</p><blockquote><p>北京强霾天气外来污染源输送占1/5</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国气象局:<a href=\"http://t.cn/Rwrkg4x\" target=\"_blank\" rel=\"noopener\">北京强霾天气外来污染源输送占1/5</a></li></ul><hr><p>视频时间:00:33:23</p><p>旁白引用:</p><blockquote><p>“在你们江苏,每三十公里就有一个电厂。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>冰冷雨天:<a href=\"http://t.cn/RwdxVk2\" target=\"_blank\" rel=\"noopener\">柴静老师,请注意一点细节</a><blockquote><p>实际上这个“三十公里”的说法出于去年两会时全国政协委员、国家电网公司董事长、党组书记刘振亚的一段发言。<br>刘振亚的发言原文是这样的:“特别是近年来东中部地区雾霾严重,给人民群众生产生活、身心健康造成严重危害。目前京津冀鲁、长三角地区单位国土面积煤电装机分别是西北地区的13倍、26倍,<strong>沿长江平均30公里一个电厂</strong>,从南京到镇江平均10公里一个电厂,严重超出环境承载能力”。<br>刘振亚的这段话就不太有说服力,比如从武汉到南京600公里长江,有20个发电厂吗?而柴静老师再一次加以改造,成了“江苏省内”这就不应该了。而且从“长江沿岸”到“江苏省内”就是从直线概念升级到了平面概念,这时只能说“每平方公里多少”,而不能还说“隔多少公里一个”了。</p></blockquote></li></ul><hr><p>视频时间:00:33:43</p><p>旁白引用:<br>“而被夹在中间的上海,它自己烧的煤也不少,它是单位燃煤量最高的国际性大都市,每一平方米的土地上要烧掉十公斤煤炭。”</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到关于“单位燃煤量最高的国际性大都市”的相关数据</li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>上海的单位城区面积燃煤量参见:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a><blockquote><p>城区面积:6341平方公里<br>2012年煤炭消费量:5703万吨<br>单位面积煤炭消费量:8.99公斤/平方米</p></blockquote></li></ul><hr><p>视频时间:00:34:29</p><p>图表引用:<br><img src=\"/images/UTD/003429.png\" alt=\"003429\"></p><p>旁白引用:</p><blockquote><p>“那这样的褐煤我们用了多少呢?这是这些年来我们生产它的曲线。”</p></blockquote><p>数据来源:中国煤炭工业协会</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>未找到中国煤炭工业协会的数据</li><li><a href=\"http://t.cn/Rw1umEf\" target=\"_blank\" rel=\"noopener\">KNOEMA世界数据图册</a><br><img src=\"/images/UTD/003429-1.png\" alt=\"003429-1\"></li></ul><hr><p>视频时间:00:35:34</p><p>旁白引用:</p><blockquote><p>“煤是可以弄干净的,英国人是百分之九十五以上都给它洗干净,那么在中国我们洗了多少呢?我们才洗了不到一半的煤,剩下的另外一半当中基本上有几亿吨是没有任何用,而且又污染又浪费的叫灰分的东西。我们就这么把它拉着,用汽车拉用火车拉,满世界地跑,再放到煤炉子把它烧出来,毫无作用。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>中华商务网:<a href=\"http://t.cn/Rw13v4v\" target=\"_blank\" rel=\"noopener\">我国煤炭入洗发展空间巨大</a><blockquote><p>美国原煤入洗率在60%左右,南非和俄罗斯在60%以上,澳大利亚和<strong>英国入洗率在75%以上</strong>,加拿大和德国为95%,而<strong>我国(2002年)只有34%</strong>。</p></blockquote></li><li>中国能源报:<a href=\"http://t.cn/Rw13vDf\" target=\"_blank\" rel=\"noopener\">国家将不再新增煤制气项目</a><blockquote><p>中国煤炭工业协会的数据显示,2013年<strong>我国原煤入洗率仅为59.8%</strong>。</p></blockquote></li><li>百度知道:<a href=\"http://t.cn/Rw13PTP\" target=\"_blank\" rel=\"noopener\">燃煤脱硫的简史及其发展</a><blockquote><p><strong>我国当前的煤炭入洗率较低,大约在20%左右</strong>,而美国为42%,<strong>英国为94.9%</strong>,法国为88.7%,日本为98.2%。</p></blockquote></li></ul><hr><p>视频时间:00:36:03</p><p>图表引用:<br><img src=\"/images/UTD/003603.png\" alt=\"003603\"></p><p>旁白引用:</p><blockquote><p>“即使在中国城市里面,我们现在都有四分之一的人是在烧散煤,他们烧的量不算大,只有百分之二十,但是这百分之二十排放出来的二氧化硫跟所有的大电厂加起来一样多,就这么厉害。”</p></blockquote><p>数据来源:环保部环境与经济政策研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到与“四分之一”相关的数据</li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>新华网:<a href=\"http://t.cn/RP5cvS5\" target=\"_blank\" rel=\"noopener\">“煤改气”难成我国雾霾防治“撒手锏”</a><blockquote><p>据环保部环境与经济政策研究中心研究员王汉臣估算,工业锅炉、家庭取暖、餐饮用煤等“散煤”的煤炭消耗量在6亿吨-7亿吨,<strong>占全国煤炭消耗量的20%</strong>,仅次于电力行业,在非电行业最多,排放二氧化硫接近1000万吨,<strong>与电力行业的二氧化硫齐平</strong>,排放氮氧化物是320多万吨,位列第三,仅次于电力和机动车,高于水泥行业排放总量。</p></blockquote></li></ul><hr><p>视频时间:00:36:17</p><p>图表引用:<br><img src=\"/images/UTD/003617.png\" alt=\"003617\"></p><p>旁白引用:</p><blockquote><p>“最要命的是,它的毒性还非常大,在北京,冬天PM2.5的毒性远远大于夏天,光致癌物的含量,就是夏天的25倍。它为什么会这么高,就是因为散烧煤炭是主要原因之一,它的致癌物的排放因子是非常高的。”</p></blockquote><p>数据来源:北京大学环境科学和工程学院</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据来源,可参考:</li></ul><ol><li>《北京市大气颗粒物中多环芳烃的组成》<a href=\"http://t.cn/RwdomQn\" target=\"_blank\" rel=\"noopener\">(黄翠玲等,2001)</a></li><li>《北京市部分地区大气气溶胶中多环芳烃污染特征及污染源探讨》<a href=\"http://t.cn/Rwdo3Qb\" target=\"_blank\" rel=\"noopener\">(曾凡刚等,2002)</a></li><li>“Seasonal Variation and Spatial Distribution of Polycyclic Aromatic Hydrocarbons in Atmospheric PM10 of Beijing, People’s Republic of China” <a href=\"http://t.cn/RwdoBlz\" target=\"_blank\" rel=\"noopener\">(Zhou et al., 2005)</a></li><li>“Annual variation of particulate organic compounds in PM2.5 in the urban atmosphere of Beijing”<a href=\"http://t.cn/Rwdodc1\" target=\"_blank\" rel=\"noopener\">(Huang et al., 2006)</a></li><li>“Molecular distribution and seasonal variation of hydrocarbons in PM2.5 from Beijing during 2006”<a href=\"http://t.cn/RwdoD30\" target=\"_blank\" rel=\"noopener\">(Li et al., 2013)</a></li><li>《京津冀地区城市空气颗粒物中多环芳烃的污染特征及来源》<a href=\"http://t.cn/Rwdosbg\" target=\"_blank\" rel=\"noopener\">(王超等,2015)</a></li></ol><ul><li>综合以上文献,北京市不同检测点冬季与夏季PAHs含量的比值在<strong>2至几百之间不等</strong></li></ul><hr><p>视频时间:00:37:42</p><p>图表引用:<br><img src=\"/images/UTD/003742.png\" alt=\"003742\"></p><p>旁白引用:</p><blockquote><p>“在中国,煤炭消耗量越多的地方,它的PM2.5的浓度几乎也就是越高。”</p></blockquote><p>数据来源:国家发改委气候战略研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>左图:未找到直接来源;或可根据:国家统计局:<a href=\"http://t.cn/z8yHJGx\" target=\"_blank\" rel=\"noopener\">国家数据</a> 推算。</li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>右图:“Global Estimates of Ambient Fine Particulate Matter Concentrations from Satellite-Based Aerosol Optical Depth: Development and Application”<a href=\"http://t.cn/zjD8Mxs\" target=\"_blank\" rel=\"noopener\">(Aaron van Donkelaar et al., 2010)</a><br>(单位应当是<strong>微克/立方米而非毫克/立方米</strong>)<br><img src=\"/images/UTD/003742-1.png\" alt=\"003742-1\"></li></ul><hr><p>视频时间:00:38:24</p><p>文字引用:</p><blockquote><p>2014年<br>74个主要城市<br>PM2.5的年均浓度平均下降了11.1%<br>但能达到国家标准的只有8个城市</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>新京报:<a href=\"http://t.cn/Rw5dmEI\" target=\"_blank\" rel=\"noopener\">去年74个城市PM2.5平均下降11%</a></li></ul><hr><p>视频时间:00:39:09</p><p>图表引用:<br><img src=\"/images/UTD/003909.png\" alt=\"003909\"></p><p>旁白引用:</p><blockquote><p>“我们的车在十年之间增加了将近一个亿,光北京2010年一年我们增加的车是多少呢?八十万。”</p></blockquote><p>数据来源:公安部交通管理科学研究所</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 18-25 民用汽车拥有量</li><li>北京晨报:<a href=\"http://t.cn/Rw1dIQK\" target=\"_blank\" rel=\"noopener\">治堵这一年 摇号减少60万新车上路</a><blockquote><p>2010年,北京新增汽车<strong>81万辆</strong>。</p></blockquote></li></ul><hr><p>视频时间:00:39:31</p><p>图表引用:<br><img src=\"/images/UTD/003931.png\" alt=\"003931\"></p><p>旁白引用:</p><blockquote><p>“北京本地的污染源当中,最大的就来自机动车。”</p></blockquote><p>数据来源:北京市环保局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>北京市环保局:<a href=\"http://t.cn/8sOWPUL\" target=\"_blank\" rel=\"noopener\">北京市PM2.5来源解析正式发布</a></li><li>知乎:<a href=\"http://t.cn/Rw8gytg\" target=\"_blank\" rel=\"noopener\">大脸撑在小胸</a></li></ul><ol><li>原答主Jun引用<a href=\"http://t.cn/8kBk1xT\" target=\"_blank\" rel=\"noopener\">中国科学院大气物理研究所张仁健研究员的论文</a>,来说明机动车对北京PM2.5的影响并非31.1%,而是4%</li><li>原答主未考虑机动车间接导致的空气污染,除了汽车尾气以外,一半的二次无机气溶胶和一半的土壤尘也由机动车造成,综合考虑后根据原答主提供的论文,也可推算出机动车对PM2.5的贡献<strong>在20-30%左右</strong>。</li><li>柴静的数据图没有问题,问题在于她对“机动车”这一项的理解。</li></ol><hr><p>视频时间:00:39:35</p><p>旁白引用:</p><blockquote><p>“某城市(杭州)一年的雾霾日达到两百天以上。”<br>“因为它(杭州)是全国人均机动车保有量最大的城市,平均两个人一辆车,所以它的机动车污染要占到(接近)百分之四十。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>中国新闻网:<a href=\"http://t.cn/Rw1dyjR\" target=\"_blank\" rel=\"noopener\">杭州机动车保有量破110万 后市场混乱或迎洗牌</a><blockquote><p>截至2013年底,浙江省杭州市机动车保有量突破110万辆,相当于<strong>每三人拥有一辆车</strong>,是国内<strong>人均汽车拥有量最高的城市</strong>。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>钱江晚报:<a href=\"http://t.cn/RwjkfUr\" target=\"_blank\" rel=\"noopener\">2014年,杭州有154个雾霾天</a><blockquote><p>2013年1到10月,据气象部门统计,杭州主城区出现了<strong>209天雾霾天</strong><br>杭州曾做过一份《杭州市大气灰霾成因及关键污染因子预防控制研究》,其中说到机动车尾气、燃煤、扬尘和其他,是杭州PM2.5的主要来源。大致的比例是这样,<strong>机动车尾气占近40%</strong>,燃煤占33%,扬尘占8%,其他则占19%。</p></blockquote></li></ul><hr><p>视频时间:00:40:19</p><p>旁白引用:</p><blockquote><p>“大家经常听到的一个问题是,东京不是比北京车多吗,我们怎么没有看到这么严重的污染呢?北京市交通委给我的解释是这样的:东京百分之九十以上的人在坐轨道交通,他们只有不到百分之六的人在开车。北京有多少呢?北京有百分之三十四的人开车,而且是在中心城区基本这么堵着,一堵的时候一怠速排放是两倍以上,一刹车重金属的排放会更多。”</p></blockquote><p>数据来源:北京交通发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国汽车报:<a href=\"http://t.cn/8sq57xA\" target=\"_blank\" rel=\"noopener\">李春利:治理交通拥堵可借鉴“东京模式”</a><blockquote><p>目前,东京是世界上轨道交通最发达的城市之一,几十条城市铁路和地铁组成的立体公共交通网覆盖整个东京圈,总里程超过2300公里,轨道交通系统每天运送旅客2000多万人次,承担东京全部客运量的86%。<strong>在早高峰时的市中心区, 有91%的人乘坐轨道交通工具,而乘坐小汽车的仅为6%</strong>。<br>出行方式的数据差距更大。<strong>北京总体的小汽车出行比例是34%</strong>,东京是13%,北京是东京的2.5倍。轨道交通出行东京达到了86%,北京是12%,东京是北京的7倍,这两个关键数据看出了北京和东京的差距所在。</p></blockquote></li><li>(虽然数据对上了,但是东京的数据用的是早高峰时,北京用的数据是整体出行比例,为什么不都用整体出行比例?)</li><li>Bureau of Urban development Tokyo Metropolitan Government: <a href=\"http://t.cn/RwdxHQI\" target=\"_blank\" rel=\"noopener\">Basic concept of public transport planning in Tokyo</a><br><img src=\"/images/UTD/004019.png\" alt=\"004019\"></li></ul><hr><p>视频时间:00:40:44</p><p>旁白引用:</p><blockquote><p>“所以在北京每天的高峰期的时候,六环以内每个小时的PM2.5排放量是多少?一吨。”</p></blockquote><p>数据来源:南开大学城市交通污染防治研究中心</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到一手数据,但可信度较高,参考报道:城市快报:<a href=\"http://t.cn/R7p2aDb\" target=\"_blank\" rel=\"noopener\">两分钟“画”出汽车“污染地图”</a><blockquote><p>日前,南开大学城市交通污染防治研究中心毛洪钧教授课题组研发了一套可完整“追踪”城市每条道路上机动车排放情况,并实时绘制出机动车“污染地图”的智能系统。该系统可为城市空气质量模型、雾霾预报预警系统提供详细的机动车“排放清单”,同时也可为政府限牌、限号、黄标车淘汰、路查路检、低排区限行等调控政策提供效果评估和改进建议。</p></blockquote></li></ul><hr><p>视频时间:00:40:53</p><p>图表引用:<br><img src=\"/images/UTD/004053.png\" alt=\"004053\"></p><p>数据来源:北京交通发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>北京交通发展研究中心:<a href=\"http://t.cn/RwdiLzx\" target=\"_blank\" rel=\"noopener\">中国城镇化机动化进程中的交通问题</a><br><img src=\"/images/UTD/004053-1.png\" alt=\"004053-1\"><br>数据应当为:1公里以内7%,<strong>2公里以内19%(7%+12%,而非12%)</strong>,5公里以内44%</li></ul><hr><p>视频时间:00:42:39</p><p>图表引用:<br><img src=\"/images/UTD/004239.png\" alt=\"004239\"></p><p>旁白引用:</p><blockquote><p>“在北京每一天在凌晨的时候,都会出现污染的峰值,而且是每天恒定出现,永远比当天下午的时候有机碳的排放大概要高出两倍。”</p></blockquote><p>数据来源:中科院大气物理研究所</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“元素碳和有机碳”一手数据,可参考</li></ul><ol><li>《北京地区夏季PM10污染的数值模拟研究 》<a href=\"http://t.cn/RwdPS1B\" target=\"_blank\" rel=\"noopener\">(王雪松,李金龙,2003)</a><blockquote><p>结果表明,北京市城近郊区夏季的PM10具有明显的时空变化规律。一般在半夜前后和早晨常表现出较高的浓度……有机碳气溶胶以及PM10的浓度水平和变化规律则主要受一次成分的影响。</p></blockquote></li><li>“Estimation of diurnal and seasonal variations of primary and secondary organic carbon concentrations using highly time resolved PM2.5 measurements”<a href=\"http://t.cn/RwdPpLi\" target=\"_blank\" rel=\"noopener\">(Kim et al., 2011)</a><blockquote><p>Diurnal variations of the estimated SOCs show peaks at around 15:00 local time (LT) while POC, TOC and EC concentrations show peaks at morning and midnight.</p></blockquote></li><li>“Elemental and organic carbon in the urban environment of Athens. Seasonal and diurnal variations and estimates of secondary organic carbon”<a href=\"http://t.cn/RwdPHXr\" target=\"_blank\" rel=\"noopener\">(Grivas et al., 2012)</a><blockquote><p>Significant evening OC levels were observed during both periods. Organics from fresh afternoon traffic emissions, combined with persisting secondary particles, were <strong>trapped in the shallow boundary layer</strong>, resulting at high levels around midnight.</p></blockquote></li></ol><ul><li>纪录片在此处试图引导观众建立以下联系:北京凌晨的有机碳污染峰值是由于夜晚柴油车排放的缘故;而上述论文一定程度上反驳了这一观点。</li></ul><hr><p>视频时间:00:44:16</p><p>图表引用:<br><img src=\"/images/UTD/004416-1.png\" alt=\"004416-1\"></p><p>旁白引用:</p><blockquote><p>“所以柴油车的比例也不算高,它跟散煤差不多,并不多,只占百分之十七,但是它的氮氧化物的排放,要占所有机动车的百分之七十,它的一次性颗粒物的排放,要占到多少呢?99%。”</p></blockquote><p>数据来源:国家环保部《2013年机动车污染防治年报》</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>国家环保部:<a href=\"http://t.cn/RwrgFPs\" target=\"_blank\" rel=\"noopener\">《2013年机动车污染防治年报》</a><blockquote><p>柴油车<strong>占16.1%</strong><br>全国柴油车排放的NOx接近汽车排放总量的70%,<strong>PM超过90%</strong></p></blockquote></li></ul><hr><p>视频时间:00:48:55</p><p>图表引用:<br><img src=\"/images/UTD/004855.png\" alt=\"004855\"></p><p>文字引用:</p><blockquote><p>从2004年到2014年,车用柴油标准三次落后于机动车排放标准的实施。</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到视频中原表格,可参考</li></ul><ol><li><a href=\"http://t.cn/RwrkU4F\" target=\"_blank\" rel=\"noopener\">Global Motor Vehicle Emissions Regulations</a><br>欧盟、美国(加州/其它)、日本、阿根廷、巴西、智利、哥伦比亚、澳大利亚、香港、印度、韩国的部分排放标准</li><li>ICCT(The International Council on Clean Transportation): <a href=\"http://t.cn/Rwrk4VG\" target=\"_blank\" rel=\"noopener\">Global passenger vehicle standards</a><br>巴西、加拿大、中国、欧盟、印度、日本、墨西哥、韩国、美国的与乘用车相关的各类标准</li><li>TransportPolicy.net: <a href=\"http://t.cn/Rwrk55X\" target=\"_blank\" rel=\"noopener\">Global Comparison: Fuels</a><br>中国、印度、欧盟、美国的燃油标准对比;点击左侧“Fuels”可查阅澳大利亚 、巴西、加拿大、智利、中国、欧盟、印度、伊朗、日本、墨西哥、俄罗斯、韩国、美国(加州/其它)及国际标准(ISO)的燃油标准</li></ol><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>中国环境报:<a href=\"http://t.cn/RwrFMbD\" target=\"_blank\" rel=\"noopener\">油品标准制定权该由谁执牛耳?</a><blockquote><p>在柴油方面,<strong>从2004年实施国2机动车排放标准以来,到2014年国4重型柴油车(机)排放标准的实施,车用柴油标准已经3次落后于机动车排放标准的实施</strong>,国2、国3排放标准是在国1柴油标准条件下实施的,国4排放标准是在国3柴油标准条件下实施的。</p></blockquote></li></ul><hr><p>视频时间:00:49:22</p><p>图表引用:<br><img src=\"/images/UTD/004922.png\" alt=\"004922\"></p><p>文字引用:</p><blockquote><p>长期比发达国家低两到三个等级,<br>每提高一个等级,排放可减少10%。</p></blockquote><p>旁白引用:</p><blockquote><p>“我们看上去有好油,我们有国IV以及以上的油,但是这部分油只占百分之三的比例。这就是杭州的痛苦,杭州有钱,想买好油,但是2013年以前就是没有,费了九牛二虎之力,你只能烧国III的油。”</p></blockquote><p>数据来源:中国环境科学院车用燃料及添加实验室</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (二)第一点</li></ul><ol><li>现阶段我国油品质量的确总体低于发达国家,不过并不存在量级的差别,并且未来目标上也将与发达国家看齐。</li><li>京V汽柴油的硫含量相比京Ⅳ油品中的硫含量从50ppm降到了10ppm,主要指标符合同阶段欧V油品标准。</li><li>国Ⅳ以及以上标准的成品油占的比例<strong>不止3%</strong>。</li></ol><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到视频中原表格,参见上一条相关数据。</li></ul><hr><p>视频时间:00:50:18</p><p>旁白引用:</p><blockquote><p>“加油站如果不加这个油气回收装置的话,它蒸发出来的量是这么大,每一升大概要蒸发出1.5克到大气空中来。而且我们平常开车的油箱,也会自己蒸发出来,它蒸发强度大到什么程度?就是在北京,我们光蒸发出来的碳氢化合物,就比汽车尾气排放出来的还要多。”</p></blockquote><p>数据来源:北京市环保局机动车排放管理处</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《汽油车加油过程油气蒸发损失试验方法》<a href=\"http://t.cn/RwrdsnO\" target=\"_blank\" rel=\"noopener\">(何仁等,2010)</a><blockquote><p>所有的汽油车都要经过加油后才使用的,如果不采取任何措施,美国环境保护署估计的排放因子为1.32g/L。现在中国使用的一般汽油车只安装活性炭罐,根据文献估计排放因子可减小为1.02g/L。<br>实验结果:(ORVR指车载油气回收系统)<br><img src=\"/images/UTD/005018.png\" alt=\"005018\"></p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwrgzEq\" target=\"_blank\" rel=\"noopener\">第七章 废气控制</a><blockquote><p>1.油箱、化油器——挥发之汽油气体,主要为HC,<strong>占HC废气的20%</strong>。<br>2.曲轴箱——经活塞与汽缸壁漏入曲轴箱之吹漏气,主要为HC,占HC废弃的25%。<br>3.排气管——燃烧后排出之气体,主要为HC、CO、NOx,<strong>占HC废气的55%</strong>。</p></blockquote></li><li><a href=\"http://t.cn/Rwrg2hj\" target=\"_blank\" rel=\"noopener\">废气排放检测的必要性</a><blockquote><p>碳氢化合物总称为烃类,是发动机未燃尽的燃料或分解产生的气体。汽车排放污染物中的HC20%~25%来自曲轴箱窜气,<strong>20%来自化油器与燃油箱的蒸发;其他55%由排气管排出</strong>。</p></blockquote></li><li>环境保护部机动车排污监控中心:<a href=\"http://t.cn/RwrgtC5\" target=\"_blank\" rel=\"noopener\">实现城市机动车高效减排的科学管理建议[PPT]</a><br><img src=\"/images/UTD/005018-1.png\" alt=\"005018-1\"></li><li>综合上述资料,<strong>油箱蒸发出来的碳氢化合物不会比汽车尾气所排放出来的碳氢化合物更多</strong>。</li></ul><hr><p>视频时间:00:51:09</p><p>旁白引用:</p><blockquote><p>“这个小标委(石油燃料和润滑剂分技术委员会)呢,就是绝大部分,大部分吧,还是这个石化行业的人,大概是百分之六十七左右;大标委(全国石油产品和润滑剂标准化技术委员会)更多,大标委几乎,我记得好像百分之九十以上都是石化行业的人;标委会的秘书处也在石化行业那儿。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>法治周末:<a href=\"http://t.cn/zRiESM8\" target=\"_blank\" rel=\"noopener\">谁在控制油品标准制定</a><blockquote><p>而油品的标准制定,也掌握在全国石油产品和润滑剂标准化技术委员会(以下简称“石化标委会”)手中。根据记者了解,在石化标委会中,<strong>90.9%的委员来自石化系统</strong>、2.3%的委员来自汽车产业系统、4.5%的委员来自环保系统;主任委员来自中石化,且<strong>秘书处设在中石化</strong>。</p></blockquote></li><li><a href=\"http://t.cn/RwrsHbv\" target=\"_blank\" rel=\"noopener\">关于成立全国石油产品和润滑剂标准化技术委员会石油燃料和润滑剂分技术委员会等六个分技术委员会的批复</a><br>根据<a href=\"http://www.cptcstd.org/Attach/20080904113621375.doc\" target=\"_blank\" rel=\"noopener\">附件1:全国石油产品和润滑剂标准化技术委员会第一届石油燃料和润滑剂分技术委员会(SAC/TC280/SC1)委员名单[doc]</a>,委员共57人,其中41人属于石化行业(中国石油化工股份有限公司21人、中国石油天然气股份有限公司16人、中国海洋石油总公司2人、河南省洛阳石油化工工程公司1人、中海油气开发利用公司1人),<strong>占比71.93%</strong></li></ul><hr><p>视频时间:00:57:56</p><p>旁白引用:</p><blockquote><p>“在中国,除了我们的油品标准低之外,我们还有另外一个,在我出发之前远远没有想到的问题就是我们有一半的油是完全失控。全球十大港口现在有七个在中国;远洋货轮带来的污染之大,在靠近海岸线四百米的地方,它排放的污染相当于五十万辆大卡车。所以在深圳,百分之六十的二氧化硫是来自轮船的。”</p></blockquote><p>数据来源:国家环保部机动车排污监控中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>深圳商报:<a href=\"http://t.cn/RhgL9ki\" target=\"_blank\" rel=\"noopener\">深圳率先开展港口船舶污染实质性治理工作 每年2亿元补贴“绿色航运”</a><blockquote><p>目前,大多航船采用含硫率3%左右的重油作为燃料,每年排放二氧化硫约1.6万吨,<strong>占我市总排放量的65.8%左右</strong>;氮氧化物1.9万吨,占总排放量的14.1%左右;排放细颗粒物(PM2.5)1600吨,占总排放量的5.8%左右。研究表明,一艘排水量10万吨的大型集装箱船一次停靠过程中的PM2.5排放量就相当于5.95万辆国Ⅳ柴油货车一天的排放量。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (二)第三点<blockquote><p>根据微信公众号“新兴产业观察者”的数据,轮船用的<strong>重柴油或者渣油占燃料油的比例接近一半(47%)</strong>,而燃料油占总油量消耗量的比例是6.6%,也就是说,<strong>船舶运输占石油消耗的比例是3%</strong>。</p></blockquote></li><li>“四百米”和“五十万辆大卡车”可能是错误引用了此文 GAS2:<a href=\"http://t.cn/SqDZS8\" target=\"_blank\" rel=\"noopener\">One Container Ship Pollutes As Much As 50 Million Cars</a><blockquote><p>But newly released data from Europe suggests that <strong>a single container ship may cause as much pollution as 50 million cars</strong> and release as much as 5,000 tons of sulphur oxide into the air annually, contributing heavily to global warming.<br>Even if the sails were just used to get the ships away from port, that would represent a major reduction in pollution since 70% of ship emissions are <strong>within 400km of land</strong> and fuel consumption.</p></blockquote></li></ul><hr><p>视频时间:00:58:38</p><p>旁白引用:</p><blockquote><p>“也有可能你会生活在机场的边上,飞机一起飞,一吨煤油没有了,一落地,一吨煤油没有了,所以(广州)白云机场(每天)排污量相当于六十万辆出租车的同时排放。”</p></blockquote><p>数据来源:广东省气象局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>广州日报:<a href=\"http://t.cn/RwrBTPf\" target=\"_blank\" rel=\"noopener\">飞机起降成空气污染物新来源</a><blockquote><p>一些以前被忽视的排放源,其排放量也开始逐年增加,吴兑认为最为明显的就是飞机等航空器起降以及巡航带来的空气污染物排放,“比如一架空客A380,<strong>起降一次耗费的燃油量将近2吨</strong>,相当于数千辆汽车的排放。以一天白云机场起降近千架航班计算,<strong>排放量相当于几十万辆汽车</strong>。”</p></blockquote></li></ul><hr><p>视频时间:01:05:05</p><p>图表引用:<br><img src=\"/images/UTD/010505.png\" alt=\"010505\"></p><p>数据来源:国际钢铁协会、中国国家统计局</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>国际钢铁协会:<a href=\"http://t.cn/RwrBOwz\" target=\"_blank\" rel=\"noopener\">1980-2013年全球粗钢产量</a></li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 3-1 国内生产总值</li></ul><hr><p>视频时间:01:12:01</p><p>旁白引用:</p><blockquote><p>“在中国,每天要消失八十个自然村,我们房屋的平均寿命只有30年,在中国一共有200多家地级市,其中184家要建成国际大都市。现在中国有13亿人, 但如果我们把目前城市规划当中这些人口都加起来的话它是34个亿。”</p></blockquote><p>数据来源:国家城市和小城镇改革发展中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>都市快报:<a href=\"http://t.cn/zl1OpUO\" target=\"_blank\" rel=\"noopener\">过去10年 全国每天消失80个自然村</a><blockquote><p>中国文联副主席、国务院参事冯骥才10月20日透露,相关部门最新统计数字显示,我国的自然村十年前有360万个,现在则只剩270万个,“比较妥当的说法是<strong>每一天消失80至100个村落</strong>”。</p></blockquote></li><li>华商网:<a href=\"http://t.cn/RwrBIBB\" target=\"_blank\" rel=\"noopener\">中国住房平均寿命仅30年(图)</a><blockquote><p>其实,“国内新住宅平均寿命仅30年”早在2006年举行“全国第七届建筑物改造学术研讨会”上就提出来了。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>人民网:<a href=\"http://t.cn/RwrBx2d\" target=\"_blank\" rel=\"noopener\">183个城市争建国际大都市 全不顾实事求是</a><blockquote><p>在2005年7月21日召开的城市总体规划修编工作座谈会上,建设部部长汪光焘批评有的城市在城市定位上不切实际,盲目追求高速度和高标准。他表示,全国竟有<strong>183个城市</strong>提出建立“现代化国际大都市”的目标,30多个城市提出要建中心商务区,这实际上是透支和浪费日渐缺乏的发展资源。</p></blockquote></li><li>新华网:<a href=\"http://t.cn/8sjnKhq\" target=\"_blank\" rel=\"noopener\">90%地级市争建新城新区 如何避免成“死城”?</a><blockquote><p>“有研究数据显示,全国<strong>新城新区规划人口达到34亿</strong>。这意味着现在中国一倍的人口也装得下。”国家发展改革委城市和小城镇改革发展研究中心副主任乔润令说,由于规划脱离实际,各个新区为招商引资互相压价,结果往往“两败俱伤”。</p></blockquote></li></ul><hr><p>视频时间:01:13:01</p><p>旁白引用:</p><blockquote><p>“国务院发展研究中心做的测算是:十五年之后,我们的耗煤量会达到六十亿吨,我们的用车量会达到四亿辆。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>中国财政部、国务院发展研究中心和世界银行:《2030年的中国:建设现代、和谐、有创造力的高收入社会》<a href=\"http://t.cn/zObF7Rf\" target=\"_blank\" rel=\"noopener\">英文版</a><a href=\"http://t.cn/zOVMypk\" target=\"_blank\" rel=\"noopener\">中文版</a><br><img src=\"/images/UTD/011301.png\" alt=\"011301\"><blockquote><p>“ In the $10 price scenario, China uses 492 million tons less coal compared with business as usual in 2030 (a difference of 15 percent). Under the $20 price scenario, coal use drops by 770 million tons compared with business as usual (a difference of 23 percent). ”<br>(原文未列出Business as usual时的使用量,但根据上文推算,<strong>该值在3280~3350 million tons左右,即三十亿吨</strong>。)</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>中国经济新闻网:<a href=\"http://t.cn/8sn0KOy\" target=\"_blank\" rel=\"noopener\">王青:中国汽车市场将面临需求和制度的双重转变</a><blockquote><p>经过四种预测方法对中国汽车中长期增长情况的预测,到2023年,中国汽车千人拥有量和总保有量将分别达到235辆和<strong>3.4亿辆</strong>,届时新车产销规模约为3800万辆。<br>(作者单位:国务院发展研究中心市场经济研究所)</p></blockquote></li><li>《中国乘用车与商用车保有量预测》<a href=\"http://t.cn/Rwr1rI6\" target=\"_blank\" rel=\"noopener\">(郝瀚等,2011)</a><blockquote><p>中国汽车保有量在2020年和2030年将分别达到2.5—2.9亿辆和<strong>4—5.2亿辆</strong>,其中乘用车保有量增长速度相对较快,且存在一定不确定性,在2020年和2030年将分别达到2.2—2.6亿辆和3.5—4.8亿辆。商用车保有量在2020年和2030年将分别达到0.33和0.47亿辆。</p></blockquote></li><li><a href=\"http://t.cn/Rwr1kRP\" target=\"_blank\" rel=\"noopener\">《中国发展低碳汽车燃料政策建议报告》</a><blockquote><p>据国际能源署 (IEA)预计,中国到2020年汽车保有量将达2.3亿辆,2030年将<strong>超过4亿辆</strong>。(IEA, 交通能源与 CO2,2009)<br>(但在IEA:<a href=\"http://t.cn/Rwr1sCr\" target=\"_blank\" rel=\"noopener\">Transport, Energy and CO2</a>中并未找到相关表述。)</p></blockquote></li><li>清华大学中国车用能源研究中心:<a href=\"http://t.cn/RwreeZ7\" target=\"_blank\" rel=\"noopener\">《中国车用能源展望2012》</a><br><img src=\"/images/UTD/011301-1.png\" alt=\"011301-1\"></li></ul><hr><p>视频时间:01:15:43</p><p>图表引用:<br><img src=\"/images/UTD/011543.png\" alt=\"011543\"></p><p>数据来源:中科院</p><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到中科院相关支持数据</li><li>北京市环境保护局:<a href=\"http://t.cn/RwrCWdm\" target=\"_blank\" rel=\"noopener\">APEC蓝的启示</a><blockquote><p>APEC会议空气质量保障措施效果初步评估结果:<br>二氧化硫(SO2)、氮氧化物(NOx)、可吸入颗粒物(PM10)、细颗粒物(PM2.5)、挥发性有机物(VOCs)(以下均用简称)等减排比例分别达到54%、41%、68%、63%和35%左右。<br>PM2.5、PM10、SO2、二氧化氮(NO2)的浓度……比去年同期分别下降55%、44%、57%和31%</p></blockquote></li><li>北京市环境保护局:<a href=\"http://t.cn/RwrC81u\" target=\"_blank\" rel=\"noopener\">减排才是硬道理APEC空气质量保障措施效果评估结果出炉</a><blockquote><p>APEC会议空气质量保障措施效果评估结果:<br>全市分别削减二氧化硫(SO2)、氮氧化物(NOx)、可吸入颗粒物(PM10)、细颗粒物(PM2.5)和挥发性有机物(VOCs)排放约39.2%、49.6%、66.6%、61.6%和33.6%,平均削减50%左右。</p></blockquote></li></ul><hr><p>视频时间:01:16:34</p><p>旁白引用:</p><blockquote><p>“中国的节能环保产业现在可能是一年的产值37000多亿,吸纳就业人口可能是3900多万人,这也是一个产业。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“37000多亿”和“3900多万人”的支持数据。</li><li>国务院:<a href=\"http://t.cn/Rwr9VSK\" target=\"_blank\" rel=\"noopener\">“十二五”节能环保产业发展规划</a><blockquote><p>据测算,2010年,我国节能环保产业<strong>总产值达2万亿元,从业人数2800万人</strong>。<br>到2015年, 节能环保产业<strong>总产值达到4.5万亿元</strong>。</p></blockquote></li></ul><hr><p>视频时间:01:17:39</p><p>图表引用:<br><img src=\"/images/UTD/011739.png\" alt=\"011739\"></p><p>旁白引用:</p><blockquote><p>“但是从1970年以来,洛杉矶的车辆还在上升,上升了这么多,三倍;但它们排放呢,降低了百分之七十五。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>知乎:<a href=\"http://t.cn/RwQGtLm\" target=\"_blank\" rel=\"noopener\">Jun</a> 修订4 原答10(回答已删除)</li></ul><ol><li>VMT的数据多了个0,扩大了10倍。</li><li>汽车和卡车用量(VMT)和排放总量(Cumulative CA vehicle emissions for NOx and HCs)的数据来源:<a href=\"http://t.cn/Rw8hGF5\" target=\"_blank\" rel=\"noopener\">Key Events in the History of Air Quality in California</a>;其中VMT数据吻合但有缺失,但Emissions和左侧Pollution数据不吻合</li></ol><table><thead><tr><th align=\"center\">年份</th><th align=\"center\">VMT(billion)</th><th align=\"center\">Emissions(million tons/year)</th></tr></thead><tbody><tr><td align=\"center\">1970</td><td align=\"center\">110</td><td align=\"center\">1.6</td></tr><tr><td align=\"center\">1980</td><td align=\"center\">155</td><td align=\"center\">1.6</td></tr><tr><td align=\"center\">1990</td><td align=\"center\">242</td><td align=\"center\">1.4</td></tr><tr><td align=\"center\">1995</td><td align=\"center\">271</td><td align=\"center\">1.1</td></tr><tr><td align=\"center\">2000</td><td align=\"center\">280</td><td align=\"center\">1.2</td></tr></tbody></table><hr><p>视频时间:01:18:35</p><p>旁白引用:</p><blockquote><p>“(洛杉矶)一千七百多万人,差不多一千三百多万辆车,都快人手一辆了。整个加州,一天之内,8.5亿的行程,所有的燃料加起来都够来回月球一千六百趟。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>California Department of Transportation: <a href=\"http://t.cn/Rwr6aAK\" target=\"_blank\" rel=\"noopener\">Historical Monthly Vehicle Miles of Travel 1972 -2014 (In Billions)</a> (每天8.5亿公里,折合至英里/每月为15.845billion,基本吻合)</li></ul><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><a href=\"http://t.cn/Rwr6PC4\" target=\"_blank\" rel=\"noopener\">City-data</a><blockquote><p>洛杉矶市2013年人口:3,884,307</p></blockquote></li><li>Wikipedia: <a href=\"http://t.cn/zO60Akd\" target=\"_blank\" rel=\"noopener\">Los Angeles County</a><blockquote><p>洛杉矶郡2013年人口:10,017,068</p></blockquote></li><li>Los Angeles Almanac: <a href=\"http://t.cn/Rwr6GzR\" target=\"_blank\" rel=\"noopener\">Vehicle Registrations</a><blockquote><p>洛杉矶郡2008年登记各类车辆共计:7,499,722</p></blockquote></li><li>STREETS BLOG LA: <a href=\"http://t.cn/8kJoEIK\" target=\"_blank\" rel=\"noopener\">Density, Car Ownership, and What It Means for the Future of Los Angeles</a><br>(此处Los Angeles指The greater Los Angeles Region)<br><img src=\"/images/UTD/011835.png\" alt=\"011835\"></li></ul><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“一千六百趟”的相关数据支持</li></ul><hr><p>视频时间:01:19:11</p><p>文字引用:</p><blockquote><p>加州柴油车1061745辆<br>每辆车平均耗油量35.2升/车/天</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到“1061745辆”和“35.2升/车/天”的支持数据</li><li>Diesel Techonology Forum: <a href=\"http://t.cn/RwrhQG5\" target=\"_blank\" rel=\"noopener\">Total Diesel Passenger Vehicles</a><blockquote><p>California 2013 Diesel Total 609,212</p></blockquote></li><li>Board of Equalization - State of California: <a href=\"http://t.cn/RwrhktQ\" target=\"_blank\" rel=\"noopener\">TAXABLE DIESEL GALLONS 10 YEAR REPORT (NET OF REFUNDS)</a><br><img src=\"/images/UTD/011911.png\" alt=\"011911\"></li></ul><hr><p>视频时间:01:19:24<br>旁白引用:</p><blockquote><p>“根据加州空气资源部的估计,南加州空气中71%的致癌物质都由柴油车产生。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><strong>并非加州空气资源部的估计</strong></li><li>实际来源为南加州空气质量管理局(SCAQMD):<a href=\"http://t.cn/RwBkXap\" target=\"_blank\" rel=\"noopener\">Major Air Toxics Study Finds Vehicles Dominate Cancer Risk November 5, 1999</a><blockquote><p>AQMD found that that <strong>diesel soot accounted for 71% of the cancer risk</strong>, 1,3 butadiene (a byproduct of incomplete combustion in engines) 8% of the risk, benzene (mostly from motor vehicles) 7%, carbonyls (including formaldehyde and acetaldehydes from both mobile and stationary sources) 3%, and other pollutants (primarily from stationary sources) 11%.</p></blockquote></li></ul><hr><p>视频时间:01:21:03</p><p>文字引用:</p><blockquote><p>全年发出重卡柴油车检查传票29913起<br>最终结案3281起<br>罚款1336880美元</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>加州空气资源部(California Air Resource Board): <a href=\"http://t.cn/RwBDrSx\" target=\"_blank\" rel=\"noopener\">2013 ANNUAL ENFORCEMENT REPORT SUMMARY</a><blockquote><p>Heavy-duty Diesel Field Inspections and Citations: conducted 29,913 inspections, issued 3,281 citations, and collected $1,336,880 in penalties.<br>(重型柴油车路检及传票:进行<strong>29913次路检</strong>,发出<strong>3281张传票</strong>,罚款1336880美元。)</p></blockquote></li></ul><hr><p>视频时间:01:23:11</p><p>图表引用:<br><img src=\"/images/UTD/012311.png\" alt=\"012311\"></p><p>文字引用:</p><blockquote><p>洁净的大气带来的效益远高于治理大气污染的成本</p></blockquote><p>数据来源:美国南加州空气质量管理局</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li><strong>未在南加州空气质量管理局(SCAQMD)网站找到相关数据</strong></li><li>实际来源为美国环境保护局(United States Environment Protection Agency,EPA):<br><a href=\"http://t.cn/RwBDtXi\" target=\"_blank\" rel=\"noopener\">Benefits and Costs of the Clean Air Act - Second Prospective Study - 1990 to 2020</a><br>(且图表数值轴应当以Billion为单位)<br><img src=\"/images/UTD/012311-1.png\" alt=\"012311-1\"></li></ul><hr><p>视频时间:01:23:29</p><p>图表引用:<br><img src=\"/images/UTD/012329.png\" alt=\"012329\"></p><p>旁白引用:</p><blockquote><p>“我们来看看伦敦的曲线,很多人都跟我将说伦敦要等了四十年、五十年才把污染治好我们也得等这么久,但真的是这样吗?看看这儿(1960~1970段),从他们开始治理污染的前十年,他们就把污染物降低了百分之八十,这是一个极大的改善。”</p></blockquote><p>数据来源:Ren 2000, Hutchingson et al. 2004 and Okubo 2013</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>本处是全片唯一一次给出了论文(或文章)出处的地方,这三篇论文(或文章)分别是:</li></ul><ol><li>“Japanese Approaches to Environmental Management: Structural and Institutional Features.”<a href=\"http://t.cn/RwRZKOg\" target=\"_blank\" rel=\"noopener\">(Ren, 2000)</a></li><li>“Air Quality in London and Tokyo: A Comparison.”<a href=\"http://t.cn/RwRZOlU\" target=\"_blank\" rel=\"noopener\">(Hutchingson et al., 2004)</a></li><li>“Mt. Fuji Lovers Get Bonus with More Days to See It from Tokyo.” <a href=\"http://ajw.asahi.com/article/behind_news/social_affairs/AJ201302250003\" target=\"_blank\" rel=\"noopener\">Okubo, 2013</a></li></ol><ul><li>单从标题就可以看出,第一篇和第三篇论文(文章)都是关于日本东京的,而和伦敦完全没有关系,这个图表其实来自于第二篇论文。<br><img src=\"/images/UTD/012329-2.png\" alt=\"012329-2\"></li><li>之所以会引用第一篇和第三篇论文(文章),其实是转引自世界银行和中国国务院发展研究中心:<a href=\"http://t.cn/RPLVQev\" target=\"_blank\" rel=\"noopener\">Urban China : Toward Efficient, Inclusive, and Sustainable Urbanization</a><br>在这本书的第453页Box7.1中,使用了伦敦和东京治理污染的对比图,其中伦敦的图是直接从Hutchinson(2004)中拿来用的,而东京的图则加入了新的数据,因为Hutchinson(2004)关于东京的数据只到1987年左右就结束了;在Box的最后有这样一段文字:Box sources: Ren 2000; Bell, Davis, and Fletcher 2004; Hutchinson and others 2004; Okubo 2013。可以猜测纪录片引用的就是这个图表,但是要么是忘了检查引用是否正确,要么是原本打算放上伦敦与东京的对比数据但后来又把东京的数据撤掉了。<br><img src=\"/images/UTD/012329-3.png\" alt=\"012329-3\"></li></ul><hr><p>视频时间:01:23:48</p><p>旁白引用:</p><blockquote><p>“伦敦大烟雾事件发生的时候,英国的能源结构中将近百分之九十是煤炭;1953年,颗粒物的平均浓度超过欧盟标准的十倍。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到直接数据</li><li><a href=\"http://t.cn/RwBmx0c\" target=\"_blank\" rel=\"noopener\">60th Anniversary: Digest of United Kingdom Energy Statistics</a><br>1948年,煤炭<strong>占主要能源的90%</strong>,<strong>占最终能源使用则为77%</strong>。<br><img src=\"/images/UTD/012348-1.png\" alt=\"012348-1\"></li><li>“Reassessment of the Lethal London Fog of 1952: Novel Indicators of Acute and Chronic Consequences of Acute Exposure to Air Pollution”<a href=\"http://t.cn/zY1RFwo\" target=\"_blank\" rel=\"noopener\">(Bell and Davis, 2001)</a> (注:TSM应当为TSP)<br><img src=\"/images/UTD/012348-2.png\" alt=\"012348-2\"></li></ul><hr><p>视频时间:01:26:46</p><p>图表引用:<br><img src=\"/images/UTD/012646.png\" alt=\"012646\"></p><p>旁白引用:</p><blockquote><p>“现在世界平均的天然气占能源结构的比例是很高的,大概百分之二十四,因为天然气是相对清洁的能源。那中国是多少呢?我们只有百分之五。为什么我们这么低?从小到大我听到的标准答案是这样的:因为我们是一个煤多、贫油、少气的国家,这是真的吗?这是真的。但这完整吗?”</p></blockquote><p>数据来源:国务院发展研究中心</p><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>《BP世界能源统计年鉴》天然气在全球一次能源消费中所占份额<br><a href=\"http://t.cn/RwBnqWH\" target=\"_blank\" rel=\"noopener\">2013年</a>: 23.9%<br><a href=\"http://t.cn/R7qKXNQ\" target=\"_blank\" rel=\"noopener\">2014年</a>: 23.7%</li><li>中国国家统计局:<a href=\"http://t.cn/RZdT4MT\" target=\"_blank\" rel=\"noopener\">《中国统计年鉴2014》</a> 9-2 能源消费总量及构成<blockquote><p>天然气占能源消费总量的比重%:2011年5.0;2012年5.2;2013年5.8;</p></blockquote></li><li><a href=\"http://t.cn/RwBm5m4\" target=\"_blank\" rel=\"noopener\">中国煤炭的清洁高效利用——中国煤炭资源与消费</a><br><img src=\"/images/UTD/012646-1.png\" alt=\"012646-1\"></li></ul><hr><p>视频时间:01:27:22</p><p>旁白引用:</p><blockquote><p>“我们的天然气探明多少呢?百分之二十二。这意味着我们还有大量的天然气埋藏在中国大地之下黑暗不知名的深处。那我们的石油探明率是多少呢?百分之三十八。在每年探明的这些储量当中,比如说天然气,探明九千亿立方米,但我们开采了多少呢?一千亿立方米。”</p></blockquote><p>数据来源:国土部矿产资源储量评审中心</p><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>国土资源部:<a href=\"http://t.cn/RZEdWAz\" target=\"_blank\" rel=\"noopener\">《2014中国矿产资源报告》</a><blockquote><p>全国油气资源评价结果显示,截至2013年底,石油地质资源量1037亿吨,<strong>探明率33.8%</strong>,处于勘探中期。天然气地质资源量62万亿立方米,<strong>探明率18.4%</strong>,处于勘探早期。</p></blockquote></li></ul><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (三)第四点</li></ul><ol><li>从国土资源部网站可以查到的数据是,2012年我国天然气<strong>新增探明地质储量是9612.2亿立方米</strong>。</li><li>国家发改委的数据是,2012年我国<strong>天然气产量是1077亿立方米</strong>。</li><li>到2013年底,我国天然气可采常规剩余储量是3.3万亿立方米,当年<strong>天然气产量是1171亿立方米</strong>,储采比是28。</li><li>我国油气勘探总体进入低品位资源勘探阶段,勘探成本不断上升。</li><li>虽然中国的天然气生产实现全球第二大增量(9.5%,99亿立方米),但仍有巨大的缺口需要通过增加进口予以解决。</li></ol><hr><p>视频时间:01:27:58</p><p>旁白引用:</p><blockquote><p>“他说拿世界上最大的天然气生产国美国来说,它们有六千三百家天然气石油公司,我们有几家呢?3家;其中百分之七十集中在一家手里——中石油。美国有一百六十家天然气管道公司,我们(主体管道)有几家呢?3家,其中百分之七十集中在一家手里——中石油。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:#0f0\"><span style=\"border:2px solid #0f0\"> 无 误 </span></center><ul><li>NaturalGas.org : <a href=\"http://t.cn/RwYIzWY\" target=\"_blank\" rel=\"noopener\">Industry and Market Structure</a><blockquote><p>Producers - There are <strong>over 6,300 producers of natural gas</strong> in the United States.<br>Pipelines - There are <strong>about 160 pipeline companies</strong> in the United States, operating over 300,000 miles of pipe.</p></blockquote></li><li><a href=\"http://t.cn/RwQwKMM\" target=\"_blank\" rel=\"noopener\">从《穹顶之下》的油气数据说起</a> (三)第四点<blockquote><p>目前,中石油已经形成横跨东西、纵贯南北,连接四大气区和主要消费市场,覆盖全国的天然气管网体系,<strong>已投运的油气管网占国内长输管道的80%左右</strong>,其中天然气管网里程超过4.7万公里,覆盖30个省、市、自治区、特别行政区,惠及5亿人口,<strong>天然气销量占国内市场份额的70%左右</strong>。</p></blockquote></li><li><a href=\"http://t.cn/RwRHs10\" target=\"_blank\" rel=\"noopener\">中国石油天然气集团公司2013年度报告</a><blockquote><p>(第23页)公司国内天然气产量占全国<strong>比重75%</strong><br>(第27页)天然气管道总里程占全国<strong>比重80%</strong></p></blockquote></li></ul><hr><p>视频时间:01:32:38</p><p>旁白引用:</p><blockquote><p>“中国有百分之一的环境争议事件进入了司法程序。”</p></blockquote><center style=\"font-size:36px;font-style:bold\"><span style=\"border:2px solid #000\"> 存 疑 </span></center><ul><li>未找到相关数据支持</li></ul><hr><p>视频时间:01:35:58</p><p>旁白引用:</p><blockquote><p>“后来我知道,在北京餐饮带来的PM2.5的污染占到至少百分之六。”</p></blockquote><center style=\"font-size:36px;font-style:bold;color:red\"><span style=\"border:2px solid red\"> 有 误 </span></center><ul><li>经济参考报:<a href=\"http://t.cn/zR4fqwx\" target=\"_blank\" rel=\"noopener\">餐饮油烟成城市PM2.5罪魁 治理遇诸多薄弱环节</a><blockquote><p>王跃思告诉《经济参考报》记者,总的来说,餐饮油烟<strong>在北京市大气中PM2.5中的比例约为13%,最高达15%</strong>,在<strong>京津冀地区所占的比例约为6%</strong>。</p></blockquote></li></ul><hr>"},{"title":"从“拿不到特奖”说开去","path":"2015/07/27/prize_otherword/","permalink":"http://sx349.github.io/2015/07/27/prize_otherword/","content":"<p>时隔八个月之后,对<a href=\"/2015/07/19/prize/\">《我知道我拿不到特奖》</a>这篇文章做一点小小的说明。</p><p>这篇文章得到了许多人的评论,嘲讽得挺少,倒有些出乎我的意料。结合最近几位学弟学妹的公众号文章,我觉得不妨从反面入手,也就是说,当我说“我知道我拿不到特奖”的时候,我不希望读者所理解的意思。</p><hr><p>首先,我最不希望给大家的一种观感,就是“怂”。有一位特奖大神评论说:“老实说,我觉得你很weak,很怂,胸怀太小。做点小破事就一味希望得到别人认可,不在主流价值观的认可里就开始吐槽。而且你的自我认同感也不够强。”</p><p>我在想,这位“大神”的自我认同感是不是过强了一点,不仅完全没读懂我的意思,还在这里瞎掺和。与他说的完全背道而驰的是,当我说“我知道我拿不到特奖”的时候,恰恰是我自我认同感最强的时候,恰恰是我最不追求得到别人认可的时候,恰恰是我意识到,主流价值观所想要的和特奖所评价的并不完全一致的时候,恰恰是我将胸怀放到无限大的时候。</p><p>这里面最重要的一点就在于,什么是主流价值观?这位“大神”还不忘说一句“自述嘛,2000字,当然是要focus咯,我总不能乱吐”,看来他心里很清楚,平时的兴趣爱好、志愿公益、社团活动,凡是没有列在特奖的评分表里的东西,都是“乱吐”,都不是“主流价值观”认可的东西。</p><p>但这真的是主流价值观吗?大学的目的,最终是为了什么?在我看来,是为了培养改变世界的人。在未来要改变世界,和今天的成绩、社工、科研有没有关系?有。但这个关系是线性而排外的吗?不是。所以“用特奖所考察的东西衡量大学对学生的培养”这种说法,和遭到工科院系普遍吐槽的“用毕业工资衡量社会贡献”又有什么区别呢?</p><p>如果你觉得上述逻辑看得有些眼晕,那不妨作个相近的类比。我在大二的时候,开始断断续续地去健身房,在大三上的时候才慢慢养成了算是比较定期的习惯。这时候,如果我说一句“我知道我拿不到马杯健美冠军”,奇怪嘛?大家一定会说,这不是废话吗?</p><p>然而这时一个肌肉壮硕的男子走过来,一看来头,嚯,正是今年马杯的健美冠军(并非实指,见谅),对我说:你前两年吃胖了竟然就不想拿健美冠军了(“不在主流价值观的认可里就开始吐槽”),而且你才锻炼了几天就自暴自弃(“你的自我认同感也不够强”),你真怂。这不莫名其妙吗?我的基础摆在这里,我明摆着不可能一天练成个瘦子,那我明确自己的目标,告诉自己,别老眼红那帮拿健美冠军的,你自个儿就这点斤两,做点自己能做到的事情,有何不可?</p><hr><p>其次,我也不希望大家觉得,当我说“拿不到特奖”的时候,就是要做一个普通人。</p><p>说实话,我自觉也算是一个比较高调的人,换句话说,就是我从不把自己当普通人看待。我哪里不普通呢?接触新事物上手比别人快一些,做某些枯燥的事情比别人能多坚持一些,寻找解决方法的能力比别人强一些,大致就是这些。但是因此我接过院刊编辑的时候马上就能把Indesign玩出花来,因此我马上就要完成Project Euler到目前为止522题的全部翻译,因此我在某位高中同学不幸过世之后,靠翻贴吧的聊天记录加搜索引擎,把一个说了大不敬的话的家伙给揪了出来。</p><p>其实,想要做一个普通人,才是更加困难的事吧?我想了想我认识的人,除了实在是点头之交,只要我深入了解一些,终究都有些奇葩的特长爱好,或是做过什么惊艳绝伦的事儿。我一直记得一个这样的例子:做一名顶尖的程序员很难,做一名顶尖的研究海豚的海洋学家也很难,但是如果你恰好是一个不错的程序员,又是一名有所涉猎的研究海豚的海洋学家,那么也许你能做出一个描绘海豚迁徙实时路线的网站,而在这个领域,你基本上就是最顶尖的人了。</p><p>这样的事情,我自己也有点相似的经历:我是一个不入流的程序员,高中学过一点编程,搞过一点竞赛;同时,在政治哲学这方面,我大概也只是略通皮毛,看过几本书听过几堂课,大概知道一些如雷贯耳的名字。但是这两个一结合,我就鼓捣出一个小程序,叫GOD KILLS,大体上来说,就是上帝打算杀N个人,这N个人各有特点,但是上帝不能乱杀,必须根据当天的道德准则,比如如果今天是康德的绝对律令,那就绝对不能杀人,如果今天是功利主义,那么就要杀那个降低社会福利的人,诸如此类。当然,这种“小破事”真是没什么好让别人认可的,只不过是寓教于乐,大家开心就好。</p><p>归根结底,我还是希望各位能努力做一个不普通的人,这又要回到我之前所说的,我们是为了改变世界而来,是为了改变世界而接受教育,是为了改变世界而培养自己。一名顶尖的程序员能够改变世界,一名顶尖的海洋学家能改变世界,一个不错的程序员兼海洋学家,说不定你能拯救一个物种,这种改变世界的方式可是比创建一个牛逼的互联网公司听起来帅气多了。</p><hr><p>第三,我希望大家不要把这篇文章看作不努力的挡箭牌。</p><p>在这篇文章中,我从来没有说过,什么事情能够靠不努力取得。拿特奖要努力,做一个不普通的人也要努力,想要躺着改变世界,最终大概只有被世界改变的份儿。</p><p>当时有同学问我,怎么像我一样靠纯手打来做课堂笔记啊?我想,同龄人中达到我的打字速度水平的,绝算不上是屈指可数。但是,愿意像我一样这么做的人,有几个呢?你得每节课都要去听课,这大概已经难倒了一大批视通识教育课为鸡肋的同学;然后你要找一个不算太偏远的位置方便听讲,然后不能迟到早退中途缺席或开小差;最后,你还得有耐心回去之后再把整个稿子重新整理润色,花费的时间绝对不比上课少。</p><p>总的来说,这篇文章说的是效用函数,效用函数直接决定的是Objective Function,也就是规划问题所需要最优化的目标函数,而这和努力与否是不相干的。认清目标是第一步,下一步是努力,再下一步是努力,再再下一步还是努力,没有其它。</p><hr><p>我觉得吧,拿特奖这个事情,有时候真的得提前规划,当然,这未必是刻意地以拿特奖为目标的,但起码我觉得要“明智”地选课,“明智”地退课,这两条我就差得远了。反过来,我之前的大学规划,好像压根就没有冲着这个来,而且随着一步步地过下去,就越发地感觉到,在大学,尤其是清华这样勉强挤进一流的大学,你能获得的机会绝对比你能够好好把握的机会来得多。能拿特奖,一方面需要抓住一路上的机会,另一方面,特奖也未尝不是这些机遇中的一种而已。</p><p>学过经济学的同学,都应该知道,理性人要避免受到沉没成本的影响。所以,无论大学是过去一年、两年还是三年,都应该更多地放眼未来,从现在开始认清自己。我想,这才是我为什么要告诉自己“拿不到特奖”,还要特地把这些陈年烂谷子重新翻出来发一通感慨的意义。</p>"},{"title":"第五讲:Kidney Exchange 肾脏交换(二)","path":"2015/07/03/ECON11/","permalink":"http://sx349.github.io/2015/07/03/ECON11/","content":"<p>三、AER(2007)<br>既然肾脏交换的轮换大小有了限定,那么这个限定到底是多少就是一门学问了,大了太复杂,小了又会有很多的轮换不能进行,所以AER(2007)这篇文章就考察了,如果我们允许轮换中的人数最多是2人、3人或4人,分别能够解决多少患者呢?</p><p>首先,我们需要进一步剖析这个肾脏交换市场。上面我们已经提到,所谓的肾脏不适合,主要就是血型不适合和组织不适合,由于总共有四种血型(这里就先不考虑RH阴性的问题啦),“患者-捐献者”的组合一共就有16种。把这16种组合分成四组来考虑:<br>I. 血型一致: A-A B-B O-O AB-AB<br>II. 只看血型,捐献者可以给患者献血的:A-O B-O AB-O AB-A AB-B<br>III. 只看血型,患者可以给捐献者献血的:O-A O-B O-AB A-AB B-AB<br>IV. 相互都不能献血的:A-B B-A</p><p>前两组是血型适合的,这样的“患者-捐献者”组合进入市场,只能是因为组织不适合,而后两组则是血型不适合的。我们对比一下第二组和第三组,显然A-O和O-A从概率上来说应该是同一数量级的,但是由于A-O只有组织不适合的进入了市场,所以市场上的A-O组合一定远远少于O-A组合;其它的对应组合也是同理。</p><p>现在我们来做一些方便我们继续下一步的假设:</p><p>假设0:所有市场上的“患者-捐献者”组合,捐献者都不能把肾脏直接给患者,或是由于血型不适合,或是由于组织不适合。<br>之所以是假设0,因为这是市场的隐含前提,所以这篇论文并没有把这个列在自己的假设当中……</p><p>假设1:(大市场假设)任意一个患者和除了自己的捐献者之外的捐献者都没有组织不适合的问题。<br>因为市场足够大,我们可以假定每个患者总能在任意一个血型中找到充分多个和自己组织适合的捐献者,写成假设1这样只是进一步的简化了而已。</p><p>假设2:(长期假设)对于第三组的任意一个组合,经过配对之后一定都会剩下至少一对。<br>很容易验证,第三组的“患者-捐献者”组合想要进入轮换,一定会消耗掉一个第二组的组合,而第二组的组合本身是血型适合的,只有组织不适合的时候才会进入市场,因此第二组的数量远比第三组的数量要少,长期来看,最后第三组的组合一定都会剩下来。</p><p>假设3:统计数据显示市场上A-B的组合数比B-A的组合数更多。<br>这就是统计数据,没说的……</p><p>好了,接下来我们要做的就是血型连连看:</p><p>1) 当最多只能进行2个组合对换时,显然的,每个第二组的组合可以解救一个第三组的组合,最后第二组全部用完,剩下第三组的。</p><p>对于任意第一组的组合X-X,如果和Y-Z进行对换,则从血型上来说,Z能给X输血,X能给Y输血,故Z能给Y输血,即Y-Z是血型适合的,要么是第一组内部解决,要么Y-Z是第二组的。如果是前者,每次匹配掉两个X-X;如果是后者,X-X每消耗一个第二组的组合,相应的第三组就少一个组合被解决,这是一换一,没有意义。所以第一组的每个组合都选择内部两两配对,最后有落单的就落单。</p><p>对于任意第三组的组合X-Y,如果和Z-W进行对换,W能给X输血,X能给Y输血,Y能够Z输血,所以W能给Z输血,但显然W和Z是不同的(否则X和Y就相同了),所以能解决第三组的一定只有第二组,在消耗完第二组的之后,第三组剩下的都不能解决。</p><p>对于任意第四组的组合,不妨假定是A-B,要么就用B-A解决,要么就用第二组的B-O,AB-A或者AB-O解决,显然后者还是一换一,没有意义,所以所有的A-B和B-A内部组合解决掉,两个组合哪个多就剩哪个,根据假设3,A-B就会剩下来。</p><p>所以当最多只能进行2个组合对换时,解决掉的组合数是:<br>N2=2*N(II)+2*N(B-A)+2*([N(A-A)/2]+[N(B-B)/2]+[N(O-O)/2]+[N(AB-AB)/2])<br>后面括号里的是下取整。</p><p>简单的解释一下,每个第二组的组合解决了两个组合,A-B和B-A中较少的那个(B-A)每次解决了两个组合,而第一组每种组合都内部解决了所有成对的。</p><p>2) 当最多只能进行3个组合轮换时,我们来看看,多出了哪些新选择:<br>首先,所有的X-X组合可以被塞进任意的对换中去了,所以X-X不会再有落单的组合了。<br>其次,AB-O这个组合可以通过和O-A A-AB或者O-B B-AB组成轮换,一次性解救两个第三组组合<br>第三,多出来的A-B组合,可以构成A-B B-O O-A或者A-B B-AB AB-A这样的轮换,从而被解救。</p><p>因此,此时能够解决的组合数变为:<br>N3=2*N(II)+2*N(B-A)+N(I)+N(AB-O)+min{N(A-B)-N(B-A),N(B-O)+N(AB-A)}</p><p>还是解释一下,首先每个第二组的组合至少解决了两个组合,但AB-O还能多解决一个第三组的组合;B-A每次解决两个组合没有变;第一组所有的组合都解决了,不用考虑落单了;最后,多出来的A-B可以加塞到B-O和AB-A的队伍中,就看是加塞的和被加塞的哪个更少一点了。</p><p>做个减法,我们有<br>N3-N2=k+N(AB-O)+min{N(A-B)-N(B-A),N(B-O)+N(AB-A)}<br>这里面的k就是第一组中落单的数量,最小是0最大是4。</p><p>3) 当最多只能进行4个组合轮换时,新增加的选择是,A-B可以构成A-B B-AB AB-O O-A这样的轮换,所以A-B可以靠加塞进AB-O来解救了,其它所有组合都没有新增成员。<br>N4=2*N(II)+2*N(B-A)+N(I)+N(AB-O)+min{N(A-B)-N(B-A),N(B-O)+N(AB-A)+N(AB-O)}<br>所以<br>N4-N3<=N(AB-O)</p><p>可以看出,N4-N3显然小于N3-N2,从成本收益的角度考虑,选择最多允许3个组合进行轮换应当是最佳选择。模拟的结果也是如此,人口总数为100的模拟中,只能两人对换时,解救的患者期望为49.708人,允许三人轮换上升到59.714人,允许四人轮换上升到60.354人,而不加限定的结果是60.39人,从两人到三人提高了10个百分点,而三人到四人只提高了0.6个百分点。</p><p>这里留一个小问题,为什么我们只考虑到四人就结束了?是因为我们发现三人到四人的提升空间已经很小了,后面越来越小不用考虑?还是因为之前说的现实情况限定了最多只能同时进行八台手术进行私人兑换?抑或是作者其实先做了模拟实验发现四人轮换的结果已经和没有限定很接近了所以就只讨论到四人?这个问题我们留到下一讲介绍肝脏交换市场的时候一并来解决。</p><hr><p>(2015.07.03 @ 微信公众号 @ 《第五讲:Kidney Exchange 肾脏交换(二)》)</p><hr>"},{"title":"第五讲:Kidney Exchange 肾脏交换(一)","path":"2015/06/30/ECON10/","permalink":"http://sx349.github.io/2015/06/30/ECON10/","content":"<p>前面的例子可能还是比较生活化的,这一讲我们将要进入的领域,肾脏交换,恐怕是大家都避之不及的内容。不过,这个领域反而是匹配理论目前最前沿的领域,接下来要提到的目前该领域的三尊大神Roth,Sönmez和Ünver,大家可以去Google Scholar上看他们最近的论文,主要都集中在这个方面。而这三尊大神,也正是这个领域最早的开拓者。</p><p>RSU三人组在肾脏交换这个领域的三篇论文,分别用了三个不同的模型,从三个不同的角度来解决这个问题,简直就是“腰花三吃”(好像不能凉拌,那除了爆炒之外还有两吃是什么……),这三篇论文分别发表在QJE(2004),JET(2005)和AER(2007),下面就用这个作为论文代号来给大家一一道来。</p><p>一、背景知识<br>首先我们来了解一下,什么叫肾脏交换,感觉平常说的都是“肾脏移植”啊,一个人的肾脏坏了,需要别人的肾脏,那又怎么“交换”呢?其实是这样的,如果某位患者肾脏需要移植,而他找到了一个亲戚或朋友,恰好可以把肾脏移植给他,那这样的人是不会进入市场的,他自己可以解决。但如果他找不到这样的亲戚和朋友,那他就需要从市场上找其他人把肾脏移植给他,这样的等待是漫长的,来看一组数据,美国每年有上十万人需要接受肾脏移植(这个数目是在不断累积的),平均每天有12人因为未能得到合适的肾脏而去世。</p><p>但是也有一种取巧的方式,比如有两个患者A和B,患者A的父亲愿意把自己的肾捐给A,但是检验发现不适合,同理患者B的哥哥也愿意把肾捐给B,同样不适合,这时候医院发现,A的父亲的肾可以给B,而B的哥哥的肾又可以给A,双方皆大欢喜,这种方式就称为“交换”。当然这种“交换”不限于两个人,在更多的人之间进行轮换也是可以的。</p><p>针对这个情况,美国建立了专门针对肾脏的清算所(Clearing House),统一调控患者和捐献者。在所有模型中,我们都假设清算所建立了一个肾脏交换的市场,在这个市场上,每个患者都对应了一个捐献者(类似于房屋市场上的禀赋),但是这两人之间是无法进行肾脏移植的,必须要和其他患者构成轮换才能进行移植。</p><p>关于所谓肾脏的不适合,医学上大概分为两个方面,其一是血型(Blood Type)不适合,规则和献血一致,O型血的肾脏可以移植给所有人,A型血只能给A和AB,B型血只能给B和AB,AB型血只能给AB型血,我们可以用一个Lattice表示这样的关系(还记得Lattice吗?在第一讲中提到的哦);其二是组织(Tissue Type)不适合,这个就比较复杂了,但我们首先做个假定,就是这两种不适合之间是独立的。</p><p>二、QJE(2004)<br>这篇论文的模型是三篇中最简单的,其实就是把肾脏交换的市场看成了一个房屋市场,假定每个患者对于市场上的所有捐献者进行排序,显然能够移植的都被排在患者禀赋的捐献者之前,然后走一遍TTC算法的流程。在上一讲已经提到了,TTC算法每一轮都能构成一个圈,我们只需要在这个圈内进行轮换的移植就可以了。</p><p>TTC满足strategy-proof,IR和帕累托最优,这是上一讲最后我们提到的定理,在肾脏交换市场上,strategy-proof是没有意义的,因为对患者来说适合不适合,或者有多适合,很多时候不是他自己说了算的;但IR和帕累托最优是有意义的,IR保证了患者不会得到一个无法移植的肾脏,最差的结果就是无法进行移植,而帕累托最优保证了这是所有情况下的最优结果,患者已经无法得到更好的肾脏了。</p><p>感觉好像这个问题解决了的样子啊,为什么还要大费周章的再去搞两个模型呢?问题出在肾脏交换潜藏的一个道德风险。</p><p>还是拿上面这个例子来看,比如患者A和B拍板决定要换肾了,先进行了两台手术,患者A的父亲把肾给了患者B,再要进行后两台手术的时候,患者B的哥哥突然表示,我不打算捐献肾脏了。一般来说,如果捐献者表示不愿意捐献了,即使之前签订了如果你捐了我就捐这样的协议,医院照样不能强制执行这样的协议,把患者B的哥哥的肾扒下来塞给患者A。但是这样的话,患者A就不乐意了,我爹的肾都给你了,你哥的肾却不给我,这他喵的不是玩我嘛!</p><p>为了避免这样的情况(据说以前还是经常发生的汗),那就只有一个办法,同时进行4台手术,这边换过来,那边换过去,谁也不吃亏。不过说是这么说,咱们考虑的只是最简单的对换,如果是三个人轮换,就要同时进行6台手术,四个人就是八台,且不说能动这种手术的医生有几个,能同时进行这么多手术的医院也很少见,往往需要好几个医院同时动手。但TTC算法在进行的过程中才不管你到底圈有多大,如果一不当心一个圈里有十几个人,那估计一个市甚至一个县(county,或者翻译叫郡,比市大比州小的行政区划)的所有医院都得帮忙才行。</p><p>所以,在实践的过程中,对于肾脏交换的这个轮换的大小,还是需要做一些限定的,比如说最多不超过同时四台,那么就只有两个患者进行对换的肾脏交换可以进行,一般来说允许六台或是八台已经是极限了。</p><p>还有一个问题是,这个模型中需要患者进行排序,这个要求实在太复杂了,很多时候患者根本不在乎这个肾是不是最好的,只要能换,先活下来再说,所以他们对于捐献者的偏好其实是二分的,要么可以换,要么不能换,可以换的里面就不区分好坏了。</p><hr><p>(2015.06.30 @ 微信公众号 @ 《第五讲:Kidney Exchange 肾脏交换(一)》)</p><hr>"},{"title":"读书札记·六月","path":"2015/06/30/notes1506/","permalink":"http://sx349.github.io/2015/06/30/notes1506/","content":"<p>书名《所罗门王的指环》 日期 2015.6.2</p><p>动物行为学的科普作品,类似于法布尔的《昆虫记》。作者对于动物的观察非常细致,而且生动有趣,尤其是许多动物展现出的类似于人类的行为,就好像看到童话故事在现实中真的出现了一样,例如作者饲养的穴乌、雁鹅等等;但同时,作为一名科学家,作者对于动物的态度也不同于一般的饲养者,并不完全是将自己的意念加诸在这些生灵之上。作者在二战期间为德军充当战地医生随后被捕直到战后才回归故里的经历,一方面造成了他所豢养的动物原理饲主的空白期,给他的动物观察创造了新的情境,另一方面也使得作者对于人与人之间的争斗有了自己的观察,并与动物作了对比。此书并不完全是一本科学著作,无论从描述上还是语言上,都非常适合不了解这一领域的人阅读,学习一些基本的动物学和动物行为学的有趣知识。</p><p>书名《逻辑学是什么》 日期 2015.6.5</p><p>介绍了逻辑学的历史和发展,从中西方古代的名学和诡辩家,到几种基本的逻辑形式,再到现代逻辑主义、形式主义和直觉主义的发展,可以作为逻辑学的入门书籍。总的来说,作者持有的观点是,逻辑学最终是为日常生活服务的,所以对数理逻辑略带有批判的态度,但在我看来,这种批判是毫无道理的,就好像因为生活中我们用的都是数而不是字母,所以转去批判代数的发明一样。从算术到代数的过程,或是从SAP/SIP这样的形式转向数理逻辑的过程,使得整个体系更为严谨,简化了许多复杂且没有意义的问题(如周延不周延,换质换位等等);我读金岳霖先生的《逻辑》一书时,读得头昏脑胀,无怪乎汪曾祺先生曾经说班上只有王浩一人能够跟上,但是最后一节讲数理逻辑的内容,却是直接豁然开朗,我想即便是汪曾祺先生恐怕也不会觉得太过困难。王浩后来的成就,也是建立在数理逻辑的基础上,这是不无道理的。</p><p>书名《我们要活得有尊严》日期 2015.6.5</p><p>和柏杨其它的书相比差得多,不推荐。还是看《丑陋的中国人》去吧,《鬼话中国正史》也不错,尚未读过柏杨版资治通鉴,可能比较有趣,待读。</p><p>书名《双城记》 日期 2015.6.8</p><p>狄更斯的经典著作,在书中将巴黎和伦敦做对比,实际上却是预言伦敦将遭遇巴黎式的底层民众推翻贵族的同样命运。我认为,放在今天来看,最有意义的地方莫过于对巴黎所谓民众革命的否定,就像开头说的,we were all going direct to Heaven, we were all going direct the other way。近期恰好在知乎上看到有人评价《湖南农民运动考察报告》,虽然有评论说那篇回答有断章取义之嫌,但是我看了《报告》原文,这种非黑即白式的“革命”,即使“胜利”了,也不是中国人民所盼望的胜利。回答中有一句很中肯,就是阶级斗争不是将阶级对立起来(当然我认为这是退一万步地说法,即首先得接受以阶级斗争为纲的理论);马克思恩格斯自己也不是无产阶级,不分好坏不分敌我扣帽子的革命必然会造成严重的后果。</p><p>书名《第22条军规》 日期 2015.6.14</p><p>这本书大致是第三或第四遍读了,其结尾的约赛连的逃亡,却免不了给人一种娜拉的出走的感觉。本书中最大的冲突,莫过于个人利益与国家利益的冲突,到底是“留得青山在不愁没柴烧”,还是“粉骨碎身浑不怕要留清白在人间”?国家利益到底是什么,是作为一个国家的利益,还是国家作为一个个人的集合体,而获得的利益?从作者的处理来看,我想他所推崇的无疑是后者:比如,米洛的M&M辛迪加,又比如布莱克上尉的签字效忠运动,这些都是一个集体打着“为了集体利益”的名号干出来的非常荒唐的事情。但是,作者并没有完全打消自己的疑惑,意大利那个老头的话难道就不荒谬吗?当个人追求个人利益而放弃国家,觉得总有一天国家会自己恢复过来,难道就是可以的嘛?</p><p>书名《万历十五年》 日期 2015.6.15</p><p>这本书是第二遍读,黄仁宇的史观经常被称为“大历史”,大概是因为他有一种独特的见微知著的视角。这本书的英文标题就很说明问题,”1587, a non-significant year”。一个值得注意的细节是,黄仁宇提到,这本书中的每个人物,无论是万历皇帝朱翊钧,张居正,申时行,戚继光,海瑞,这些人最终都是悲剧人物,而他们的悲剧命运是由于这个皇朝的行将就木所造成的,而这个皇朝在此时的倾颓,又是因为在洪武皇帝时所定下的文官治国的理念导致的。但文官治国的理念,又是朱元璋借鉴唐朝藩镇割据而得出的结果。这里面,是不是有一种必然呢?如果是的话,这种必然又到了什么程度呢?它是由于明朝的某些特征,比如是继一个少数民族强大政权的覆灭而兴起的,所造成的必然,还是由于民族的某些特征所造成的必然呢?我大概还是倾向于后者吧。</p><p>书名《数据挖掘导论》 日期 2015.6.15</p><p>因为助研需要所看的一本书,偏计算机科学方向,技术性很强,不推荐没有需要的同学去看。不过,确实起到了拓宽视野的作用,如果有一天人类不再受限于计算能力(比如量子计算机),那么从已有数据中寻找模式还是比较可行的,比如前两年出现的剪刀石头布高胜率程序,就是根据上百万次的真是猜拳过程,发掘人类在猜拳时“随机”出招背后的规律。现在的人工智能仰仗于人来编写对答程序,一些非监督学习的算法尚处于初步阶段,但未来一定会更加重要吧。</p><p>书名《城市迷宫》 日期 2015.6.18</p><p>联合国教科文组织《信使》杂志的汇编,介绍了一系列有关城市的话题:人口、建筑、环境、设计理念、文化交融等等。由于《信使》杂志在2002年即停办,所以许多话题都还是13年前的样子,比如对人口的预测以及特大城市的预测,现在的发展状况早已超出了上世纪末(1999年)的预期;与此同时伴随而来的,就是城市的问题也远远超过了我们的预期。当时预计的特大城市,全球范围内有36个,而如今,也许仅仅中国就已经超过了这一数目,想起柴静在《穹顶之下》中提到的那个数字,全国有一百多个城市要建设成国际化大都市,不禁感叹,在城市日益扩大成为迷宫的同时,人们对于这种假大空的追求同样已经陷入迷宫不能自拔。</p><p>书名《生命是什么》 日期 2015.6.19</p><p>著名的物理学家薛定谔的著作,由于当时的科学研究所限,其中有些内容在今日看来已然并不正确,特为尤甚的是所谓“基因与量子跃迁”的关系。然而,不可否认的是,其中一些思想确实成为了后人的指路明灯,例如遗传密码的存在,又比如耗散结构(当然薛定谔还是只看到了负熵,不过这一想法已经很好地解决了拉普拉斯妖的悖论),我写的短文《未及柳絮因风起》隐约有了一点这个想法的影子,虽然当时我并没有意识到。看到有人评论此书时说,威尔金斯和克里克都是因为受到此书的影响而从物理学转入生物学,更觉得热血澎湃;也许,这是一本外行写的内行书所能达到的最高境界了。</p><p>书名《锯齿啮痕录》 日期 2015.6.19</p><p>伤痕文学的代表作。只要我想起仅仅五十年前,在我们这样一个庞大的国家,曾经遍地都是这样的人,这样的事,而且今人还为之涂涂抹抹遮遮掩掩,不能不令人齿冷。所谓威权主义莫过于是吧?所谓个人崇拜莫过于是吧?呜呼,在那个年代,那个燃烧的金字塔,下面烧死了多少草民。怪草民乎?怪中间不愿自己被烧而向下烧草民的人乎?还是应该怪金字塔尖上那个放火的人乎?流沙河君可说了,“五一六”的时候,可是没有红卫兵的呀。</p><p>书名《人间失格》 日期 2015.6.29</p><p>太宰治的作品,很多时候都直接地把假、恶、丑,把肉欲、贪欲、自私、幻想都泼在纸上,然后告诉你:看!这就是我!这就是你!这就是现实!轻小说写美女喜欢帅哥,那跟大部分人是无缘的;《巴黎圣母院》写美女喜欢丑八怪,那在现实中是绝无仅有的;而他的作品——渣男,花心,搞外遇,缺钱,人前一套人后一套,但却又散发着女人无法抗拒的魅力——有谁敢说,这不是现实呢?再比如,明明可以直接告诉他:给我去上学,有人替你付学费;却偏偏要说:学费啊,确实是贵的,然而最重要的是你的心意,你要诚心——什么,你的心意是要去学画画?你一定是不诚心的,再去想想。相比之下,“——你听我解释……——我不听我不听!”这样的对白简直是幼稚粗鄙到了极点,让人觉得作者把读者当猴耍——然而毕竟还是有人吃这一套。</p><hr><p>(2015.06.30 @ 微信公众号 @ 《读书札记·六月》)</p><hr>"},{"title":"第四讲:Housing Market 房屋市场(下)","path":"2015/06/26/ECON09/","permalink":"http://sx349.github.io/2015/06/26/ECON09/","content":"<p>现在我们回归一下现实,由于我的禀赋只有房子没有钱,我只能靠卖了自己的房子得到的钱去买新房子,所以新房子价格最多和我持有的房子一样贵;另一方面,在所有我买的起的房子中,新房子必须是我最喜欢的房子,否则我完全可以去买更喜欢的房子。假定我们找到了核中的一个匹配,然后又找到了一个对应的价格安排(这个价格安排也叫做Competitive Price)满足上述要求,这就称为Walrasian Equilibrium(瓦尔拉斯均衡……?感觉很少看见这个译名的说……),学过公财的同学可能在福利经济学第一定理中看到过这个名字,这个均衡的经济学意义是,在一定的价格安排下,每个人最大化自己福利的决策恰好使市场出清。Shapley & Scarf在1974年的论文中证明了,每个房屋市场都存在Walrasian Equilibrium,其中的匹配就是上面提到的TTC给出的匹配,而价格是如下决定的:第一轮的所有房子价格定为N-1,第二轮定为N-2,依此类推。</p><p>自然而然的一个问题就是,这样的均衡是唯一的吗?从价格的角度讲,实际上只要每一轮的房子价格比后一轮更贵就可以了,甚至如果在一轮中存在多个Top Trading Cycle,因而分成了几轮才都安排完,这些Cycle中的房子其实都可以定为相同的价格。但是说来说去,我们讨论的都是TTC给出的这个匹配,从匹配的角度来说,Walrasian Equilibrium是唯一的吗?</p><p>其实我这么问的时候大家应该已经猜出答案了,TTC其实是核中唯一的匹配,这个定理是由Roth和Postlewaite给出的。显然这个定理的一半已经证完了,TTC肯定是在核中的;要证的另一半就是,在核中的匹配一定是TTC,这个也很容易,如果核中还有不同于TTC的某个匹配,我们称它为新匹配,然后我们用核的定义来验证一下:首先,我们把在TTC中第一轮拿到房子的人拉出来组成一个联盟,因为他们在TTC中拿到的都是他们最喜欢的房子,如果新匹配不能给他们TTC中的房子,那么这个联盟的人总能通过互相交换禀赋的房子达到TTC然后better off,矛盾出现了,好吧,那么新匹配给这帮人的房子就得跟TTC一样;然后把这些人和他们的房子拿走,把TTC第二轮拿到房子的人拉出来,新匹配又得给他们和TTC一样的房子;然后再拿走,第三轮……依此类推,最后我们发现,新匹配每一轮给的房子都跟TTC一样,说好的不同于TTC的匹配呢!矛盾,好吧,核中除了TTC还是TTC,不可能有别的匹配了。</p><p>没想到TTC这个简单粗暴的算法竟然是这个问题的唯一解吧(此处应有“我伙惊”……),但是这么个简单粗暴法很难不让人想到我们的老朋友Boston Mechanism,而BM最大的问题之一就是,它不满足strategy-proof。但核中的独苗匹配TTC显然是身经百战,见得多了,所以TTC竟然满足strategy-proof也没有那么令人惊讶了吧……</p><p>再结合之前关于TTC是核中唯一匹配的结论,Ma(1994)”Strategy-Proofness and the Strict Core in a Market with Indivisibilities “给出了定理,即满足IR、帕累托最优和strategy-proof的匹配只有TTC给出的匹配。不过我感觉其实这只是个填坑的工作,因为TTC给出的匹配是唯一满足IR和帕累托最优的匹配之前已经证明过了,而TTC匹配满足strategy-proof是1982年Roth论文的结论,那我们已经知道满足IR和帕累托最优的只有TTC了,而TTC又满足strategy-proof,这个定理难道不是显然的嘛……</p><p>Roth的证明还是很赞的,我们来扒一扒,相关论文是Roth(1982)”Incentive compatibility in a market with indivisible goods”。</p><p>假定有一个人i打算谎报自己对房子的偏好,而且在他没有谎报的时候,他在第k轮拿到房子x,在他谎报的时候,他在第k’轮拿到房子x’。<br>如果k比k’要小或者k等于k’,那么在第k轮开始之前,x和x’都还在没拿走的房子里,这时候i最喜欢的房子就是x,谎报拿到的房子x’一不会比x更好(可以就是x),这时候他没有谎报的激励。<br>如果k比k’要大,那么在第k’轮的时候,说了谎的i一定处在某个Top Trading Cycle当中,他拿的房子x’原本的主人是j0,然后j0拿了j1,j1拿了j2,直到最后jn拿了i自己禀赋的房子。<br>那么如果i没有说谎,在进行到第k’轮的时候,这个Top Trading Cycle并不存在,只缺了最后一环也就是i拿了j0的房子。而在从第k’轮到第k-1轮的过程当中,j0有可能处在某个Top Trading Cycle当中,从而把房子x’给了别人吗?不可能,我们可以倒过来看,因为i还在市场上,所以jn最喜欢的房子一直在市场上,他肯定始终指着i,所以他肯定没有出现在某个Cycle里,否则i就会进入这个Cycle,从而在第k轮前就拿到房子,这和我们的假定矛盾;同理,jn-1也始终指着jn,所以他也没有出现在某个Cycle里,否则就又把i带进这个Cycle了,然后倒推回j0,他在第k轮之前,一定没有进入任何一个Cycle。<br>这意味着,如果i说谎了,他拿到了x’这套房子,而如果i没有说谎,在到了第k轮的时候,x’这套房子(以及它的主人j0)仍然留在市场上,这时候i选择了他最喜欢的房子x并得到了这套房子,显然x至少和x’一样好,所以i还是没有激励谎报。<br>所以无论如何,i都没有谎报的激励,也就证明了TTC是strategy-proof的。</p><hr><p>(2015.06.26 @ 微信公众号 @ 《第四讲:Housing Market 房屋市场(下)》)</p><hr>"},{"title":"第四讲:Housing Market 房屋市场(上)","path":"2015/06/23/ECON08/","permalink":"http://sx349.github.io/2015/06/23/ECON08/","content":"<p>在第二讲的时候提到了所谓的单边市场,也就是一边是人,一边是物的市场。这样的市场,最简单的例子莫过于房屋交易市场。</p><p>在这个市场上,每个进入市场的人都有自己的禀赋(endowment),也就是一套房子(总不能让人在市场上交易不成功回去连住的地方都没有吧>_<),然后互相交易。每个人对市场上的所有房子有一个偏好,但是房子不会挑人,随便谁住都可以。</p><p>这时候,什么样的匹配是稳定的呢?因为房子不会挑人,也就等同于人对于房子来说是无差异(indifferent)的,而Pairwise Stability中的私奔需要双方都喜欢对方胜过现在的匹配对象,所以我们不能再用PS作为稳定的条件了。</p><p>但是,直观上,有一种显然的不稳定的情况:比如说现在有两套房子,一套在海边一套在山里,住在海边房子里的人喜欢爬山,住在山里房子里的人喜欢冲浪,双方对于对方的房子的评价都比自己的房子要高,假定两套房子其他方面都差不多,这时候让他们两个人交换,那么两个人都会得到更好的房子。当然,这个情况可以构造得更复杂,比如说一个城市里的三套房子,分别临近水果店、电玩城和游泳馆,结果房子的住户偏偏讨厌自己临近的这家店,水果店旁的住户喜欢打电玩,电玩城旁的住户喜欢游泳,游泳馆旁的住户喜欢吃水果,那好,三个人轮换一下,正好每个人都得到一个更好的房子。</p><p>我们把市场上的一部分人(以及他们的房子)捆绑在一起,称为联盟(Coalition)。在这个市场上的所有匹配中,存在有一种特殊的匹配,无论怎样选择联盟,这个联盟中的人都不能够通过互相交换自己的禀赋房子,使得至少一个人得到的房子比这种匹配给出的更好,这种匹配构成的集合被称为核(Core)。实际上,在双边市场上,所有的稳定匹配也构成了核:无论你选择哪些男生和女生,都不能够通过在他们内部重新匹配,使得每个人都得到更好的配偶。这些稳定匹配满足两个条件,分别是个体理性(Individual Rationality)和配对稳定(Pairwise Stability)。而在单边市场上,核中的匹配同样需要满足两个条件,分别是个体理性和帕累托最优(Pareto Efficient),证明很简单就不再赘述啦。(或者也可以留作课后习题哈哈~)</p><p>那么有没有一种方便的办法,像DA那样,只要我们给出有哪些人,有哪些房子,每个人对于房子的偏好,就能够得到一个在核中的匹配呢?答案当然是有的,而这个算法是由DA的提出者之一Gale提出的(虽然1974年提出这个算法的论文是Shapley和Scarf写的),名为Top Trading Cycle Algorithm(中文大概是叫首位交易循环算法,不过好复杂啊还是叫TTC算法好了反正你们都懂得╮(╯▽╰)╭)。</p><p>顺便扒一句,其实在匹配理论这一块,Gale和Shapley这两个数学家(主要是数学家吧……像1962年介绍DA的论文和1974年介绍TTC的论文都是发在数学期刊上的)是最有资格拿诺奖的,但是直到20世纪90年代,诺奖才慢慢开始转向博弈论的领域,而稳定匹配理论的获奖一直等到了2012年,若不是Gale老先生于2008年仙逝,他一定有机会分享这份荣誉。Roth作为将匹配理论从数学领域打入经济学领域的第一人,获奖倒也算是实至名归。Shapley获奖并不奇怪,比较奇怪的反而是他并没有因为在合作博弈方面的贡献获奖(比如Shapley Value,这个应该是合作博弈耳熟能详的名词吧)。</p><p>TTC算法的流程非常简单粗暴:第一轮,每个人都指向自己最喜欢的那套房子,如果构成了一个圈(A指向B的房子,B指向C的房子,C指向D的房子,等等等等,最后一个人指向A的房子),那么这些人就换房子,然后这些人带着房子离开市场。<br>第二轮,所有还在市场上的人,都指向所有还在市场上的房子中最喜欢的那套房子,如果构成了一个圈,同样的,换房子,离开市场。</p><p>然后第三轮,第四轮,直到所有人都离开市场。</p><p>等下,好像刚才说的是……“如果”?如果不构成一个圈呢?</p><p>当然是一定可以构成圈的啦否则我还在这里说个什么鬼……证明如下:假设没有圈,任取一个人A,A不能指向自己,一定指另一个人B;B不能指A和自己,所以只能再指另一个人C;这样进行下去,最后所有人都被指到了,而最后那个人又不能指已经指过的人,否则就会有圈,出现矛盾,假设不成立。</p><p>算法论述完毕,再来看看这个算法的结果是不是在核中。假定我们划了一批人构成了联盟,首先看这个联盟当中那些在第一轮就拿到房子的人,显然他们拿到的是最喜欢的房子,不会选择和其他人换;然后是第二轮,他们没法和第一轮的人换(因为第一轮的人不愿意跟他们换),也不会和其他人换(因为拿到的是所有第一轮的人没拿的房子里最喜欢的房子);然后是第三轮,第四轮,依此类推。结果是联盟中的人都不能通过换房子得到更好的房子。</p><hr><p>(2015.06.23 @ 微信公众号 @ 《第四讲:Housing Market 房屋市场(上)》)</p><hr>"},{"title":"第三讲:More on College Admission 择校问题补遗(三)","path":"2015/06/19/ECON07/","permalink":"http://sx349.github.io/2015/06/19/ECON07/","content":"<p>之前我们讨论的一直是学生可以操纵最后的结果,但实际上学校也仍然有一定的自主权,虽然priority是由政府决定的,但是学校可以假装自己没有那么多位置,反而能获得更好的学生。肯定会有同学表示:这怎么可能呢?多了一个位置,就好比是扩大了你的决策集,应该总是better off的呀?但是请注意,现在我们面临的机制设计问题,本质上还是博弈论问题,这是一个动态的决策,很多时候减少自己的选择是能够帮助玩家取得更好的结果的。</p><p>我们先简单地考虑一个例子,有两个学校A和B以及两个学生1和2,学校A有两个位置,学校B只有一个位置,学校A更喜欢学生2,学校B更喜欢学生1,但是学生1更喜欢学校A,学生2更喜欢学校B。</p><p>假定我们按照由学校向学生“求婚”的DA算法来进行择校过程,学校B选了学生1,学校A两个都选了,此时两个学生都会去学校A。但是学校B还没招满,于是它又去招揽学生2,学生2就乖乖地去学校B了,学校A只留下了自己的次选学生。</p><p>归根结底,学校A做错的不就是好死不死地在第一轮把学校B喜欢的学生给抢走了嘛,那么如果学校A假装自己就只要一个学生会怎么样呢?学校A在第一轮就不会要学生1了,学生1留在学校B,学生2留在学生A,对学校来说皆大欢喜(对学生来说自然是worse off咯)。</p><p>聪明的同学可能已经发现了,如果是由学生向学校“求婚”呢?这时候,不论学校A是不是假装减少一个位置,终归是学生1去学校A,学生2去学校B,这种情况我们就称为“不可通过容量操纵的(Non-manipulable via capacities)”。</p><p>首先开始更系统性研究这个问题的paper是Sönmez(1997,JET) “Manipulation via Capacities in Two-sided Matching Markets”。他告诉我们,当学校和学生(在论文中是医院和实习生,一个道理咯)至少有两所和三名时,所有的稳定匹配规则都是有可能通过容量操纵的(所谓有可能,指的是存在特定的偏好和配额,使得其中至少有一个学校减少自己的配额能够得到更好的结果)。</p><p>啊,我是不是没有讲什么叫稳定匹配规则?其实上面提到的由学校向学生“求婚”的DA算法(因为得到的是对学校(文中是医院)最优的结果,称之为HOSM,即Hospital Optimal Stable Matching)或是学生由学校“求婚”的DA算法(IOSM,即Intern Optimal Stable Matching),都属于稳定匹配规则,它们保证了对于任意的偏好R和配额q,都能给出一个稳定匹配。实际上,稳定匹配规则并不关心中间的过程(分成几轮啊,谁向谁求婚啊,拒绝或是不拒绝谁啊),而是一个黑箱,输入任意一个(R,q)的组合,给出一个匹配μ对于(R,q)都是稳定的,就算做一个稳定匹配规则。也就是说,我完全可以制定这样一个稳定匹配规则:如果学生1认为学校A是accpetable的,那么就给出HOSM的稳定匹配,否则给出IOSM的稳定匹配。</p><p>简单的情况下,比如当只有一个学校的时候,它一定会用完自己的配额,所以不会主动减少容量;当只有一个学生的时候,他会选择自己最喜欢的学校,学校减少配额是没有意义的,要么就是那个学生本来就不选你,要么就是那个学生选了你,你把名额从大于一的任意数减少到一,结果也不会变,如果再减少到零连唯一的这个学生都没有了。当有两个学校和两个学生的时候,可以证明IOSM是无法通过容量操纵的。(咦突然发现作者没有证明如果有两个学生和多于两个学校的情况……不过好像这时候显然IOSM也是无法通过容量操纵的……)</p><p>那么至少有两所学校和三名学生的时候会发生什么呢?其实只需要一个反例,这个反例是这样的:</p><p>先看恰好有两所学校和三名学生的情况,这时候的偏好和priority分别是<br>(注意这里学校的priority是用集合的形式给出的,这是正式的写法。如果写成之前的1 2 3,无法看出学校会更喜欢{1}还是{2,3})<br>PA: {1,2,3} {1,2} {1,3} {1} {2,3} {2} {3}<br>PB: {1,2,3} {2,3} {1,3} {3} {1,2} {2} {1}<br>R1: B A<br>R2: A B<br>R3: A B</p><p>当两个学校的配额容量都是2的时候,稳定匹配只有一个,那就是(A 2,3)(B 1)<br>当两个学校的配额容量都是1的时候,稳定匹配也只有一个,那就是(A 1)(B 3)<br>(同学们可以拿出纸笔自己验算一下~)<br>但如果学校A的配额容量是2,而学校B的配额容量是1的时候,稳定匹配会有两个,其中一个是(A 2,3)(B 1),另一个是(A 1,2)(B 3)</p><p>这时候让我们来构造稳定匹配规则吧,我们只需要考虑在这三种不同配额的情景下,分别给出哪一种稳定匹配就可以了,显然前两种情景已经不用选了,关键看第三种情景:<br>如果选(A 2,3)(B 1),这时候把第三种情景作为初始情况,学校A把自己的配额从2减少到1变回第二种情景,对它来说{1}比{2,3}更好。<br>如果选(A 1,2)(B 3),这时候把第一种情景作为初始情况,学校B把自己的配额从2减少到1变为第三种情景,对它来说{3}比{1}更好。<br>所以无论我们在第三种情景下怎么选,总存在一组(R,q),使得其中有一个学校减少配额能够得到更优的结果。</p><p>还没完,因为我们要证明的是“至少”两所学校和三名学生,那么我们在已经有两所学校和三名学生的基础上,加上更多的学生和更多的学校,这些新增的学生不愿意去任何一所学校,而新增的学校不愿意招收任何一名学生,显然在所有的稳定匹配中,这些学生和学校都保持“单身”,不影响结论成立。</p><p>当然,这种举反例的方式有点无赖,而且基本上不可能普遍适用。但无论如何,这篇paper开辟了一个新的领域,也说明了这个问题值得得到重视(毕竟这个结论是“几乎都能被操纵”)。后续阅读推荐Ehlers(2010) “Manipulations via Capacities Revisited”,进一步归纳和总结了一些性质和结论。</p><p>顺带提一句,如果我们抛弃稳定性而只考虑帕累托最优和防止通过容量操纵的话,其实有一种很简单粗暴的匹配规则:首先把学校编个号,然后学校1先过来挑,挑到配额全部用完或者所有acceptable的学生都挑完为止,然后学校2来挑,然后学校3……对于每个学校来说,减少配额是没有意义的,假设这个学校的配额是p,在这个机制中挑了q个学生(p>q),如果把配额减少到q,q+1,q+2,…,p-1,结果不变,还是挑q个学生,如果把配额减少到q-1,q-2,…,1,0,那连q个学生都挑不满,而前面的学校又不会因为你减少了配额给你留一点更好的学生,结果只会更糟。这个规则也有一个名字,叫作Serial Dictatorship(中文大致叫“序贯独裁”吧)。</p><hr><p>(2015.06.19 @ 微信公众号 @ 《第三讲:More on College Admission 择校问题补遗(三)》)</p><hr>"},{"title":"第三讲:More on College Admission 择校问题补遗(二)","path":"2015/06/16/ECON06/","permalink":"http://sx349.github.io/2015/06/16/ECON06/","content":"<p>在介绍波士顿机制Boston Mechanism时我们提到,学生会发现,如果自己谎报志愿,和报真实偏好相比,可能能够进入更好的学校,这说明BM不是strategy-proof的。而且,如果所有学生都意识到这一点,那么就构成了一场填报志愿的博弈,最后的均衡结果中最好的恰恰就是DA算法给出的最优稳定匹配。</p><p>但是,并不是所有的学生都意识到了这一点,如果有的学生仍然选择报上自己的真实志愿(这些学生称为”Sincere students”,或者也叫”Naive students”),而有的学生意识到自己可以谎报(这些学生称为”Sophisticated students”,或者也叫”Strategic students”),那结果会是怎么样的呢?</p><p>接下来登场的这篇论文是Pathak & Sönmez (2008, AER) “Leveling the Playing Field: Sincere and Sophisticated Players in the Boston Mechanism”(Sönmez的那个o上有两个点哦)</p><p>首先我们考虑这样一个问题,为什么那些谎报志愿的学生可以更好呢?其实很简单,因为他们可以把自己想去的学校往前提,挤掉那些虽然学校更喜欢但是却把志愿填在后面的学生。那么哪些学生不会被他们影响呢?我们来看这个例子。</p><p>有四个学生1<del>4,四所学校A</del>D,每个学校招一个学生,偏好和Priority分别是:<br>R1: B A C D(Sophisticated)<br>R2: B A D C(Sophisticated)<br>R3: B A D C(Sincere)<br>R4: B A C D(Sincere)</p><p>PA: 4 2 1 3<br>PB: 3 2 1 4<br>PC: 1 2 3 4<br>PD: 1 2 3 4</p><p>如果所有人都是真实填报,结果是<br>1 C 2 D 3 B 4 A</p><p>但是如果学生1和2试图通过改变志愿来使得自己更好,最终的结果是<br>1 C 2 A 3 B 4 D (学生2将学校A提到学校B之前)</p><p>注意到,由于学生3最喜欢学校B,而学校B也最喜欢学生3,即使学生3本身不能谎报,他也一定会被录取。而学生4则没有那么幸运了,因为他不能谎报,所以学生1和学生2都可以把学校A提到最前面而把学生4挤掉,但是由于学生2在学校A的priority中靠前,因此学生2会被学校A录取,学生1最好的结果也只有学校C,但是学生4被挤到了学校D。</p><p>总结一下,只有学生3这样,把学校B填在了首位,而且学校B又将他列在了可能谎报的学生之前的学生,才会不受影响。</p><p>从另一个角度看,这个结论也就意味着,对于任意学校A来说,可以谎报的学生都比不能谎报且没有把学校A填在首位的学生更有可能进入学校A,这其实也是一种变相的priority的体现。那么我们是不是可以通过调整priority来直接得到最后的均衡结果呢?答案当然是YES。</p><p>对于学校A,新的priority是这样组成的:<br>(i)所有可以谎报偏好的学生,和所有将学校A作为首选学校的学生,按照原先的priority进行排列;<br>(ii)所有将学校A作为次选学校的学生,按照原先的priority进行排列;<br>(iii)所有将学校A作为第三选择学校的学生,按照原先的priority进行排列;<br>……依此类推</p><p>还是拿上面的例子来看<br>P’A: 2 1 4 3<br>P’B: 3 2 1 4<br>P’C: 1 2 4 3<br>P’D: 1 2 3 4</p><p>在新的priority下,所有的稳定匹配就是在只有部分学生可以谎报志愿时所有的纳什均衡。如果所有学生都能谎报志愿,那么新的priority和原本的priority是一致的,所有的稳定匹配就是所有的纳什均衡,这个之前已经讨论过的结论就是新结论的一个特例啦~</p><hr><p>(2015.06.16 @ 微信公众号 @ 《第三讲:More on College Admission 择校问题补遗(二)》)</p><hr>"},{"title":"第三讲:More on College Admission 择校问题补遗(一)","path":"2015/06/12/ECON05/","permalink":"http://sx349.github.io/2015/06/12/ECON05/","content":"<p>择校问题是个筐,什么都能往里装,啊不对,也不是什么都能往里装,但是确实这个筐里还是装了很多好玩的东西,值得拿出来扒一扒。今天首先要扒的,就是各位高三考生马上要面临的一件大事——平!行!志!愿!(Chinese Parallel Mechanism)</p><p>上一讲介绍了波士顿机制,童鞋们有没有觉得这个模型很熟悉?好像有点像中国的高考志愿对吧?这两者之间确实有些相似之处,但又不尽相同,经济学家们自然关注到了这个问题。</p><p>今天要介绍的这篇paper是Chen & Kesten (2013) “From Boston to Chinese Parallel to Deferred Acceptance: Theory and Experiments on a Family of School Choice Mechanisms”名字很长,文章也很长(83页……),但是其实主要的逻辑线很简单。</p><p>首先登场的当然是相爱相杀的BM和DA两兄弟的血泪史,由于DA的强势崛起,BM被波士顿公立学校委员会抛弃了,但是随后一大票经济学家又了出来,表明BM在效率上各种血虐DA。不料此时作者荡开一笔,将目光移开这场没有硝烟的战争,在神秘而遥远的东方,有一条沉睡了千年的巨龙,它的名字叫作中国,在大学录取时采用了一种神奇的机制,称为Shanghai Mechanism,因为这套机制最早出现在上海的高中录取流程当中;当然, 后来它的名字就变成了Parallel Mechanism,也就是平行志愿。各地的高考录取慢慢接受了这套机制,当然也做了一些小小的改动,比如北京就是第一志愿+三个平行第二志愿(这是作者说的,不知道是不是现在还是这样哈),这些变种就被称为Partial Parallel Mechanism;所有这些,统称为Chinese Parallel Mechanism。</p><p>虽然看起来BM、DA和平行志愿大不相同,但其实我们可以把它们统一归纳为一族机制。在这一族机制中,这三种机制唯一的区别只在于一个参数e,对于BM来说,e=1,对于DA来说,e=∞,对于不同的Chinese Parallel Mechanism, e=2~∞(当e=2时,就是Shanghai Mechanism)。</p><p>那么这个参数e到底是什么意思呢?它表示的是,在一个分阶段进行的不删档录取机制中,每一阶段可以填写的学校数量。<br>对于BM,e=1,也就是每个学生每阶段只能写一个学校,这就相当于每次挑自己未被拒绝的偏好最高的学校进行申请,如果被录取了,学校就不会再招新的学生进来了。<br>对于DA,e=∞,也就是每个学生每阶段能把所有的学校按自己的偏好填进去,这就是一个普通的DA算法。<br>对于平行志愿,比如e=2,每个学生每阶段能够填2个学校,然后学校和学生就根据这个有限的偏好,进行一轮DA。在这一轮中,学校对于后来的优秀学生会踢掉原先稍差一点的学生。但是这一轮DA结束后,已经招进去的学生就不会被踢掉了,只有那些没有被招进去的学生,和没有招满学生的学校进入下一轮,每个没学上的学生还是填2个学校,再进行一轮DA,依此类推。大体上来说,可以理解为本一填两个志愿,如果录取了就录取了,如果没有录取,就进入征求平行志愿,这时候已经招满学生的学校就不会再有名额了。</p><p>这样一来,平行志愿就成了BM和DA的一个折衷。我们知道,BM的优点在于帕累托最优,也就是有效率(是这一族机制中唯一一个);而DA的优点在于给出稳定匹配,而且对于学生这一方来说是strategy-proof的(也是这一族机制中唯一一个)。折衷的结果就是,平行志愿比DA更有效率,比BM更稳定,而且更不会被操纵。当然,随着参数e的不同,这些优点也不尽相同,e越大(越倾向于DA),可操纵性就越低,e越小(越倾向于BM),效率就越高(会有更多的学生进入他们的首选志愿学校)。至于稳定性方面,如果e’是e的整数倍,那么前者比后者更稳定,但对于非整数倍之间就无法比较了。</p><p>从另一个角度来看,平行志愿其实可以通过调整学校的priority,转化为DA的结果,这其实是证明上述结论时的一个小trick:</p><p>首先,根据e的大小,把学生的实际偏好分段,每段有e个学校。学校据此调整自己的priority:</p><p>对于学校A,假设学生i把学校A填在第ai段,学生j把学校A填在第aj段,如果<br>(i) ai<aj,那么学生i会在学生j之前先进入学校A的选拔,所以i要排在j前面;<br>(ii) ai>aj,同理;<br>(iii) ai=aj,那么学生i和学生j将会在同一轮进入学校A的选拔,那么根据原本的priority顺序排列。</p><p>举个栗子,e=2,有六所学校A<del>F,三名学生1</del>3,学校A的priority原本是1 2 3,而三名学生的偏好分别是<br>R1: C D E F A B<br>R2: E F A B C D<br>R3: E F A B C D</p><p>学生1在最后一段才会填学校A,而学生2和3在第二段就会填学校A,所以学校A把2和3排在1前面,最后调整过的priority就是2 3 1。而在这个新的priority下,DA的结果就是参数为e的平行志愿的结果。</p><p>作者还做了实验检验在不同数目的学校下填报志愿的模拟结果,但是由于这个一点都不有趣感兴趣的同学们可以自己去读这个论文哈(再次强调,83页……)</p><hr><p>(2015.06.12 @ 微信公众号 @ 《第三讲:More on College Admission 择校问题补遗(一)》)</p><hr>"},{"title":"第二讲:College Admission 择校问题(下)","path":"2015/06/02/ECON04/","permalink":"http://sx349.github.io/2015/06/02/ECON04/","content":"<p>让大家久等啦,今天我们接着讲择校机制~</p><p>四、波士顿机制的性质</p><p>对于每一种机制的考量主要分为三个方面,效率(efficiency),稳定性(stability),和是否存在使学生谎报偏好的激励(strategy-proofness)。</p><p>所谓效率,就是学生是否得到了可能获得的最好的学校。在DA算法中,我们证明了,如果由学生主动“求婚”,能够得到最好的学校,但这里的“最好”指的是在所有稳定匹配中最好的学校。如果我们考虑不稳定匹配,实际上可能存在帕累托更优的不稳定匹配,每个学生至少和在最好的稳定匹配中的处境一样好,且至少有一个人匹配到了比稳定匹配中更喜欢的学校,而这正是BM的情况。实际上,BM给出了一个帕累托最优,但是不一定稳定的匹配。</p><p>考虑下面这样一个例子,有六个学生1<del>6,四个学校a</del>d,学校ab各有两个招生名额,cd各有一个招生名额。</p><p>学生对于学校的偏好是<br>R1:a 后略<br>R2:a 后略<br>R3:a b 后略<br>R4:c 后略<br>R5:c a b d 后略<br>R6:c a b d 后略</p><p>学校对于学生的priority是<br>a: 6 1 2 3 后略<br>b: 5 6 3 后略<br>c: 4 5 6 后略<br>d: 5 6 后略</p><p>根据BM,得到的匹配是<br>a: 1和2<br>b: 3和5<br>c: 4<br>d: 6</p><p>匹配的最优性是显而易见的,因为每一轮BM都把每个学生匹配到了当前最好的学校,对于任何一个学生来说,他比BM所匹配的学校更喜欢的那些学校在任何匹配中都不可能招收他。而这个匹配的不稳定性在上面这个例子也展现出来了,学校a喜欢学生6胜过学生1和学生2,而学生6喜欢学校a胜过学校d,说明上面的这个匹配不是稳定匹配。</p><p>观察一下学生6的偏好,可以发现他在第一轮选择了自己最喜欢但是却并不是那么喜欢自己的学校c,这是一个严重的失误,因为同时本来最喜欢他的学校a已经招满了学生,即使他试图在第二轮挽回(选择了学校a),仍然一再错失机会,最后只得到了排在最后的学校d;这就是“一旦拒收,一落千丈”和“浪费首选位置”的真实体现。</p><p>在这里,又不得不提到最后一个考量标准,即strategy-proofness。在单边市场中,学校的priority是不能更改的,但学生仍然可以不按照自己真实的意愿来报告自己对学校的偏好。当学生面对BM时,确实存在谎报偏好使得自己能够进入更好学校的可能,仍以上面这个例子为例,如果学生6意识到自己可能被学校c拒绝并最终不得不进入学校d,那么他完全可以谎报自己喜欢a胜过喜欢c,这样在第一轮他就把学生2挤走,成功进入了学校a,尽管他没能进入最想进入的学校c,但是仍然比进入学校d要更好。</p><p>综上所述,BM虽然能够给出帕累托优的匹配,但是这个匹配是不稳定的,而且学生有谎报自己偏好以进入更好学校的可能。</p><p>五、BM与DA的联系</p><p>不过拍脑袋一想,不对啊,既然你可以改自己的志愿,我也可以改嘛。因此,在BM真正开始分配之前,也就是学生们填报志愿的时候,就已经开始了一场没有硝烟的战争,学生们都会调整自己的志愿,使得自己能够上到更符合自己偏好的学校,这就构成了一场新的博弈。在这场博弈中,每个学生都是参与者,他们的策略就是不同的志愿填报顺序,最终的结果(outcome)就是根据每个人填报的志愿进行BM分配所得到的学校,相应的支付(payoff)就是是否获得自己更偏好的学校。</p><p>在这个博弈中,存在一系列的纳什均衡,而神奇的是,这些纳什均衡的结果恰好就是所有的稳定匹配,这个命题的证明可以在这里稍微表述一下:</p><p>首先要证明所有的纳什均衡结果都是稳定匹配,这是显然的,如果存在纳什均衡的结果是不稳定匹配,那么也就意味着有学校S想要学生s来踢掉别人或者填补空位,学生s也想要学校S而非现在匹配的学校,那么学生s只需要把学校S提到自己志愿的第一位就可以了;换句话说,学生s有进一步改变自己志愿的激励,不满足纳什均衡的要求。</p><p>其次要证明所有的稳定匹配都是纳什均衡的结果,这就更显然了,只要每个人都把自己在这个稳定匹配中匹配到的学校填在第一位,第一轮百分之百全部录取;任何人单方面的背离都不能使他更好,因为其他人把所有别的坑都填满了,他要么最后还是回去填了自己的坑,要么就没学上。</p><p>根据集合的性质,如果集合A和集合B互相包含,那么这两个集合相等,因此所有的纳什均衡结果就是所有的稳定匹配。</p><p>这告诉我们什么呢?如果将学生在BM下选择不同的志愿视为一场博弈的话,非常神奇的是,这个博弈的所有纳什均衡结果恰好是所有稳定匹配,而我们又知道,在所有稳定匹配中,学生的最优解可以由学生主动“求婚”的DA算法来得出,因此,一切又回归到DA算法,将学校招生视为双边市场,考虑学生主动“求婚”的DA算法,得到的解满足stable(稳定),student-optimal(在所有稳定匹配中对于学生来说最优)以及strategy-proof(学生没有激励去伪造自己的真实意愿,可以证明,此处略去)。</p><p>顺便说句题外话,其实在某种特殊的情况下,BM和DA一定会给出同样的结果,那就是如果所有学生对于学校的排序都是一样的,同时学校对于所有学生的排序都是一样的,这时候最好的学校招最好的学生,差一点的学校招差一点的学生,最差的学校招最差的学生,依此类推。我们常常诟病的“一考定终生”就是这样的情况,学校对于学生只有单一的衡量标准,也就是分数,如果学生对于学校本来就划分了三六九等,恰好就满足了这样的条件。当然,这种条件的出现是比较少见的,像目前中国的高考算是比较接近的例子,但高考的录取机制又不完全是BM或者DA,这个就留到下一讲吧。</p><hr><p>(2015.06.02 @ 微信公众号 @ 《第二讲:College Admission 择校问题(下)》)</p><hr>"},{"title":"第二讲:College Admission 择校问题(上)","path":"2015/05/29/ECON03/","permalink":"http://sx349.github.io/2015/05/29/ECON03/","content":"<p>讲完婚姻市场之后,我们接下来讲Matching Theory的另一个重要应用,就是College Admission,择校问题。</p><p>择校问题的市场上还是有两边,一边是学生,一边是学校,不同于婚姻市场,学生仍然只能选择一所学校,学校却可以选择多于一个学生,这是一个一对多双边市场。</p><p>实际上,像择校问题这样一对多的市场,是比婚姻市场这样一对一的市场更加普遍的一种情况,可以认为后者就是前者的一个特例;Gale和Shapley第一篇关于DA算法的论文讨论的两个例子,就是择校问题和婚姻市场。</p><p>一、一对多双边市场<br>实际上,只需将在一对一双边市场上的延迟接受算法稍作修改,就能够得到一对多双边市场上的延迟接受算法,这些修改大致包括如下几项:</p><p>(1)在婚姻市场上,男生对于所有女生进行排序,女生对于所有男生进行排序,这是非常直观的;在学校招生时,学生只能去一所学校,所以学生对于学校的排序也是很容易理解的;但是,学校对于学生的排序就有两种不同的情况:一种是,学生与学生之间是相互独立的,学校对于单个的学生建立偏好(也称为responsive或者separable preference);另一种是,学生与学生之间是有关联的,学校招生的时候对于学生的组合建立偏好。(严格地说,前者是后者的一种特例,对于responsive preference,同样可以把这个偏好写成学生的组合上的偏好)</p><p>举例来说,如果一共有3个学生1,2,3,那么前一种情况就是,学校觉得1比2好,2比3好,或是3比2好,2比1好,这是建立在“单个的学生”上的偏好;另一种情况是,学校觉得同时招1和2最好,其次是1和2和3,再次是2和3,再次是1和2,等等,这是建立在“学生的组合”上的偏好。后一种情况在什么时候会发生呢?举个例子,T大每年都要参加帝都高校长绳比赛,长绳需要12名队员,选拔长绳队员的时候,并不是从各个院队里一个一个选拔出来(“单个的学生”的偏好),而是选拔了最好的一个院队(“学生的组合”的偏好),因为这些学生已经经过长期的磨合和练习,把他们组合在一起将会有协同效应,换上别的院跳得好的同学,反而有可能拖后腿。</p><p>后一种情况,正如上一讲所说的,存在反例说明可能并不存在稳定匹配。然而,对于前一种情况,稳定匹配总是存在的,原因在下面进一步阐述。</p><p>(2)引入配额(quota)的概念,即学校的招生名额,这是很自然的,一个学校能够招收的学生数量总是有限的。</p><p>(3)由于配额的存在,对于稳定匹配的要求也发生了变化:<br>IR: 学生不会选择unacceptable的学校,学校不会选择unacceptable的学生;<br>Pairwise Stability: 不存在如下情况:<br>(i) 学校S有空位,且学生s对于学校S是acceptable的,且对于学生s来说,学校S比目前的情况(没有学上或是被另一学校S’录取)更好;<br>(ii) 学校S没有空位,但在录取的学生中有一名学生s’,对于学校S来说,学生s比学生s’更好,且对于学生s来说,学校S比目前的情况更好。</p><p>回到上面这个问题,为什么当学校对于学生的偏好是建立在“单个的学生”上时稳定匹配总是存在的呢?在一对一双边市场上,稳定匹配的存在性是由DA算法构造出来的,因为构造出来的这个匹配是稳定的,所以稳定匹配存在。而对于一对多双边市场,我们不需要再按照证明一对一双边市场上那样完全走一遍,实际上,当学校的偏好建立在“单个的学生”上时,我们完全可以这样做:</p><p>学校A有三个名额,那么我就把学校A拆成学校A1、A2、A3,分别只有一个名额,它们对于学生的偏好和学校A是一样的,而学生认为学校A1、A2和A3是一样好的;对其它学校也是一样处理。DA算法并不要求偏好是严格(strict)的(这个要求对于稳定匹配的存在性或者说DA算法并没有影响,只影响了最后的稳定匹配是否构成Lattice),这样一拆,相当于构造了一个一对一双边市场,新市场上的稳定匹配存在(由DA算法给出),则原市场上的稳定匹配只需要将A1、A2、A3重新合并回学校A即可。</p><p>二、单边市场</p><p>实际上,对于学校招生,还有另外一种模型,即单边市场模型。在单边市场中其实还是存在两边,但是一边是人,另一边是物品(object),由于我们只关注人的行为,而被分配的物品是没有“行为”的,所以称为“单边”。</p><p>最基本的单边市场是房屋市场,每个人对于房屋都有自己的偏好,房屋自己不会挑人,随便谁住都可以;但是,当我们考虑学校招生这样一个单边市场模型时,学校对于学生肯定不是随便挑的,每个学校也会有自己的“偏好”,但是这个“偏好”不是通常意义上的preference,而被称为priority,两者在形式上是一致的,但是内涵大不相同。</p><p>不同于人的偏好是自己决定的,学校的priority是受到限制的,一般来说,首先由政府制定一系列的规则,然后根据这套规则,每个学校就自动产生一套priority。举个栗子,政府规定,学校优先招纳距离学校近的学生,在距离差不多的情况下,招纳学分绩更高的学生,在学分绩也相同的情况下,优先招纳女生,这样一来政府完全可以不需要学校参与决定priority,在这个priority中也完全没有体现出学校自己的意愿,相应地,当我们考虑此时市场的效率(efficiency)时,我们就不需要将priority考虑进来,而只考虑学生这边是否进入自己心仪的学校。</p><p>由于priority取代了学校的preference,对于稳定匹配的第二条要求就有了一个新的名词,叫作respecting priority,基本上等价于双边市场中的pairwise stability。</p><p>三、波士顿机制(BostonMechanism)</p><p>在DA算法之前,有一个模型在美国学校招生中非常流行,这个模型称为波士顿机制(Boston Mechanism),因为其最先在波士顿试水并推广而得名。如果没有猜错的话,纽约在2003年之前使用的就是BM或者BM的变种,因为相关报道中有两句话非常符合BM的实际情况:</p><p>“所以有些目标高远、填报了多家热门学校的学生,一旦被写在第一栏的学校拒收,就会在选校表中一落千丈。”<br>(这是BM机制中常有的,参见下面的例子中的学生6)</p><p>“家长们担心如果选错了,他们的孩子会‘浪费’至关重要的首选位置。”<br>(这句话虽然写在新机制DA运作之后,但表达的是BM的流毒,即浪费首选位置的想法,同样参见下面的例子中的学生6)</p><p>首先来简要介绍一下什么是BM,从流程上来说,它类似于由学生主动向学校“求婚”的DA算法,但差了这么一句话:</p><p>在BM中,如果学生s被学校S录取,那么之后不会被替换掉;而在DA算法中,如果后面出现更好的学生,学校S会踢掉学生s。</p><p>简单地说,DA对于BM最重要也是最神奇的改进只在于:允许“见异思迁”。</p><p>咱们先中途下课,下次课的时候,我们将继续介绍波士顿机制的性质,以及它和我们的老朋友延迟接受算法之间的种种爱恨(划掉)纠葛。</p><hr><p>(2015.05.29 @ 微信公众号 @ 《第二讲:College Admission 择校问题(上)》)</p><hr>"},{"title":"第一点五讲:Adjusted Preference 偏好调整","path":"2015/05/25/ECON02/","permalink":"http://sx349.github.io/2015/05/25/ECON02/","content":"<p>上一讲给大家介绍了DA算法,不过老是给大家讲这些五十年前的内容也挺没有意思的,所以这一讲我们来讲点新东西~</p><p>在上一讲中我们提到,对于匹配其实有两种不同的解读方式,一种是Centralized Market,另一种是Decentralized Market。在Decentralized Market中,稳定匹配的含义是市场上的男男女女会不断地分分合合,直到最后不再变动为止。</p><p>但实际情况并非如此,我们会观察到,不是每个人都愿意分分合合的,有时候就算能够有机会去找更好的对象,但是因为各种成本的问题,也许就会选择继续和现在的对象在一块儿了;就比如说,有些父母虽然感情不和,但是考虑到孩子还在读书的缘故,可能就会选择不离婚,这样的新闻也是挺常见的。</p><p>这种情况,我们可以统称为离婚成本(Divorce Cost)。在一个Decentralized的婚姻市场上,有些匹配本来应该是不稳定的,但是由于存在离婚成本的缘故,原来更好的选择反而不如保持现在的对象,结果就变成“稳定”的匹配了。由于在原本的设定中,我们对偏好并没有采用赋值的形式,而是采用了排序的形式,所以我们在表示离婚成本的时候,也不能直接说它是某个值c,而是考虑对偏好进行顺序上的调整。</p><hr><p>在最简单的情况下,假定现在市场上已经有一个禀赋的匹配μ,我们依照这个匹配来调整每个人的偏好:某个男生m匹配的对象是μ(m),在他的偏好中,比μ(m)高一位的是女生w;在考虑离婚成本的情况下,m把μ(m)上移一位,排到了w的上面,μ(m)同时也把m上移一位。当然,如果对象本身已经是第一位,那仍然保持不变就可以了。</p><p>如果匹配μ在调整过偏好之后的市场上是稳定的,那么我们就称它为1-稳定匹配,这里的1表示我们把匹配对象上移一位。如果上移k位,那就叫k-稳定匹配。原本的稳定匹配,我们也可以称为0-稳定匹配。</p><p>举个例子,现在有三男三女,分别是a,b,c和1,2,3,他们的偏好如下所示:</p><blockquote><p>Ra: 3 2 1<br>Rb: 1 3 2<br>Rc: 1 2 3<br>R1: a b c<br>R2: a c b<br>R3: b a c</p></blockquote><p>如果不考虑离婚成本,那么唯一的稳定匹配是:(a 3)(b 1)(c 2)<br>如果考虑离婚成本,而市场上本来有一个禀赋的匹配μ=(a 1)(b 3)(c 2),那么我们相应地调整偏好为</p><blockquote><p>R’a: 3 1 2<br>R’b: 3 1 2<br>R’c: 2 1 3<br>R’1: a b c<br>R’2: c a b<br>R’3: b a c</p></blockquote><p>在调整过的偏好下,μ=(a 1)(b 3)(c 2)是稳定匹配,所以它是一个1-稳定匹配。</p><p>现在,让我们换一个角度。我们之前提到,稳定匹配首先要求个体理性(Individual Rationality),其次要求配对稳定(Pairwise Stability),也就是说,首先匹配对象至少要和单身一样好,其次不存在一男一女,喜欢对方胜过自己的配偶。如果我们直接从这个定义上类比地去看k-稳定匹配,那么要求就应该是:首先匹配对象至多比单身差k位,其次不存在一男一女,喜欢对方胜过自己的配偶k位。</p><p>可以证明,这两种对于k-稳定匹配的定义是等价的,而由后面这种定义,我们很容易发现,随着k的增大,k-稳定匹配组成的集合是向前包含的,也就是说,1-稳定匹配中包括了所有0-稳定匹配(也就是原本意义上的稳定匹配),2-稳定匹配包含了所有1-稳定匹配,以此类推。当k充分大的时候,k-稳定匹配将会包含所有的匹配,因为所有人会把自己在当前匹配下的对象上移到第一位。</p><p>因为这样一个向前包含的缘故,如果我们只考虑一方的福利,越来越多的选择一定意味着效用边界在不断向外推进,比如上面的例子中,对于女生1、2、3来说,(a 1)(b 3)(c 2)比起(a 3)(b 1)(c 2)来说就更优。但是,如果我们要同时考虑双方的福利,仅仅通过调整偏好,我们还是没办法得到比任何稳定匹配帕累托更优的匹配,如若不然,假定有一个稳定匹配μ和一个帕累托优于μ的匹配ν(未必是稳定匹配),不妨假定ν(m)优于μ(m),显然ν(m)在μ中的对象不是m,那么m又优于μ(ν(m)),我们就有了一对阻碍μ的blocking pair (m,ν(m)),和μ是稳定匹配矛盾。</p><p>由于在k-稳定匹配下,我们仍然有k-blocking pair的定义,即“存在一男一女,喜欢对方胜过自己的配偶k位”,所以,我们可以将Roth & Vande Vate(1990)的结论复制过来,也即从任意一个匹配出发,以正概率随机选择一个k-blocking pair解决掉,最终一定以概率1到达一个k-稳定匹配;原论文的证明也可以直接套用。</p><p>在上文中提到,对于偏好的定义,我们采用的是排序而不是赋值的形式。如果我们采用赋值的形式,也就是说,对于男生m,和女生w1匹配时,获得效用u(m,w1),和女生w2匹配时,获得效用u(m,w2),依此类推,这时候我们就可以用一个数c来表示离婚成本了。通过恰当地定义效用和离婚成本,我们可以模拟出排序形式下的k-稳定匹配的定义,也就是说,采用赋值的形式来表示偏好,是对于原本问题的一个推广。</p><p>我们重新定义稳定匹配如下:一方面,如果和现在的配偶离婚回到单身能够获得收益,则收益不能超过离婚成本;另一方面,不存在一男一女,通过更换配偶所得的收益能够超过离婚成本。所有我们已经得到的结论,包括单方面的效用边界拓展,或是通过随机过程达到稳定匹配,在这个情境下依然适用。</p><p>突然发现,咱们是不是还没有介绍这是哪篇论文?其实这是我自己撰写的一篇论文,大家引用的时候记得注明Xu(2015)哦<del>最后,诚招co-author继续研究这个问题哦哈哈</del></p><hr><p>(2015.05.25 @ 微信公众号 @ 《第一点五讲:Adjusted Preference 偏好调整》)</p><hr>"},{"title":"第一讲:Marriage Market 婚姻市场","path":"2015/05/19/ECON01/","permalink":"http://sx349.github.io/2015/05/19/ECON01/","content":"<p>匹配理论(Matching Theory)讨论了许多非常生活化的经济学,让你充分意识到,经济学本质上解决的是稀缺资源如何分配的问题,而不仅仅是GDP、收入、消费等等这些一听就像要钻进“钱”眼里的名词。在本讲中,我们首先介绍婚姻市场(也可以叫恋爱市场)中的匹配模型,参考书目是Roth & Sotomayor(1992)“TWO-SIDEDMATCHING:A GAME-THEORETIC MODEL AND ANALYSIS”。</p><a id=\"more\"></a><hr><h1 id=\"婚姻市场\"><a href=\"#婚姻市场\" class=\"headerlink\" title=\"婚姻市场\"></a>婚姻市场</h1><p>首先,婚姻市场是一个双边市场,分为男生和女生。由对称性,以下的讨论我们都站在男生的视角,但同样的结论对于女生也是成立的。</p><p>每个男生对于所有女生有一个偏好(就是对女生有一个排序),反之女生对于所有男生也有一个偏好。不过,在男生的偏好中,除了所有女生之外,还包括他自己;所有排在自己之后的女生都被视为unacceptable(换句话说,宁愿单身也不愿意接受这些女生),反之女生也是如此。</p><p>为了方便起见,我们一般默认偏好是严格(strict)的,也就是说,对于任何两个女生,男生总是喜欢一个胜过另一个,不会出现觉得两个女生一样好(indifference)的情况。这个要求对于我们后面介绍的Deferred Acceptance算法其实没有影响,但是对于其它一些重要结论则是必要的条件。</p><hr><h1 id=\"匹配\"><a href=\"#匹配\" class=\"headerlink\" title=\"匹配\"></a>匹配</h1><p>一个匹配(matching),就是将婚姻市场上的一部分男生和女生配对,然后让剩下的男生和女生保持单身,我们用函数μ(·)来表示这个匹配,μ(x)就是x的配偶。μ(x)满足以下三条要求:1. 男人的配偶不是自己(单身)就是女人;<br>2. 女人的配偶不是自己(单身)就是男人;<br>3. 每个人的配偶的配偶都是本身。(不考虑同性恋和一夫多妻/一妻多夫)</p><p>但是,匹配有很多种,比如说每个人都保持单身,这也是一个可行的匹配,但是这样的匹配几乎是不可能维持的(因为单身狗也有春天……)。那些可以维持的匹配,一定存在一些特殊的特点,我们把这些匹配称为稳定匹配(stable matching)。具体来说,稳定匹配指的是满足以下两个要求的匹配:</p><ol><li>Individual Rationality(简称IR,大致可以翻译为“个体理性”),每个人不会接受unacceptable的配偶。</li><li>Pairwise Stability(大致可以翻译为“配对稳定”),不会同时有一男一女喜欢对方胜过自己的配偶。(如果有的话,我们称这一男一女为blocking pair,他们阻碍(block)了这个匹配μ)</li></ol><p>一般来说,第一条是很容易满足的,比如说每个人都保持单身,就是满足IR的匹配。但是第二条就不一定了,而且在现实世界中很有可能就发生这样的事(NTR)了。那么问题来了,挖掘机技术到底哪家强?啊不对,怎么找到稳定匹配呢?</p><p>在进入下一部分之前,首先来解释一下稳定匹配的意义,或者说,为什么我们要找出稳定匹配。实际上,对于寻找稳定匹配的过程,存在两种不同的解释,或者说就是两种不同的市场:其一称为”Decentralized Market”,就是把一堆男生女生扔进市场,然后他们就开始不断地追人甩人,最后他们达到了某种稳定的状态,这种状态就是稳定匹配;其二称为”Centralized Market”,就是现在开了一家T大婚姻介绍所,所有T大的单身狗把自己的信息提交过去,然后男生给女生排个序,女生给男生排个序,然后T大婚介所根据双方的偏好信息,排好一套匹配,然后告诉男生“你命中注定的女神就是XXX”,告诉女生“你命中注定的男神就是XXX”,如果双方皆大欢喜,不存在NTR的可能,那么也是稳定匹配。下面介绍的Deferred Acceptance算法,看起来好像是按照前一种市场进行的,但实际上我们可以假定已经收集到了所有的男女生信息,然后让计算机模拟一下流程,得到了最终的稳定匹配,然后咱们的婚介所就可以关张了(当这个问题是静态时,这样的婚介所开一次就可以关门了……)。</p><hr><h1 id=\"寻找稳定匹配\"><a href=\"#寻找稳定匹配\" class=\"headerlink\" title=\"寻找稳定匹配\"></a>寻找稳定匹配</h1><p>知乎上有一句经典名言:不问有没有就问为什么都是耍流氓(以及其它变种……)所以在找稳定匹配之前,一定先要问一句有没有。如果有同学表示这他喵的不是显然的么,那就让我们(等会儿)用一个实际例子来糊他一脸~</p><p>实际上,婚姻市场这样一个双边市场是非常特别的,它有这样两个性质:</p><ol><li>它把所有市场中的人分成了两边,互相进行选择;</li><li>它是一对一的;</li></ol><p>这两个性质为什么特殊呢?因为如果不符合,那就可能没有稳定匹配;具体来说,如果市场上只有一类人(室友选择问题),或者有三类人(家庭配对,有父亲、母亲和孩子)或者市场上是一对多的(劳动力市场,一个公司招好几个工人,而且工人之间有协同效应),都能够找到反例。比如以下这个例子来自Gale & Shapley,是室友选择问题提供的一个不存在稳定匹配的反例。</p><p>有4个同学a,b,c,d分在两个寝室,每个寝室有两个人,每个同学对另外三个人各自有一个偏好(当然,这时候就不允许保持“单身”啦)。</p><ul><li>P(a)=b,c,d(a最希望和b分在一个寝室,然后是c,最差是d,以下同理)</li><li>P(b)=c,a,d</li><li>P(c)=a,b,d</li><li>P(d)=arbitrary(d同学显然是最不受欢迎的室友= =)</li></ul><p>4个人分在两个寝室,一共就只有三种分法</p><ul><li>μ1={(a b)(c d)} 不行,因为b和c希望和对方住一块儿均胜过自己的室友;</li><li>μ2={(a c)(b d)} 也不行,这回闹别扭的是a和b;</li><li>μ3={(a d)(b c)} 还是不行,这回闹别扭的是c和a;</li></ul><p>于是这个室友选择问题就没有稳定匹配了。<br>那双边市场有没有稳定匹配呢?有。为什么呢?我们往下看。</p><p>假定我们先接受了这一事实,然后开始尝试找稳定匹配。一个比较直接的想法是,首先大家自由配对,选择自己能够接受的配偶,比如说先让所有人单身(避免了Individual Rationality的问题),这时候的匹配基本上不可能是稳定的,不满足Pairwise Stability嘛。那好,我们在所有阻碍了这个匹配的男女(还记得阻碍(block)的定义吗?不记得的向上翻)中,找一对出来,让他们俩成了,然后让他们原本的配偶(如果有的话)变成单身,就得到一个新的匹配,比原来那个稍微稳定了一点吧?然后就不断重复这个过程就行啦~</p><p>所以这个想法对吗?也不对也对。说不对,是因为Knuth找到了这样一个例子,如果选择这对(狗)男女选得特别不巧的话,还真有可能进入一个死循环。说对,是因为Roth 和 Vande Vate 给出了一个定理说,对任何一个匹配来说,一般不会只有一对男女阻碍了这个匹配成为稳定匹配,如果我在所有的匹配中随机地选一对,把他们解决掉,然后在新的匹配中随机地选一堆,然后解决掉,不断重复下去,只要每种选择的概率都是正的(都有可能被选上),那么最终这个过程以概率1达到一个稳定匹配,这篇论文“Random Paths to Stability”(Roth & Vande Vate, 1990)短小精悍,证明也很巧妙,不妨一读。</p><p>但是,这种办法实在太折腾了,而且听起来太靠运气了,有没有不那么麻烦的算法呢?那就是Gale & Shapley在1962年提出的著名的Deferred Acceptance Algorithm,简称DA算法,翻译过来叫作延迟接受算法,简单地说,就是(假定男生主动求婚)允许一个女生在接受某个男生之后,如果碰到更心仪的男生可以把前一个给甩了。</p><p>接下来就是揭露真相的时刻啦,为什么我们说双边市场有稳定匹配呢?因为Gale 和 Shapley 说,只要按照我们这个DA算法走一遍,得到的一定是稳定匹配,所以稳定匹配就一定存在。换句话说,他们就是通过用DA算法构造出一个稳定匹配,证明了双边市场上稳定匹配的存在性。</p><p>然后我们来看一看这个具体的流程(假定男生主动求婚);</p><ul><li>第一轮,每个男生向自己的偏好中排名最高的女生求婚;每个女生拒绝掉所有unacceptable的求婚者,然后在剩下的求婚者中,选择在自己的偏好中排名最高的男生,然后和他订婚(keep engaged),并且拒绝掉其他人。</li><li>之后的每一轮,每个还没有名草有主的男生,向自己的偏好中排名最高的没有拒绝过自己的女生求婚;每个女生仍然是拒绝掉所有unacceptable的求婚者,然后在剩下的求婚者以及上一轮订婚的那个男生中,选择在自己的偏好中排名最高的男生,和他订婚(渣女!哼!),并且拒绝掉其他人。</li><li>依此类推,每一轮那些没有被接受的男生都会向下一顺位的女生求婚,直到被订婚或者所有acceptable的女生都拒绝了他。</li><li>如果没有男生求婚了,算法结束(这个过程可以保证是有限的),此时所有已经订婚了的男生和女生结婚,剩下的人保持单身。</li></ul><p>可以证明,这个时候给出的匹配一定是稳定的。否则假定男生m和女生w打算私奔,那么m一定喜欢w胜过目前的配偶μ(m),根据算法的流程,m一定在和μ(m)配对前向w求过婚并被拒绝了,那这意味着w遇到了一个比m更好的男生m’,而w目前的配偶μ(w)至少和m’一样好,所以μ(w)比m更好,那么女生w就不应该愿意和男生m私奔,矛盾。</p><hr><h1 id=\"稳定匹配的好坏\"><a href=\"#稳定匹配的好坏\" class=\"headerlink\" title=\"稳定匹配的好坏\"></a>稳定匹配的好坏</h1><p>实际上,稳定匹配自然不是只有DA算法给出的这一个(或者说两个,男生主动和女生主动各自得到一个可能不同的稳定匹配),那么这些稳定匹配各自是有好有坏的。Gale 和 Shapley进一步证明,如果偏好是严格(strict)的,那么由男生主动求婚得到的这个稳定匹配对男生来说是最好的,称之为M-optimal;反之,由女生主动求婚得到的稳定匹配就是W-optimal的。</p><p>让我们先往前回一点,先定义一下对于双方来说什么是好的匹配,什么是不好的匹配。我们用”>m”(m应该是下标)来表示对男生来说的“优于”,假设我们现在有两个匹配μ和ν,μ >m ν的意思就是每个男生在μ中的配偶至少和ν中一样好,且至少有一个男生在μ中的配偶比ν要好;相对应的,女生也有这样的一个比较关系“>w”。(不是”>w<”谢谢)</p><p>牛逼轰轰的Knuth又一次站了出来,他告诉我们,如果两个稳定匹配μ和ν满足 μ >m ν,那么一定有ν >w μ。也就是说,男生认为更好的配对,女生一定认为更不好,反之亦然。这个定理最牛逼的地方在于,它有一个非常直截了当的推论:</p><p>由男生主动求婚得到的M-optimal的稳定匹配中,每个男生都匹配上了他在所有的稳定匹配中可以匹配到的最好的女生(当然如果他在所有稳定匹配中都只能单身就……),每个女生则匹配上了她可以匹配到的最差的男生(或者始终保持单身……);反之,由女生主动求婚得到的W-optimal的稳定匹配中,每个男生匹配上了最差的女生,每个女生则匹配上了最好的男生。</p><p>更加神奇的是,即使我们拿在手里的是两个不能比较的稳定匹配(也就是有些男生更喜欢μ,有些男生喜欢ν),那么我们可以让每个男生在μ和ν的两个配偶中选一个更好的,每个女生在μ和ν里的两个配偶中选一个更差的,然后:</p><ol><li>一定能得到新的匹配(也就是不会出现两个男生选同一个女生,或者两个女生选同一个男生,或者某男生选了某女生,该女生却没有选该男生);</li><li>新的匹配也是稳定匹配;</li><li>新的匹配对每个男生来说比μ和ν都要好。</li></ol><p>数学上,这意味着所有的稳定匹配μ构成了一个代数,称为lattice,它有一些很好的性质,这里就不再展开。</p><p>最后给各位男生女生一句忠告,由于算法要求自始至终只有一方propose,所以主动不主动和算法流程的关系不大,但真正有关系的在于偏好啊,你不主动一点男神/女神一直把你放在unacceptable的人里面那你就一点机会都没有了啊!</p><hr><p>(2015.05.19 @ 微信公众号 @ 《第一讲:Marriage Market 婚姻市场》)</p>"},{"title":"第零讲:Grading Exams 评分标准","path":"2015/05/15/ECON00/","permalink":"http://sx349.github.io/2015/05/15/ECON00/","content":"<p>本学期,学校推出了学分绩改革,从原来的百分制改为等级制。学校这一创举的初衷我们无从揣测,也许主要是为了和国际接轨。不过,这么好的一个policy shock(政策冲击),不拿来琢磨琢磨实在有点可惜。</p><p>不过,我对于实证实在不太感冒,所以Event Study啊,Regression Discontinuity啊,Difference-in-difference啊之类的就留给后人了(其实只是因为不太可能从学校那边拿到大家在学分绩改革前后的成绩),作为一个立志跳进理论经济学大坑的本科僧,我就给大家说道说道,理论经济学家对这个问题有什么高见。</p><p>今天我要给大家介绍的paper是Dubey and Geanakoplos(2010) “Grading exams: 100, 99, 98,… or A, B, C?” 不得不说,这个名字起得就极为切合这一讲的主题。这篇论文发表在Games and Economic Behaviour上,在博弈论方向也算得上是除掉五大顶级期刊外的一流刊物。</p><hr><p>不同于我们平时把这两种打分方式称为“百分制(percentage grading)”与“等级制(letter grading)”,作者在这篇文章中把这两种打分方式分别称为”fine”和”coarse”,也就是“精细打分”和“粗糙打分”。一般来说,我们总觉得精细比粗糙要好,首先看起来更有区分度,更何况在现行的学分绩计算方式下,差一分估计就差了好几个名次。其次,如果我告诉你90分及以上是A,85~89分是B,很多人都会拼尽全力去拿90分,但是却不会愿意花功夫去再多拿几分,所以一度广为流传这样一句话:“六十分万岁,多一分浪费”。在经济学上,这也是一个常见的现象,有时函数上会出现向外凸的折点(kinky points),效用最大化的结果往往就是大多数人都集中到这个折点上。</p><p>但这并不能说明粗糙就一定不能比精细要更好:区分度方面,有时92分和93分之间实际上并没有显著的差距,打出这样的分数对于某些课或是某些教授来说都是凭感觉,结果这一点感觉就主宰了同学们的命运,如果老师当时大笔一挥统统给A,反而没有这样的问题;更进一步地说,粗糙还有可能给所有人同时创造激励选择更努力的学习,比如说下面这个例子:</p><p>现在两个童鞋学霸A和学渣B马上面临只有这两个人参与的期末考试,他们不在乎具体的成绩高低,但在乎能不能超过对方。假定成绩不仅受到学生自己的特征(学霸还是学渣)和采取的行动(学习或者不学习)的影响,还会受到一些随机因素的干涉,因而会在一定范围内波动(假定是均匀分布),但学霸A即使是在不学习且成绩波动到最差的情况下,依然要比学渣B学习且成绩波动到最好的情况更好。如果我们直接打分,那两个人都不会好好学习,因为学渣B知道自己一定比学霸A差,学习也没用,所以索性不学,而学霸A意识到这一点,在学渣B有占优策略不学习的时候,学霸A的最优反应还是不学习。</p><p>但如果教授突然宣布,现在我们来采取等级制而不是百分制,而且成绩设定是这样的(假定他非常清楚学生们的各项波动的上下限等等),只有超过学霸A学习时的平均值才能得A,而只要超过学渣B学习时波动到的最差值就能拿B,而学生B不学习的情况拿的是C。学生仍然关注的是自己的相对成绩,但是他们只能看到等级而不知道分数,所以对B来说,选择学习有一定的概率和A拿到一样的等第,显然比不学习要好。同样的,对A来说,选不学习只能得到B,那就和学渣B一样了,但选择学习则仍有可能超过学渣B;最终双方都会选择学习。</p><p>当然,作者考虑的粗糙打分还是和我们所想的有那么一点不一样的,比如说,他给了教授更大的自主权,教授可以根据学生的实际情况,决定到底多少分到多少分是A,多少分到多少分是B,依此类推。这初看下来好像和咱们学分绩改革的规定内容不符,给的等级和相应的学分区间不是已经写出来了吗?但是细想想,教授完全可以先按照实际成绩(比如期中30%,期末60%,作业10%)给你打个分,然后根据这个分数估算各档等级成绩的分数段给出等级,然后再根据学校规定的什么等级在哪个分数区间给换算回来。说起来,不就是我们一直在经历的“调分”嘛!</p><p>说到调分,我们平时也称为“curve”,就是按照某种分布曲线(比如正态分布),根据同学的相对成绩高低,分别安排到曲线的不同位置上,从而调整得分;也就是说,这是根据相对成绩而非绝对成绩指定的打分标准。作者在论文中特意之处了这两种不同的打分方式:一种称为“绝对得分(absolute grading)”,你是多少分就打多少分(打相应的等第),上90分就是A,不管有多少人;另一种称为“相对得分”,或者作者更多地称为“调分得分(grading on curves)”,你是前百分之几,我给你打相应的等第,比如只有前5%才能拿A+,就算你考了99分,如果前5%的人拿了100,你就只能是A而不是A+。在这两种基本的打分机制之外,我们还能玩出更多的花样,下面我们还会讲到。</p><hr><p>好了,以上这些大概就是背景知识,不过信息量还是很丰富的,接下来我们就可以开始看看这篇文章的各种结论了。先来罗列一下:</p><ol><li>如果学生之间是“分离的(disparate)”,就像我上面举的那个例子,好学生就算不学习也比一般学生学习了要好,那么最优的绝对得分打分机制永远是粗糙的;</li><li>如果学生之间是“同质的(homogeneous)”,也就是说,如果一个学生学习,他的成绩总是服从某个对所有学生而言相同的分布,而不学习,也总是服从另一个对所有学生而言都相同的分布。在这种情况下,如果我们仍然考虑所有的绝对得分打分机制,一般粗糙打分会是更好的,但也有时精细打分会是最优解,取决于学习和不学习时这两个分布的性质。</li><li>无论是上面哪种情况,只要学生的成绩在相互之间是独立的,使用绝对得分都至少和使用相对得分一样好。</li><li>作者最后带过几笔有关不完全信息、异质学生和期中考作为拓展的一些讨论。</li></ol><p>这里面,最不好理解但又最好解释的,是让我们感觉很奇怪的第三条。说它不好理解,是因为我们既然假设学生关注的就是自己的相对成绩而不是绝对成绩,那为什么不直接用相对得分的打分机制呢?更何况,在现实生活中,相对得分用的也是很广泛的。但说它好解释,是建立在作者允许教授根据学生的情况自由裁量每个等级对应的(原始)分数的区间的基础上。作者证明,在这种情况下,教授可以通过调整这些区间,使得自己使用绝对得分达成使用相对得分所得到的任意结果。而显然最优的绝对得分打分方式建立在最优的区间选择上,也即是说最优的绝对得分结果至少和相对得分结果一样好。</p><hr><p>回过头来看一下第一条和第二条。首先是学生之间分离(disparate)的情况,上面给大家举了最简单的例子,说明粗糙打分能够激励学生学习。这种经过精心设计以激励学生学习的粗糙打分机制还有另一条隐含的性质,大致可以称为金字塔型(pyramiding),也就是说获得每个等级的学生数量,从顶至下是越来越少的;比如在上面这个例子中,即使学霸A选择学习,也只有1/2的概率获得A等级,而一般概率的学霸A加上学渣B,就一共有3/2人次可能获得B等级了。</p><p>对这种情况的推广,比如说有更多类型(即能力水平)的学生(仍然是分离的),或是每个类型的学生有多于一个,我们总能够合理地选择各等级的区间,构造出一个最优的粗糙绝对得分打分机制,满足以下两个性质:首先,每个学生在这个机制下,都会有选择学习而非不学习的激励;其次,由高到低,每个类型的学生中,只有一部分人能够得到对应顺序的那个等级,唯一的例外是最差的那个类型全部得到最差的等级;除此之外,最差的等级也还包含了一些次差类型的学生,这使得他们的人数遥遥领先其它等级,成为金字塔型夯实的地基。</p><p>既然我们已经知道,绝对得分比相对得分好,在所有绝对得分机制中,粗糙打分又是最优的,那是不是粗糙绝对得分打分机制就是所有机制中最优的呢?并非如此。正如我上面提到过的,我们还可以在机制上继续玩花样,特别地,可以在off-equilibrium path上放更加恶劣的情况,构造复合机制。作者举的例子正是一个复合机制:教授承诺,根据最后的考试结果,如果所有人都选择学习,那我们就选择实施粗糙绝对得分打分机制(也就是等级制),这样会有一半的学霸只能拿等级B;但如果发现有学霸选择不学习,那教授会给所有学习的学霸最后等级均为A。这时候,和不学习的学霸同样在等级B的将近一半学霸会成为等级A,只能导致他自己吃亏。这种复合机制并没有影响结果,但确实对学霸提供了更强的学习激励。</p><hr><p>再来看学生之间同质(homogenous)的情况,作者首先引入了“随机占优(stochastic dominance)”的概念,大致可以理解为,当两个变量都在某个区间段上实现时,随机占优的那个更有可能取到更大的值。借用作者在另外的地方举的例子来说明一下,学霸和学渣去考试,学霸的得分稳定在90分上下,不外乎85。90、95这三种可能,而学渣背下了去年的卷子,不是0分就是100分。那么在90~100这个区间内,学渣的成绩是随机占优的,因为一旦学渣的成绩落在这个区间,那他一定是100分,而学霸的成绩落在这个区间,要么95要么90,都不如学渣高。</p><p>作者由此给出了两个结论:首先是说,如果选择学习一致随机占优选择不学习,那么精细绝对得分打分机制是所有机制中最好的;其次是说,粗糙绝对得分打分机制是最好的绝对得分打分机制的充要条件是存在一个区间划分P,使得在P以外,选择学习一致随机占优选择不学习,而在P的各个划分内,选择不学习随机占优选择学习。证明中有很多数学,虽然我确信我应该是能看懂的,但是这不是咱们的重点,所以也就不再赘述;单从直观上来理解,第一个结论大概是说,如果学习肯定比不学习要好,而且学习的回报通过精细打分体现出来了,所以大家都有激励选择学习;第二个结论则可以通过和分离类型时的现象做个对比,在每一个区间划分内,高分段是高类型而不学习的学生,低分段是低类型而学习的学生,这样就给予了学生学习的激励,因为他们从不学习转到学习,将能进入下一个等级,和一部分高类型(能力水平更高而不学习)的学生获得相同等级;由于整个模型是一个零和市场,这些人所获得的效用正是从那些选择不学习的人身上剥夺来的。</p><hr><p>最后,我们进一步贴近现实,考虑以下三个扩展:不完全信息、异质学生和期中考。</p><p>不完全信息的状况,作者总结道:如果一共有有限种类型(作者此处隐含的假定是每个类型的学生是分离的),且每种类型出现的概率已知,而学生们是从这些类型中随机地抽取了一种类型,那么教授总是可以把这个不完全信息的问题转化为一个连续类型完全信息的问题,只需要使用分离情况下最优的粗糙绝对得分打分机制,把本来每个类型的人数替代为每个类型的概率,然后求对称纳什均衡就可以了。</p><p>有关异质学生的问题,作者做了一个非常有趣的假设:按学生的能力从小到大记为1至N,如果他们学习,那么每道题答对的概率就递进1/N,如果不学习,就和能力水平差一个等级的学生选择学习时一样;再假定一共有K道题。如果我们选择精细绝对得分打分机制,这时候最高能力和最低能力的学生都没有激励去学习,而中间段的学生则很有学习的欲望,如果我们此时引入粗糙绝对得分打分机制,虽然会压制中间段学生学习的欲望,但反之会激励最高能力和最低能力的学生去学习,总的来说提高了整体的最低努力值。不过,这些内容并不完全是理论分析的结果,而是作者所做的一个模拟实验,可以参见他们的工作论文Dubey and Gealakoplos(2005) “Grading in games of status: Marking exams and setting wages”</p><p>对于期中考,作者说的比较简单,只是表达了在有期中考的情况下,粗糙打分更加有必要,相关内容同样可以参阅他们的工作论文。</p><hr><p>(2015.05.15 @ 微信公众号 @ 《第零讲:Grading Exams 评分标准》)</p>"},{"title":"我知道我拿不到特奖","path":"2014/11/03/prize/","permalink":"http://sx349.github.io/2014/11/03/prize/","content":"<p>杨老师说“只有特奖才是真牛,别的都是扯。”</p><p>我知道我拿不到特奖,所以我来扯两句。</p><p>我的GPA从大一开始单调递减;物理学导论期中考,本班(经26)平均88,我只有75,去查卷发现助教多给了12分,第二天在篮球场上接到老妈的电话痛哭流涕,虽然不是第一次,但确实是最深刻的一次怀疑自己;优秀学术人才培养计划面试那天,我从面试场出门的时候其实隐约就有预感自己要跪,果不其然地跪了。虽然大家都在说是老师没眼光,但是我相信这里面肯定有原因,也许不适合学术就是其中之一。如果还要再加上放弃了挑战杯SRT的项目,放弃了今经乐道的比赛,因为感觉压力太大连着两学期都在期中退了一门数双(同时有些人两学期四门数双无压力),我觉得自己对于学习也就只有这样了。</p><p>社工我也做得一般,虽然这学期身兼数职,但大多是因为秋季学期不交换的学生太少的缘故。而在这之前,从组织组打杂,到临到最末一天才报名到调研组当组长(结果也没有做成什么大事,唯一一个大型的调研项目主要是鸡丁在负责),后来又到沟通中心编月刊,手上攥着毕业刊的材料一直没有做出来;而现在,已经有几个思源的小伙伴是副主席乃至主席了,尤其崇拜倩爷整天去跟那些大人物对话,感觉大家在社工上都撵了我十条街了。</p><p>体育就更不用说了,除了我必定会参加的马杯拔河以外,我唯一的另外的马杯经历就是去年的马杯象棋,结果最后证明我不过就是去凑数的(当然,说的好听点,就是田忌赛马的),不仅只赢了一轮,而且那一轮经管是四胜一负,换句话说我去不去完全没有影响最后成绩。每年体测对我来说都是半个噩梦,引体向上从来都是挣扎着做起来几个就已经很满意了,估计如果按全年级来排身体素质,我大概就是在后一百名到两百名这个范围内了。</p><p>到明年这个时候,又是一年特奖,我相信候选人名单里会出现我熟悉的名字,但肯定不会有我的名字。那时候我一定会回想我过去三年在清华到底做了些什么,为什么他们能够读几百本专业书,拿到十几门满分,从学生会到团委到国旗仪仗队到体育专业队,而我却啥都没做呢?甚至也不用等到那时候,如果让我现在回过头想想,估计已经落下这群潜在特奖得主一大截吧~</p><p>但是……遗憾吗?</p><p>其实不遗憾。我只是把他们“夯实基础”的时间拿来做别的事了而已。</p><p>我进大学之前完全不知道什么是PS,没用过Indesign,没听说过Prezi,没剪过片子,没消过音,人人主页还是个新鲜事物。现在我都学会了,虽然我只会用PS做个协会杂志的封面,拿Indesign放几张图放点文字就当作是院刊,拿Prezi搞个花里胡哨的展示,剪片子的时候还是不会让声音淡入淡出,消音之后伴奏里还是会人声超过音乐,我们的人人主页也只活跃了一年。但是我学会了。</p><p>做西方文明和中国文明的笔记,最初的念头只是想做速记,结果一发不可收拾。其实这事儿看起来挺蠢,耗时间,西方文明大概是课上记录和课下整理1:1,到中国文明起码要1:3到1:4,基本上隔了一年我才把中国文明整理完毕。但是我相信,我学到的东西比任何一个只是上了课的同学都要多,比那些考得比我更高的同学也都要多。后来彭刚教授把我的笔记要过去作为他的《西方文明史》的补充,后记里面有一句感谢,我觉得还是挺心满意足的,原本还想要“敲诈”点稿费的想法也烟消云散了。更遑论,就是因为这份笔记的缘故,我在经管结识了许许多多的人,顺带也帮助了一些人(提供了一些不去上课的激励,哈哈),这绝对是比拿到满分更加让人高兴的事。</p><p>我喜欢C楼地下水木的旧书架,每次去必定会买50元上下的旧书,有时候明知道图书馆就能借,或是网上也能买,但是我喜欢那种发现一本自己喜欢的书的感觉。而说道去图书馆,相比今年某位大神借了几百本书,90%是本专业的,我大概90%都不是经济类的书籍,我看小说,看杂文,看一些莫名其妙的理论研究(曾经被人调侃看一本讲解从古希腊到现代各类宗教的《神的起源》,和妹子在琉璃厂买笔墨结果最后带回去一本王力老先生的《汉语音韵》,大致如此),我觉得钻图书馆如果只是为了去找一本有助于自己写论文或是做作业的书,或者泛泛地说,找一本为了自己学习的书,绝不是我所希望图书馆给予我的。</p><p>我花了很多时间在人人上,加上这篇日志,我的日志总数一共是291篇,带编号的杂记到了HD,也就是第192篇,抖机灵的状态估计也写了不下数百条,因为一次四级考试做了半天的网红结果被父母勒令删掉,曾经也一度好友数达到上限不得不一批一批地删好友。水人人肯定是对学习没有帮助的,而且也占用了不少本来可以做别的事情的时间,何况我也不像杨老师或是王教授那样,成为著名禅师或是著名段子手,来访量也一直是不温不火的增长着,就算是“术业有专攻”,我也绝算不上是把人人营销得很成功的那一类。我想,拿到特奖的人,他们大概不会像我这样咋呼地关注着苏格兰公投,或是有事没事写一两篇阅读量都不会破百的日志自娱自乐的吧?但是,这就是我的乐趣吧。就好像李博学姐的“清华小生活”微博一样,只是自己的小生活而已。</p><p>也许有人会说,到清华这样一个国内顶尖高校来上学,难道不就应该学得最顶尖的毕业吗?不就应该每个人都像特奖得主那样勤奋、努力、每门课都力争90+、95+乃至99+吗?不就应该把社工做到最好,从干事一路到部长再到副主席再到院会主席最后争取校会主席吗?不就应该冬练三九夏练三伏,做不了引体向上就三个月狂减40斤魔鬼变身吗?不就应该每个人的毕业论文都拿了SCI,不是去顶尖学校念博士、硕士就是到行业巨头去发挥力量吗?</p><p>但是,特奖也只有10个人而已啊,每门课的成绩curve之后也不是有那么多100啊(还都被特奖的这些人拿了),院会主席和校会主席一届也只能有一个啊。</p><p>我知道我拿不到特奖,所以,过得像个拿不到特奖的人,不行吗?</p><p>不过,杨老师的话说得未必也太满了,拿了特奖的人,就一定比我过得辉煌吗?</p><p>可以一样辉煌嘛,个人的效用函数又不是单峰的。</p><p>(2014.11.03 @ 人人 @ 《杂记HD——我知道我拿不到特奖》)</p>"},{"title":"蜘蛛的房租","path":"2014/07/31/novel07/","permalink":"http://sx349.github.io/2014/07/31/novel07/","content":"<hr><p>3月17日 晴 适宜织网</p><p>伟大的斯拜德之神在上,<br>我愿向您敬献我的蛛网。</p><p>在您的蛛网之光的庇佑下,卑微的埃米斯终于找到了一处合适的居所。按人类的说法,这是一间平房,大约3米(或者说,9.84夫特)高,主要的构成应该是干燥的无腐殖土。尽管土质有些影响,但是卑微的埃米斯认为神的恩赐一定有其道理,空气中的水分很充足,前来接受洗礼的异教徒也非常活跃,看来近期就要有一场大雨,卑微的埃米斯蒙神指引从野外进入这间平房,一定能够在这个即将来临的雨季中生存下去。</p><p>说到接受洗礼的异教徒,卑微的埃米斯感到非常满意。这里的异教徒族群不仅种类很丰富,数量也很庞大,但是由于落脚初定,卑微的埃米斯只来得及在墙角搭起最基本的三角网,而且还差点在指引一只硕大的弗莱从迷途中回归时被它挣破了。卑微的埃米斯将花费一段时间努力加固自己的蛛网,至少要达到坚固的六角网,如此才能让为斯拜德之神传教的地方容纳更多的异教徒。</p><p>辛勤的工作是卑微的埃米斯的天职,只有如此卑微的埃米斯才能得一日之安心。</p><p>顺颂八爪安祺!</p><p>卑微的埃米斯上</p><hr><p>3月24日 雨 适宜传教</p><p>伟大的斯拜德之神在上,<br>我愿向您敬献我的蛛网。</p><p>雨季在等待了六天之后终于姗姗来迟,卑微的埃米斯得以做好万全的准备工作,将墙角的六角网搭建完毕,目前已经开始加固第二层。空气中的水分每天清晨都会在第一层蛛网上摇摇欲坠,这种稍带有腐殖气息的甘露既是最好的饮品,也是吸引异教徒前来朝拜斯拜德之神的圣水。</p><p>卑微的埃米斯在这一周见到了数次这间平房的主人,卑微的埃米斯判断这是一个雄性人类,这是从他的气味初步判断出来的。卑微的埃米斯曾经在野地中生活过,其中许多腐败的野果也会发出这种醇厚的香气,这种香气对于弗莱有着强烈的诱惑力,它们会大量的蜂拥而来,然后围着野果跳起疯狂的弗莱之舞,其中有些会突然摔到地面上。卑微的埃米斯从来不涉足这些不义之食,只有那些凑巧撞在了蛛网上的弗莱,卑微的埃米斯才会向它传授斯拜德之神的教义,指引它从迷途中回归。掉在地上的弗莱则大多成为安特们的大餐,作为斯拜德之神的忠诚的朋友,卑微的埃米斯自始至终都对安特们保持应有的尊敬及友好,对于误入传教地的安特,卑微的埃米斯也按照神的指引将它们送回地面。</p><p>人类对于弗莱的态度真是非常之古怪,他们应该是非常讨厌弗莱的,这是神谕在《弗莱百问》中解释过的,但奇怪的是,他们对于死掉的弗莱也有着同样的厌恶。既然讨厌弗莱,那么对于弗莱的死亡不应该感到高兴和欢庆吗?</p><p>在雨季到来之前的一天,卑微的埃米斯开始将每天的一部分时间用于野外探索,除了摸清附近的地形,以拓展传教之路外,卑微的埃米斯还找到了一些非常坚硬而闪光的圆饼。卑微的埃米斯在很多地方看到这些散落的圆饼,它们大致分成三种,最大的银白色的是银可因,中等大小金黄色的是金可因,小一号的银白色的是小可因。这些可因都是在这间屋子的主人从未涉足的地方发现的,卑微的埃米斯认为,这应该是一种天然出产的作物,因此各采集了一块回来,敬献给伟大的斯拜德之神,祈求神的祝佑,让卑微的埃米斯的传教之路能够更加顺利。</p><p>顺颂八爪安祺!</p><p>卑微的埃米斯上</p><hr><p>3月31日 阴 适宜探索</p><p>伟大的斯拜德之神在上,<br>我愿向您敬献我的蛛网。</p><p>卑微的埃米斯本来还感到略微的失望,伟大的斯拜德之神竟然拒绝了敬献,并且要求卑微的埃米斯将这些圆饼放置在人类能够看见的地方。尽管当时卑微的埃米斯不甚理解,但违抗神的旨意是渎神的行为,卑微的埃米斯是伟大的斯拜德之神虔诚的信徒,必须完成神所授予的任务。</p><p>但是今天,卑微的埃米斯终于理解了神的伟大:卑微的埃米斯将找到的几个可因一一放了出去,其中有一次,不知道是不是错觉,卑微的埃米斯感觉到一种被注视的感觉,很有可能就是那个人类。等到卑微的埃米斯累死累活地搬完这些可因之后,竟然有一块与卑微的埃米斯体型相当的腐殖块落在了蛛网前,伟大的斯拜德之神在上,这样一块腐殖块能够将周围的弗莱都吸引过来,卑微的埃米斯将让更多的异教徒接受洗礼,播撒伟大的斯拜德之神的荣光。</p><p>神所谓的“房租”,卑微的埃米斯并不是太能够理解,应该指的是与人类的一种交换吧?卑微的埃米斯觉得,人类一定也是斯拜德之神的忠诚的朋友,因此才会如此慷慨,为了一些看起来漂亮却不能吃的可因,而赠予卑微的埃米斯这样一份大礼。</p><p>顺颂八爪安祺!</p><p>卑微的埃米斯上</p><hr><p>4月7日 晴 适宜织网</p><p>伟大的斯拜德之神在上,<br>我愿向您敬献我的蛛网。</p><p>卑微的埃米斯又回到了野地,对于人类与斯拜德之神的关系也有了重新的认识,看来人类中也分为两类,其中一类与斯拜德之神是友好的,平房的主人(现在是前主人了)就是这一种,另一类则与异教徒同流合污,企图破坏斯拜德之神的传教事业。</p><p>就在几天前,一名拥护异教徒的人类取代了友好的人类,成为了平房的主人,卑微的埃米斯的传教地也被捣毁,幸好卑微的埃米斯不是第一次经历这样的劫难,才逃得一命。尽管卑微的埃米斯试图再多拿出一点“房租”,但是已经找不到多少可因了;这位新来的人类几乎将整个房间都翻了个遍,找到了许多卑微的埃米斯都没有找到的可因,而且理所当然地收进了自己的口袋;异教徒也纷纷被剿灭,魂归伟大的斯拜德之神。</p><p>幸好,现在雨季也已经过去,在野地中,卑微的埃米斯将继续为伟大的斯拜德之神传教的事业,愿斯拜德之神永远赐福于卑微的埃米斯,让更多的异教徒接受洗礼。</p><p>顺颂八爪安祺!</p><p>卑微的埃米斯上</p><hr><p>(2014.07.31 @ 人人 @ 《杂记HA–蜘蛛的房租》)</p><hr>"},{"title":"数学家的言情小说","path":"2014/07/29/novel06/","permalink":"http://sx349.github.io/2014/07/29/novel06/","content":"<hr><p>数学家改行了,写起了言情小说。</p><p>他熟练地打开Vim,在光标的跳跃中开始构思他的处女作。</p><p>首先要有两个主人公,一男一女……男的是个数学家……什么方向的呢?几何……分析……ε-δ语言……讲微积分可以追得到妹子,也不知道是谁想出来的……复分析学得不错的话,可以去试试丘赛的分析……好像是一位师兄说的?……丘成桐学的是什么方向……证明了卡拉比猜想,以他命名的卡拉比-丘流形……师从陈省身……吾日三省吾身,高乎?帅乎?富乎?……微分几何最高奖……从来没有学懂过,就它了。</p><p>女主角呢……还是数学家?不行,数学家一般也不会找数学家吧……何况,数学系的女生是珍稀生物……找个文科的女生也不错……有多大的可能呢?据说有前辈成功了,但好像是在从数学系转出去之后……或者就是理学院的女生吧,学物理的或是学化学的……物理?化学?</p><p>数学家在屏幕上轮流地打着这两个词。</p><p>物理化学?听起来就像代数几何一样……有个冷笑话,上几何与代数的学生买了一本代数几何的教材……学物理化学的,那该算是学化学的吧……</p><p>数学家打下了“物理化学”四个字。</p><p>两人读了同一所大学,否则可能不太有机会接触……虽然说数学家不是宅男……窝在自习室不能算宅,数学家的事儿!……穿长衫而站着喝酒,排出九文大钱……茴香豆的茴有几种写法?一共有多少种不同的单群?……</p><p>喝酒,这两个人会喝酒吗?……喝酒会让数学家的头脑变得迟缓……数学是一门精确的学科?……数学是确定性的丧失……哪个克莱因说的?……哥德尔不完备性定理,康托对角线,图灵的停机问题……什么模糊数学,多值逻辑,都是捣糨糊……罗素的理发师,包含所有集合的集合,选择公理……选择什么呢?啤酒还是红酒……两个人在一起应该是喝红酒吧……拿着高脚杯,一个完美的抛物线……不对,不够完美,最后收敛成了平行线……</p><p>真的会喝酒吗?……听说有人靠着LSD来支撑自己做实验……凯库勒的苯环,马车背后的蛇,门捷列夫的扑克牌……去年出去开会和同事们打的扑克牌,竟然要卖30块一副……LSD……Longest Shortest Distance……最长的最短路径……树的直径?</p><p>说起来,确实应该要有棵树……那就得有个院子……什么树呢?……梧桐,香樟,水杉……不行,这种树种在院子里也太奇怪了……李树,枣树,梨树……二叉树,AVL树,红黑树,PQ树……应该是一棵可以爬的树,有果子,可以摘下来吃……</p><p>给谁吃呢?最好有个青梅竹马……自家的院子还是大院?……郎骑竹马来,绕床弄青梅……床前明月光,胡床是不是床?……青梅竹马……是谁的?男主人公还是女主人公?……</p><p>数学家努力回想着自己看过的为数不多的几本言情小说。</p><p>就算是男主人公的吧……这一回得是文科生了……中文系,历史系……北大二院五院出CP……交叉学科,元培……中国学?1年制的硕士,写论文也来不及吧……三角地,静园,东区浴室……一边是精神上的,一边是肉体上的……</p><p>不行不行,到现在为止还没有什么波澜……男主人公和女主人公不能光喝酒啊……应该要有争执,有争执就有矛盾,有矛盾就有波澜,有波澜就有看点……君子动口不动手,唯女子与小人难养也……吵起来得是原则问题……0是不是自然数?调和级数不收敛?……怎么不收敛?芝诺的乌龟跑不过拉普拉斯的兔子……爱丽丝也有一只兔子,还有三月兔……红皇后,白皇后……八皇后问题?这个问题可没有争议……得找没有解决的问题……</p><p>数学家很头疼,他感到思路有些卡壳了。</p><p>赶紧收敛回来……怎么收敛?测度收敛,弱收敛,一致收敛……Riesz定理,Egroff定理,还有那些没有名字的乱七八糟的收敛定理……有Chebyshev的份嘛?……俄国人的习题集真厚,吉米多维奇,菲列波夫,波波夫……莫斯科大学,还是叫莫斯科中央大学?……莫斯科不相信眼泪……</p><p>应该让他们一起出现……同一时间,同一地点,同一……三一律,同一什么来着?……男女结伴去莫斯科旅行……还是墨西哥?……莫斯科在西伯利亚,墨西哥在赤道,一个冷,一个热……</p><p>还有那个青梅竹马的……在莫斯科偶遇?那么大的莫斯科……清华的主楼和莫斯科大学是同样的造型……同款!……国母同款努比亚……没钱买,弄个国母同款发型吧……如果是在莫斯科学习呢?这样就有理由了,两个人结伴去拜访,偶遇……学中国学的,跑到俄罗斯去……当初谁签的割地条约?……骂李鸿章签字,配了张曾国藩的图……不是旅行,难道还能是去收复失地?……开什么玩笑……</p><p>还是要去的,不是去旅行……去当访问学者?……俄罗斯学派,靠的主要是直觉而不是严谨的证明……听起来倒像是拉马努金……不世的天才几乎都短命,拉马努金死的时候不超过40岁吧?……菲尔兹奖要发给40岁以下的数学家,是怕来不及拿奖吗……拿诺贝尔奖的七老八十的老头,硬是撑到了颁奖之后……数学界的天才,不是疯子就是短命……阿贝尔没等到柏林大学的聘书,伽罗华没等到刘维尔……</p><p>数学家看了看表,已经写了快一个钟头了,剧情连个头都没有开始。</p><p>但是数学家没有放弃,学数学的人,最大的优点就是有耐心。</p><p>他写了又改,改了又写,就这样过去了很多天。</p><hr><p>数学家改行了,不写言情小说了,改写现实主义小说。</p><p>他噼里啪啦地敲起了键盘。</p><p>“从前有个数学家,他一直找不到女朋友;也不知道听谁说的,写言情小说很容易勾搭女读者;于是,数学家改行了,写起了言情小说……”</p><hr><p>(2014.07.29 @ 人人 @ 《杂记GZ–数学家的言情小说》)</p><hr>"},{"title":"太阳的质量","path":"2014/07/01/massofsun/","permalink":"http://sx349.github.io/2014/07/01/massofsun/","content":"<p>“已知1+1=2,求太阳的质量。”</p><p>这是一个流传已久的题目,初看起来颇为棘手。</p><p>“……首先我们来看已知条件,1+1=2,也许有人会觉得这是基本公理,持有这样想法的人显然对于数学史缺乏足够的了解。对自然数的理性认知(相对于我们常见的发自直觉的感性认知而言),我个人认为可以追溯到1889年意大利数学家皮亚诺发表皮亚诺公理;然而,目前最常用的加法公理的证明,往往都会引用ZFC体系以及20世纪初冯·诺伊曼对于自然数的集合论定义的贡献……”</p><p>他一直觉得皮亚诺公理的第一条应该是“0是个自然数”,不过她不同意,因为在数论中自然数总是从1开始的。这是原则问题,她总是这么说。他觉得这不重要,把0从自然数集合里去掉,不影响他明天早上看到太阳从东边出来。他把他的想法告诉了她,她说:爱一个人,就应该为对方做一些改变,不是吗?</p><p>她终究还是找了一个认为自然数从1开始的人,因为这是原则问题。表面上他好像无所谓,其实他仍旧坚持着自己的原则,所以他不会,也并没有为了她而改变什么,她最后对他如是说。</p><p>“……当然,这样的考虑也许夸大了作者本来的意图,毕竟这里只是单纯采用了1+1=2这样一个形式化的等式。对于自然数的认识,大概远古时期就已经诞生,具体时间已不可考,但是其形式化符号的出现却有迹可循:等于号‘=’由RobertRecorde在1557年发明,因为这个符号由两条等长的平行线构成,鲜明地表达了两个量相等的含义;‘+’作为加号最早出现在14世纪NicoleOresme的手稿中,但是晚至1489年才由JohannesWidmann正式地引入;目前欧洲人所使用的阿拉伯数字,或者更准确地说,印度-阿拉伯数字的拉丁版,据信与早期印刷术同时诞生于15世纪末至16世纪初;综合来看,1+1=2这样一个形式等式的出现,最早也要到16世纪的中期……”</p><p>她在自己的领域俨然是个权威,但是她仍不满足。他惊诧于她的效率,她似乎是块海绵,除了大部分的时间要奉献给阅读文献和专著外,总还能够挤出足够的时间,让她晚上去看一场歌剧,让她周末赶去听演唱会,让她能够来得及和他慢悠悠地吃一顿晚餐;而他的时间只够自己踉跄着赶上老师的进度。</p><p>印度-阿拉伯数字和印度-阿拉伯数字系统是两回事,现在常见的0~9的写法,其实只是欧洲文化殖民的结果;而在阿拉伯地区,仍保留着一支带有典型伊斯兰烙印的阿拉伯数字;这还是她告诉他的。</p><p>他感到自己在不断地积累烦躁,那是一种郁闷、羡慕、嫉妒相互揉杂形成的感受。高负荷并不是人人都能承受的,但是她的焦虑却往往让他歇斯底里,他不知道这是她给的压力,还是自己给自己的压力;他最终选择了出逃。</p><p>“……我们再从另一个角度来看看这个问题。如何求太阳的质量?对于21世纪的人类来说,只要随手查一下搜索引擎就能得到较为精确的数字;当然,如果这个问题出现在高考的试卷上就又是另一回事了。高中生们手中的工具并不比一两百年前的数学家和物理学家更为先进(反过来说也同样成立),他们所用的最基本的武器是牛顿在1687年《自然哲学的数学原理》中发表的万有引力定律,以及亨利·卡文迪许于1797年至1798年期间借助扭秤实验精确测量得出的万有引力常数,还有法国天文学家拉朗德在1771年首次推算出的地日距离。值得一提的是,拉朗德的观测资料是在1769年英法两国交战时取得的,法国政府特别下令海军不仅不得攻击观测用船奋进号,还必须为其保驾护航。对于有心人来说,在万有引力定律发表之后,太阳的质量就不再是个难以解开的谜题,等到1771年最后一环完成,太阳的质量就已经唾手可得……”</p><p>他一度以为他真的要嚷嚷着“Scienceis my lady”来渡过余生,让平时自嘲的“注孤生”一语成谶。科学上的巨匠往往在情场上颗粒无收,牛顿因为科学研究而冷落了与其订婚的药剂师的继女,据说他虽然终生未娶,却始终对这段情愫保留着一段美好的回忆;卡文迪许同样孤独终老,他对于交际的恐惧曾使他留下面对特意寻来的贵客落荒而逃的轶事。</p><p>他周围不乏同病相怜的家伙,但找不到女朋友与找不到女朋友之间,却很难轻易地划上等号;毕竟,脱光这种事情,是绝没有可能传染的。</p><p>“……回顾我们已经提到过的时间点,首先,在16世纪中期,人们就已经能够写出1+1=2这样的形式等式了;其次,太阳的质量在1771年得到了初步的破译。如果问题是在1771年之后提出的,那么简直可以说是毫无意义,毕竟这个数字已经是明摆着的了。排除这样一种可能性之后,我们不妨做个大胆的猜想,这个问题的提出其实是在16世纪中期至1771年之间,而在历史上,恰好有一件事情与此相契合:那就是文艺复兴。如果事实果真如此,那么问题的提出者其实并不在乎最终太阳的质量是多少;这个问题背后的深意在于,太阳不再被视为一个神祇,它和地球、甚至和人类一样,都有着其基本的物理属性——质量。文艺复兴时期是一个神陨落而人诞生的时代,是一个信仰破灭而公理建立的时代,由最基础的等式1+1=2开始,破除对于以太阳神为代表的众神的敬畏,科学颠覆宗教的大幕正在被缓缓地拉开……”</p><p>他想要的是一段平淡的爱情,一个人与一个人的爱情。他很清楚这一点,但是要做到,实在太难。</p><p>在爱情之中,总会有人不自觉地成为神,完美,没有瑕疵,让信徒心甘情愿地奉献一切,然后呢?或许是放逐吧。文艺复兴带来的是人的精神的复苏,但是神又何尝没有潜藏在人的灵魂之中,背影之后?</p><p>笛卡尔的二元论其实并没有摆脱神的成分,将灵魂视为一种与肉体同等高贵的事物,实际上是为规则创造神格,以规则造神。他对怀抱中的她讲述着他的观点,她微微地蹩着眉,不知是没听懂,还是表示对于挑战正统理论的不满与反对。</p><p>唉呀,不说了,都是我瞎想的。午饭想吃什么呢?</p><hr><p>(2014.07.01 @ 人人 @ 杂记GY–已知1+1=2,求太阳的质量)</p>"},{"title":"瞎扯·2014·二","path":"2014/06/30/tweet-2014-2/","permalink":"http://sx349.github.io/2014/06/30/tweet-2014-2/","content":"<p>《瞎扯·2014·二》为《瞎扯》的第七辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>@致我们呼啸而过的青春:<br>清晨想到的第一个人 和夜晚想到的最后一个人 不是让你幸福的人 就是让你痛苦的人<br>(2014.05.18)</p><hr><p>刷完《当我们变成一堆数字》,看到一系列NB的数据分析公司在码农或非码农的手里被打造出来,最深刻的体会是:之所以“只缺一个码农”会被嘲笑,关键只是你想的Idea连码农都能想到,这样的Idea绝对是赚不到钱的啊……<br>(2014.04.27)</p><hr><p>时间满怀着恶意从我头上缓缓流过,让我觉得每一分钟都是如此漫长。——《秒速五厘米》<br>(2014.04.19)</p><hr><p>@酒話:<br>你喜欢一个人会怎么样” “看他发的状态就像在做阅读理解”<br>(2014.04.07)</p><hr><p>@我想谈场永不分手的恋爱:<br>“你能模仿下啄木鸟吗?” “怎么模仿?” “就拿我的脸当树.”<br>(2014.04.04)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>怪不得我们不怎么用北大教材<br>@孟祥溪.bme:<br>北大科研部长周辉:“编书人的水平都是很低的,你考的分数越高,就和他的水平越接近。”<br>(2014.06.30)</p><hr><p>@李欣宜:<br>今早在作业研究所发生了骇人的一幕。王后雄见袁莲花在场,于是拿出一本作业,大声喊话:“你们买不买?不买就没的做了!” 袁莲花赶紧跪下来求王后雄不要撕掉,王后雄朝袁莲花喊道:“全套王后雄,买不买?1000一套!没答案、没解析!不买我就撕了” 说完,袁莲花含泪买下作业,研究所其他正在默默做题的教授也纷纷放下了手中的作业强忍着泪水买下了一套又一套作业。王后雄拿得钞票,向围观的葛军和姚天行挥舞。围观的做家起哄叫好,发出“做!做!”的喊声,还有人竖起大拇指嘉奖教授们对作业的爱惜和善做。<br>(2014.06.24)</p><hr><p>所以最后不顾这些孩子的生命用了球状闪电……?<br>@余额宝yui:<br>球状闪电里有一幕到今天还记得:一群极端环保主义者为了反核绑架了一群学生,然后要炸核电站。<br>(2014.06.22)</p><hr><p>没有本科学历还想出家?<br>@P大树洞:<br>好痛苦,像我这种学渣来北大有什么意义?每天混日子而已,什么都没学到,考前看一看,分数低得不堪入目,又不甘心,然而并不想做什么来改变……还有救吗?当初不如直接出家算了。//from Web<br>(2014.06.10)</p><hr><p>在这报考季来黑一发北大吧(其实也说不上是北大= =)在校医院偶然遇见一位大爷一位大妈在闲聊,大妈说:我是52年进清华的;大爷说:诶我也是啊,怎么没见过你;大妈说:我是先去了燕大,有个叔叔在清华,我就过来了,不过倒没有靠关系,是因为成绩好;大爷说:那真巧,那会儿燕大先发了榜,我就先去了燕大,学费都交了,上了一个礼拜的课,清华发榜了,我就来清华了。(摊手)<br>(2014.06.09)</p><hr><p>@贝肯鲍尔:<br>在世界杯历史上,仅有巴西、土耳其、哥斯达黎加三支国家队战胜过中国国家队。<br>(2014.05.31)</p><hr><p>焦裕禄式的干部。<br>@吴科.chm: 刚刚新闻联播播放了村干部段爱平,自己捐钱给学校,自己儿子却还要还房贷,雨天一直在外抢修给村民修房子,自己的房子却在漏雨,已经得了淋巴癌,却不在医院而跑出来上班,对于朝廷台树立的这种典型,我感觉很惊悚。。<br>(2014.05.17)</p><hr><p>@ANTI历史震惊你:<br>【女孩扮学生乞讨 90后城管出数学题拆穿骗局】已知全集U{1、2、3、4、5、6、7}A{2、4、6}B{1、3、5、6、7}C{6},求A∪(B∩C)的集合。一个女孩假扮高中生跪路边乞讨,毕业于闽南师范大学数学系的90后城管,认为是骗子,出数学题试她。女孩做不出,默默收拾东西走了。——————————今天在手机网易新闻上看见这个新闻,再去看了评论,一大半网友表示自己也不会做,我为我以前经常和网易评论者们过不去感到愧疚。扶助关爱社会弱势群体人人有责,你们以后都不要黑人家了。<br>(2014.05.14)</p><hr><p>@王逸超:<br>“佩妮!佩妮!佩妮!”“天哪,我的老伙计,你一定要每次敲三下吗?看在上帝的份上,饶了我的门吧!小心我用靴子狠狠地踢你的屁股!我发誓我真的会这样做!”<br>(2014.04.29)</p><hr><p>其实是门通识课吧。<br>@THU日2: 恋爱是门必修课,有人总是中期退课,还有人就tm没选上过<br>(2014.04.27)</p><hr><p>张小平:(讲四色问题)但是不能有飞地啊。比如美国的例子,美国是中国的飞地…啊不是,阿拉斯加是美国的飞地……<br>(2014.04.15)</p><hr><p>辩论小组赛结束,突然想到之前讨论时的一个梗:昌婷@卢昌婷 说咱们找学霸@周恩泽 作为卧底打探一下计算机系的立论吧,大家纷纷表示学霸一定会持中立态度。但是,学霸到底会持哪种中立态度呢?是守序中立:我不告诉这边,也不告诉那边;还是邪恶中立:我既告诉了这边,又告诉了那边;还是混乱中立:我告诉两边的都是错的?<br>(2014.04.07)</p><hr><p>@修新羽!!!:<br>闲下来一定要在清华组织个行为艺术协会……随心所欲地做稀奇古怪的事情,在操场中间的草坪上大声背诗啊,在图书馆前的水池撒花瓣一边撒一边自言自语“他爱我他不爱我他爱我”啊……每天花一小时逢人就问“同学你愿意喝下那杯醉生梦死吗”…约好某天从教学楼回宿舍大家都沿着马路中间那条白线骑车啊……【人生在世一件很重要的事情就是先意识到自己最真实最奇葩的地方,再意识到每个人都是独特的,这样一来即使自己“奇葩”了也不会感到孤独。】<br>(2014.04.05)</p><hr><p>(看见小雨回答建功说“屈指可数”之后自动脑补:)各位同学,今天来上课的人这么少就算了,竟然真的有同学在数来了多少人;数来了多少人也就算了,这位同学也用不着把自己的鞋脱了来数吧?好吧,你脱了自己的鞋也就算了,你也用不着把边上的那个同学的鞋也脱了呀……(其实小雨一只手有21根手指……)<br>@夏乔雨: 中外艺术史一共来了41个人 呵呵呵呵<br>(2014.04.04)</p><hr><p>@口袋英语:<br>某些外企白领说的“中文”:这个project的schedule有些问题,尤其是buffer不多。另外,cost也偏高。目前我们没法confirm手上的resource能完全take得了。anyway我们还是先pilot一下,再follow up最终的output,看能不能run的比较smoothly,更重要的是evaluate所有的cost能不能完全被cover掉。<br>(2014.04.03)</p><hr><p>张小平:“你们看Huffman这个长者…啊不这个学者和他的这个Huffman树…”<br>(2014.04.01)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>授人以氢,考人以氕氘氚……(话说高中化学不就是这么考的么……)<br>@陈景莉❤: #转自杨震环# “授人以鱼,考人以鱽鱾鲀鱿鲃鲂鲉鲌鲄鲆鲅鲇鲏鲊鲋鲐鲈鲍鲎鲝鲘鲙鲗鲓鲖鲞鲛鲒鲚鲜鲟鲔鲕鲑鲧鲬鲪鲫鲩鲣鲨鲡鲢鲤鲠鲥鲦鲺鲯鲹鲴鲶鲳鲮鲭鲵鲲鲰鲱鲻鲷鲸鳋鳊鳁鳀鲾鲼鳈鳉鳃鳄鲿鳇鳂鳆鳅鲽鳌鳒鳎鳏鳑鳐鳍鳘鳛鳕鳓鳙鳗鳚鳔鳖鳜鳟鳞鳝鳡鳠鳢鳣鳤。”<br>(2014.06.17)</p><hr><p>@石小田:<br>呆在石家庄机场,忽然想起一件往事。当年论坛上有人感慨:时间与隔阂产生美。一句“吾乃常山赵子龙”,听着英气勃发;你要换成现在的地名,来一句“吾乃石家庄赵子龙”试试? (@马伯庸)<br>(2014.06.09)</p><hr><p>突然想到“Current Liability”里面有一项叫作“Long-term debt due within a year”<br>@刘喆汀:<br>In the long term, we are all dead. However, in the short term, no zuo no die~</p><hr><p>今日大风又起……微积分告诉我们:“连续不一定可倒,可倒一定连续”……而实变函数告诉我们:“单调不连续,则几乎处处可倒”(Lebesgue定理)(所以这就是为什么没有靠在一起的自行车也倒了么……因为它们太单调了…………)<br>(2014.05.27)</p><hr><p>西西弗斯用python写一个项目,但是,每到夜晚,所有的缩进都会消失……<br>@程序猿每日一黑:<br>西西弗斯被伐用 Perl 写一个项目。但是,每到夜晚,所有的注释都会消失。结果,每个黎明,西西弗斯都要从头开始。<br>(2014.05.23)</p><hr><p>瞬间飘过“欲练此功,公司金融”……<br>@王怡人 irene: 给跪…也许我大电子可以是厚德载物,电子电路<br>@马鸣: 厚德载物,高等代数!<br>@物理系学生会:<br>自强不息,数学分析…给你大数学系马杯开幕式口号跪了…<br>(2014.04.27)</p><hr><p>Bond, Zero Coupon Bond. (我叫邦德,赛罗·库旁·邦德)<br>(2014.04.04)</p><hr><p>明明是服务器响应的错误,电脑躺枪= =剧情其实应该是:本机=某ds,服务器=女神,然后……<br>@笑得肚子疼:<br>有时候觉得,电脑就像一个孤高的女子。代码400,是她冷冰冰地说:“我听不懂你在说什么”;401,是她无情地转身:“我不认识你,别说那些奇怪的话”;403,是她残酷的拒绝:“我听懂你的话,也认出你的脸,可我不爱你”;404,是她紧闭心门:“我这儿没有你想要的东西”;503是“呵呵我去洗澡”。<br>(2014.04.04)</p><hr>"},{"title":"战国野史之小国的灭亡","path":"2014/04/13/novel04/","permalink":"http://sx349.github.io/2014/04/13/novel04/","content":"<hr><p>公元前359年,秦孝公用商鞅变法。</p><p>商鞅为了取信于民,在国都市集的南门立一根三丈之木,声称如有人能将其搬至北门,奖励十金;人们觉得很古怪,都不敢去搬;商鞅于是把奖励提高到五十金,有一个人试着把木头搬了过去,果然拿到了五十金。商鞅又声称如有人将木头从北门搬回南门,同样奖励五十金,又有一人上前搬走了木头,也拿到了五十金。于是众人纷纷信服。</p><p>秦国在商鞅变法之后国富民强,使得周围小国纷纷效仿。</p><p>有一小国,在市集南门立一根三丈之木,声称如有人能将其搬至北门,奖励十金;结果行人纷纷上前,往返于南门与北门之间,国库很快挥霍一空。</p><p>又有一小国,为了防止来人过多掏空国库,规定一日只有十个名额;于是民众尽日争夺这十个名额,不顾正业,国家因缺粮而在别国入侵时献城投降。</p><p>又有一小国,因国库承受不住压力,将奖励降为五金,民众不满而上街抗议,别国敌对势力趁机制造内乱,最终灭亡。</p><p>又有一小国,为了减缓国库压力,国家不得不提高税率,习惯了轻傜薄赋的民众纷纷外逃,人口骤减,国家名存实亡。</p><p>又有一小国,由于大量货币涌入市场,发生严重的通货膨胀,物价飞涨,民不聊生,民众发动起义,推翻了国君的统治。</p><p>又有一小国,坊间流传“拿到十金的人都被偷偷地干掉了,家产充归国有”,又有人说“拿到十金的都是安排的托”,于是民众对国家政策半信半疑,不久敌国入侵,国家征兵政策反响平平,不少民众充当带路党,国都被破。</p><p>又有一小国,以实物券形式支付酬劳,民众至商贾处交割时,商贾以国家拖延付帐为由拒绝提供实物,双方一言不合引发争斗,国家分裂为无产阶级与资产阶级两部分,内斗中被邻国吞并。</p><p>又有一小国,有谋臣接连上书陈说南门立木的害处,君主不听,怒而杀之,几名大夫不忿,密谋后联合杀死国君,因谁来作国君而自相残杀,国家因无元首而陷入混乱。</p><p>又有一小国,民众觉得国君有如此做法,必然难以在战国立足,举家搬迁者十之八九,于是国家灭亡。</p><p>终于,有一小国认识到,光学商鞅南门立木的做法是没有用的,更重要的是学习其变法的内容,励精图治,才能国富民强。于是这个小国接受了商鞅《垦草令》重农抑商、改革户籍制度实行“连坐”、允许土地私有及买卖、烧《诗》《书》尊法家、重刑厚赏、推行郡县制等等政策,最终……</p><p>因为离秦国靠得太近,很快就被灭掉了。</p><hr><p>(2014.04.13 @ 人人 @ 《杂记GX——战国野史之小国是怎么灭亡的》)</p><hr>"},{"title":"未若柳絮因风起","path":"2014/04/09/novel05/","permalink":"http://sx349.github.io/2014/04/09/novel05/","content":"<hr><center>白雪纷纷何所似? 未若柳絮因风起。 ——题记</center><hr><h1 id=\"PART-1-柳絮\"><a href=\"#PART-1-柳絮\" class=\"headerlink\" title=\"PART 1 柳絮\"></a>PART 1 柳絮</h1><p>年复一年,暖春的四月里,柳絮如约而至,被风所裹挟,在半空中转着圈。<br>他刚搬到她的对门,都是八九岁的年纪,都上三年级,都喜欢看书。<br>她读的是《诗经》、《老子》、《世说新语》,他则珍藏着一整套《十万个为什么》。</p><p>“你知道什么叫PM2.5嘛?”<br>“不知道,但是我妈昨天跟我说,今天PM2.5太高了,不让我出去玩。”<br>“那我告诉你,就是我们吸气的时候,会把一些很小的东西吸进来,这些小东西只有2.5微米那么大,所以就叫PM2.5啦!”<br>“什么叫微米啊?”<br>“微米是长度的单位,比咱们平时用的厘米、毫米还要小。”<br>“那什么叫PM啊?”<br>“呃……不知道……反正就是小东西的意思吧。”<br>“哦,这样啊。你懂得好多啊!”<br>“没有啦,这是书上看来的。”<br>“但是我读的书里面都没有这些东西。”<br>“那你读的书真没意思。”<br>“哪有!我读的书可有意思了,我给你念一段?”<br>“好啊。”<br>“我正好看到这儿:东晋的名士谢安和他的侄子侄女一起赏雪,谢安出了上句:白雪纷纷何所似?他的侄子谢朗回答说:撒盐空中差可拟。侄女谢道韫则说:未若柳絮因风起。”<br>“哈哈,哪有拿撒盐和下雪的时候比的。”<br>“对啊!你看,很有意思吧?”</p><hr><h1 id=\"PART-2-白雪\"><a href=\"#PART-2-白雪\" class=\"headerlink\" title=\"PART 2 白雪\"></a>PART 2 白雪</h1><p>今年的初雪来得有些晚,但看着皑皑白雪覆盖了清华园的每一个角落,总是一种赏心悦目的体验。<br>他和她走在绿园的小径上,周围的樱树点缀着白雪,好像又回到樱花漫天的时节。</p><p>“好漂亮的雪啊!我都不想回去赶论文了。”<br>“你在写什么论文呀?”<br>“科学哲学,一门选修课,期末要交一篇论文。我写的题目叫作《发展与冗余》。”<br>“科学哲学?”<br>“我也搞不清楚,反正上课的老师特别能扯,每节课都讲一些科学发展的小故事。”<br>“听起来有点像我们的‘科技史讲座’。”<br>“也不是,他讲得特别人文。我的论文题目就跟他上次讲的一本电影有关系,就是阿西莫夫的《I, Robot》。”<br>“那部电影我看过,中文名是叫《智能叛变》来着?”<br>“对。我的观点就是,在社会发展的进程中,伴随着各种形式的‘冗余’,这些冗余有一部分简化了我们平时的日常工作,还有很大一部分都是无效的内容。但是就像电影里所描述的那样,机器人利用自己编程语句中的无效语句发生了自主进化,社会发展中看似冗余的部分,也许有一天反而会成为社会进一步发展的动力。”<br>“呃……大概听懂了……”<br>“算了算了,没指望你这个工科脑袋能听懂。你最近在忙什么?”<br>“这不马上要期末考试了么,据说结构力学考试特别难。”<br>“结构力学?”<br>“啊,文科生不需要理解。”<br>“哼,敢跟我这样说话,皮痒了是不?”<br>“我说了你也不懂嘛。比方说吧,现在你面前的这棵樱树,上面盖了这么厚的一层雪,结构力学会让你算一下它在枝杈的地方应力有多大,会不会断掉之类的。”<br>“无聊无聊无聊……”<br>“也不一定无聊,上次上课老师给我们讲过一个很有趣的例子。你觉得一般的雪花有多大?”<br>“不一定,李白不也写过‘燕山雪花大如席,片片吹落轩辕台’么。”<br>“这有点夸张了。雪花是冰的晶体,从空中落下的时候,它既要保证在一定速度下空气的浮力和阻力、重力相抵消,又要保证在有风的情况下晶体的结构不会被破坏,这里面就涉及到结构力学的问题。”<br>“这不还是很无聊嘛。”<br>“真的吗……我还觉得挺有意思的呢……”</p><hr><h1 id=\"PART-3-烟尘\"><a href=\"#PART-3-烟尘\" class=\"headerlink\" title=\"PART 3 烟尘\"></a>PART 3 烟尘</h1><p>联合国大会的会场上,人们摘下整日带着的防毒面具,呼吸着空气清洁器制造的新鲜空气。<br>窗外一如当年的暖春时节,但飘飞着的却是一团团染黑的“柳絮”,它们在半空中猖狂地遨游。<br>他和她作为特邀专家来到了大会现场。</p><p>“……在我们日常的生活中,经常能观察到这样的情况:长期积累的灰尘会自发地形成网格状或者片状,这是微小颗粒在外加场和相互作用下自发的结果。这样的结构本身是很不稳定的,只要一吹就能够被彻底破坏。但是,在近几年捕获的‘柳絮’状烟尘中,我们检测到几种具有催化剂性质的物质,这种物质不符合我们平常对污染物的定义,但它们能够使灰尘和空气中的硫、氮化合物发生反应,改变灰尘之间相互作用,使灰尘构成类似富勒烯的球状结构。这种结构不仅增强了灰尘之间的结合作用力,同时又在最大程度上降低了其在高风速下的应力,并且由于其内部存在笼状空洞,有可能通过内部结构的动态改变,借用空气对流涡流产生的升力,扩大其传播的范围……”</p><p>“……在一般工业工程设计的过程中,往往都考虑到了收益、产率和可见的污染,而对中间阶段的杂质处理没有给予足够的重视。随着工业产品的多样化,中间阶段的数量不断增加,在中间阶段产生的杂质种类数量更是以更高的量级不断增长,这其中不乏为了简化生产而进行的冗余操作和由此带来的冗余杂质。由于这些杂质一般都在传统上需要考虑的数量级之下,因此只有很少一部分在工业进程的控制范围之内。尽管如此,随着冗余发展理论的应用不断扩大,在工业工程设计方面,我们发现,即使只有很少的、几乎可以忽略的数量的杂质,当其中包含的杂质种类达到充分大的数量级时,发生‘杂质失控’的概率几乎是百分之百。根据理论的量化估计,我们调查了几家企业,它们都具有工艺流程极其复杂、中间产物数量极多的特点,从其中部分企业的排放物中检测到了造成‘柳絮’状烟尘的催化物质,这与根据气象学特点反推烟尘来源的结果基本相符……”</p><p>“好美的柳絮。”<br>“对啊,真的好美。”<br>“老子说过,理想的社会就是‘鸡犬相闻,老死不相往来’。男耕女织,自给自足,不需要科技,不需要工业,全靠自己的一双手。”<br>“但人们回不去了。”<br>“真的回不去了吗?”<br>“冗余发展理论说到底就是熵增原理。最早提出智能具有负熵特征的人没有想到,在现代社会的发展中,人的智能的推动作用远远超出了阻碍作用,结果必然是不断加剧着整个结构的混乱度。”<br>“但是人类并没有做错什么呀。”<br>“不,人类只是什么都没有做而已。”</p><hr><p>(2014.04.09 @ 人人 @ 《杂记GW–未若柳絮因风起》)</p><hr>"},{"title":"瞎扯·2014·一","path":"2014/03/31/tweet-2014-1/","permalink":"http://sx349.github.io/2014/03/31/tweet-2014-1/","content":"<p>《瞎扯·2014·一》为《瞎扯》的第六辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>Here lies one whose name was writ in water。<br>(这里躺着一个人,他的名字写在水上。)<br>——济慈墓志铭<br>(2014.03.29)</p><hr><p>所以原po你在做什么……<br>@汪成:<br>世人如果能忘记海子,忘记卡夫卡,就好了。因为他们生前不愿被人记起,死后也不愿留下虚名。他们所有的文学理念,一生的情怀,都已在自己短暂的生命中完成。无需后人画蛇添足。<br>(2014.03.27)</p><hr><p>“when subjects play zero-sum games repeatedly, individual players often choose different pure strategies over time. But they seem to mistake alternation for randomization; that is, they switch their choices more often than true randomization would require.” (选自Dixit ‘Games of strategy’ )<br>(2014.03.11)</p><hr><p>@史宗恺:<br>当我们经历痛苦和挣扎,甚至煎熬的时候,我们仍需坚持与坚守,保有我们的心怀意念,记得我们最初的梦想,不随波逐流,不放任自己。<br>(2014.02.24)</p><hr><p>原句明明是托尔斯泰的《安娜·卡列尼娜》的吧……<br>@经典电影台词:<br>幸福的人家彼此都很相似,可是不幸人家的苦难却各不相同。——《刺猬的优雅》<br>(2014.02.23)</p><hr><p>读季羡林《清华园日记》,再想起后来成为一代大家,不由想到:就好像FIFA里面的球员,就算一开始再烂,等他踢到30岁的时候,上个80的总评总是有希望的,如果踢得好,总是让他上场,也未必不能上90;但还是有些人到二十几岁就能世界顶级的。<br>(2014.02.18)</p><hr><p>@ChaseFuture:<br>当你老了,回顾一生,就会发觉:什么时候出国读书、什么时候决定做第一份职业、何时选定了对象而恋爱、什么时候结婚,其实都是命运的巨变。只是当时站在三岔路口,眼见风云千樯,你做出抉择的那一日,在日记上,相当沉闷和平凡,当时你还以为是生命中平凡的一天。<br>(2014.02.18)</p><hr><p>@二十几岁的你应该知道:<br>我们不应为某件事情过分高兴或者过分悲伤,原因之一就是一切事物都在改变,另一个原因是我们对于何为有利,何为不利的判断是虚幻的。因此,几乎每个人都曾经一度为某件事情悲伤不已,但最后那却被证明是一件天大的好事。又或者,我们曾经为之兴高采烈的事情,却变成了我们极度痛苦的根源。—— 叔本华<br>(2014.02.18)</p><hr><p>“Economists working in the Marxian-Sraffian tradition represent a small minority of modern economists, and that their writings have virtually no impact upon the professional work of most economists in major English-language universities”, according to George Stigler.<br>(2014.01.15)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>@孙画画:<br>刚才叉院大神@贾志鹏 问我怎么找女票,我说加社团什么的找机会,他说没时间,我说那你也一定没时间谈恋爱了。<br>(2014.03.30)</p><hr><p>@青年历史发明家:<br>“这座山受到了精灵的加护,只要你往上爬就能获得巨大的能量”“喔,听你这么一说确实感觉全身充满了能量,是什么能量呢”“势能!”“势能……”<br>(2014.03.28)</p><hr><p>@邢畅XC:<br>谁转过2048那个游戏麻烦@我一下……我昨天才知道那个游戏是要把方块上的数字打到2048而不是分数正好是2048<br>(2014.03.20)</p><hr><p>@王嘉兴:<br>今天带外校同学吃饭,路过学堂路的时候…一妹子看到了足球赛的宣传板,然后说,“你们学校的体育赛事是以每年的生肖命名的嘛~” 笑疯了 [ zz ]<br>(2014.03.18)</p><hr><p>@陈绮贞:<br>各国论坛对MH370的阴谋论:中国论坛一水儿的在怀疑美国下大棋;美国论坛在指责索马里海盗和基地组织;印度论坛在猜测中国为进入印度洋在演苦肉计;斯里兰卡人叫嚣着要荡平泰米尔人,说是泰米尔人给飞机加了油;巫统和安瓦尔在马来西亚的论坛上相互扣帽子;越南人哀叹国家衰弱到连马来西亚都敢骗它;以色列人推断伊朗两个人是特工;伊朗人立即指出阿富汗的塔利班最有嫌疑;欧洲人注意到飞思卡尔居然有20多员工在一架航班上,十分惊奇,觉得他们一定是在执行秘密任务;俄国人:“别转移注意力,把克里米亚还给我们!”<br>(2014.03.17)</p><hr><p>结合前一段时间Yao的真理:不努力虽然赚的少一点,但是赶工作的时候会累一点嘛!<br>@福报:<br>努力不一定成功,但是不努力会很轻松哦!<br>(2014.03.01)</p><hr><p>@陈一舟:<br>美国海关:先生,我们遗憾的通知您,我们不得不拒绝您入境。您从北京来,行李里连个口罩都没有,我们有理由怀疑您没打算回去。 中国海关:先生,我们遗憾的通知您,我们不得不拒绝您入境。您从美国来,行李里连个口罩都没有,我们有理由怀疑您没打算活着回去。<br>(2014.02.26)</p><hr><p>@超级课程表:<br>刚刚看到一个段子,“一个姑娘问我,搞学术的为什么都睡得那么晚,难道只有到晚上才有灵感?不是,姑娘,搞学术不靠灵感,靠的是碌碌无为的白天引发的愧疚心。”(太对了……分分钟撕破了伪学霸的真面目)<br>(2014.02.26)</p><hr><p>寒假前,钱YY来30号楼视察,惊呼:还有比这更糟糕的生活环境吗?!于是春节结束后施工队入驻。如果钱YY今天再来视察的话,一定会由衷地对施工队的各位大哥们说:你们辛苦了,我终于知道,真的有比那更糟糕的生活环境……<br>(2014.02.25)</p><hr><p>@国雨薇:<br>学胖,只要学习就必须吃东西,否则浑身难受,最后成绩长没长不知道,体重是一定长了吃饭。。谁中枪╭(╯ε╰)╮<br>(2014.02.21)</p><hr><p>今天被人问到一个问题:你平时听歌是用耳机还是麦克风?……你给我用麦克风听一个试试!<br>(2014.02.20)</p><hr><p>莫名想到上次去西北旺采访的时候为骗认同自称是北财的,结果竟然采访到一个北大的……(幸好没说自己是T大的,否则估计就不接受采访了= =)<br>@李梅皓 no mercy:<br>清华学子:同学你是哪个学校的? xxx:理工的。 清华:哦,我是清华的~ xxx:你好厉害。 清华:其实也不算难啦~你当初要是认真点学也能考上的。对了,你是大连理工还是北京理工? xxx:麻省的。<br>(2014.02.19)</p><hr><p>其实离“餐桌上有只鸡就要把鸡腿夹给某人的年纪”也不远了吧……<br>@韩澍Myth:<br>我已经过了餐桌上有只鸡就一定能吃到鸡腿的年纪。<br>(2014.02.16)</p><hr><p>@设定控:<br>@Solidot: 【中国开发者推出马克思币】一个中国团队宣布了一个新的山寨币——马克思币(白皮书PDF)。<br>(2014.01.30)</p><hr><p>@黄鑫鹏:<br>第一眼看成了李泽<br>@冯悦∫volGLkamp:<br>李泽厚抽桑德尔的新书即将由三联书店出版。<br>(2014.01.25)</p><hr><p>@侯泽华:<br>My goal for 2014 is to accomplish the goals of 2013 which I should have done in 2012 because I promised them in 2011 and planned in 2010.<br>(2014.01.18)</p><hr><p>@钱虹良:<br>室友们渐渐考完了,感觉就像成功进化成人的三只猩猩,用充满同情的眼神看着他们昨天的玩伴——那只还在努力进化的猩猩,从他们给我的香蕉中我读出了无限的同情和当我变人后把门锁好整理完卫生再走的期待。<br>(2014.01.18)</p><hr><p>@中国工程院作业工程苏州研究所:<br>名人都有高端大气的外号。袁莲花教授也打算给自己起个响亮的外号。大家纷纷献计献策:“作业界的拿破仑”“作业界的普京”“作业界的马云”“作业界的梅西”“作业界的利威尔”……袁莲花教授脸色铁青:“你们是不是在黑我的身高?”<br>(2014.01.01)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>谁说只有一本了……在有多本的情况下二分算法直接退化。<br>@王锐:<br>早晨一女生背着一堆书进了阅览室,结果警报响了,管理员大妈让女生看看是哪本书把警报弄响了,那女生把书倒出来,准备一本一本的测。大妈见状急了,把书分成两份,第一份过了一下,响了。又把这一份分成两份接着测,一会儿就找到了,大妈用鄙视的眼神看着女生,说:”O(n)和O(logn)都分不清”<br>(2014.03.31)</p><hr><p>想到了这道题:A声称B否认C说D是说谎的,其中BCD说谎概率为1/3,那么A说谎的概率是多少?<br>@唐德源manju:<br>中翻英:马航否认之前否认交通部长否认过的由大马军方否认的否认消息。<br>(2014.03.16)</p><hr><p>@向重洋:<br>本人今年虚岁21i。<br>(2014.03.14)</p><hr><p>张小平:“这个dijkstra是荷兰人。河南人?他长得像河南人嘛?!”“把你们关在小黑屋里面,不给你们饭吃,你们能搞出这个旅行商问题/最短路径的算法吗?搞不出来?你们又跟从你们的内心了,上面一个从,下面一个心。”<br>(2014.03.11)</p><hr><p>@左晗君:<br>“这个模式匹配方法是由Brute和Force提出来的,所以被称为B-F算法”<br>(2014.03.03)</p><hr><p>与各位分享一个冷笑话:我刚刚把自己的眼睛变成了一个“与”门,具体操作方法是:在坏了的路灯快速地一闪一闪的同时随机地眨眼,观察路灯的亮(1)与暗(0)。很想知道怎样把自己变成一个“异或”门……<br>(2014.02.23)</p><hr><p>@李泽峰:<br>终于知道当年鲍林为何要把氟的电负性定为4.0了,原来他毕生的梦想也是满绩<br>(2014.02.20)</p><hr><p>@化学:<br>索契成环反应(Sochi annulation)的产率为80%。@郭谦益<br>(2014.02.08)</p><hr><p>@李旭昇:<br>【钓】为什么美国一直不敢惹中国?因为中美之间隔着国际日期变更线,中国今天发射一枚导弹,昨天就打到美国去了,根本来不及拦截。<br>(2014.02.08)</p><hr><p>马原政经的逻辑:(1)贸易保护是生产过剩的表现(别看我,这是别人写的论文= =),所以多么多么不好,发达国家贸易保护阻碍发展中国家自由贸易……(2)自由贸易是资本主义强加给殖民地来赚钱的,幼稚产业保护是资本主义国家发家致富的根本,自由贸易不是万能的,国家干预贸易保护才是发展中国家应当采取的策略……<br>(2014.01.17)</p><hr>"},{"title":"人与路","path":"2014/03/03/novel02/","permalink":"http://sx349.github.io/2014/03/03/novel02/","content":"<hr><h1 id=\"一\"><a href=\"#一\" class=\"headerlink\" title=\"一\"></a>一</h1><p>年轻的时候,我特别喜欢到处乱跑。</p><p>在大西北的草原上,我第一次遇见了那个人:他一边走,一边脚下就出现了路。</p><p>其实这话说得不对:我只遇到了他一次;之后每次我遇到的,只是跟他穿了同一双鞋的人罢了。</p><p>那是一个牧羊人。我见到他的时候,他骑在一只羊的背上,他身后的草原横七竖八地交错着一条条土路。</p><p>那双神奇的鞋是被他家的羊从地里拱出来的。正好原来那双鞋已经破烂不堪,他就索性换上了这双新鞋。结果麻烦来了。</p><p>他走到哪里,地上的土路就铺到哪里,第一天下来,草原上就好像被划了十几条杠杠。</p><p>他舍不得鞋,又不想把草原给毁了,于是他索性每天骑着羊赶羊。</p><p>我觉得这双鞋特别神奇,想跟他买下来,但是被他一口回绝了。</p><p>钱对他来说没什么用,他宁可要几只羊羔。</p><p>当然,我是不可能随身带着小羊羔的。</p><hr><h1 id=\"二\"><a href=\"#二\" class=\"headerlink\" title=\"二\"></a>二</h1><p>四五年之后,我都快忘了这件事了,偶然间却又听到了有关那双鞋的传闻。</p><p>那是在一次饭局上,一个小老板说要给我们讲个故事。</p><p>小老板的厂子是做手工艺品的,雇了几个老工人,订单虽然挺多,但是干活的人少,所以也不敢接太大的单子。</p><p>有一天,一个中年人背着一个大包找到了他。中年人从包里掏出一些精巧的小玩意儿,希望能够跟他合作。</p><p>小老板仔细看了看,手艺确实没得说,不过当中年人说起他们那个村子时,小老板觉得这桩生意要黄了。</p><p>小老板也算半个本地人,中年人说的那个村子他知道,藏在深山老林里,车都很难开进去。</p><p>中年人摆了摆手,笑了。他跟小老板说,明天到村里去看看就知道了。</p><p>第二天,小老板将信将疑地跟着中年人往那个村子去,快到记忆里特别难走的小路时,小老板惊呆了。</p><p>原本顶多两个人并排走的小路,现在却变成了一条宽敞的水泥马路,小轿车一路平稳地开进了村。</p><p>生意谈成了,自然要一醉方休,小老板灌醉了那个中年人,套了半天话才知道,原来中年人有一双神奇的鞋。</p><p>听到这里,我不禁竖起了耳朵。</p><p>牧羊人和那双鞋的事儿,不知怎么地传到了村子里。村里几个管事儿的一合计,让中年人跑去找到牧羊人,许给他十五只羊羔加一双新鞋,换回了那双能走出路来的鞋。</p><p>没想到,这一回走出来的竟然不是土路而是水泥路。村里几个年轻小伙子轮番上阵,没两天的工夫就把敞亮的大路给修好了。</p><p>后来那双鞋呢?我赶紧问了一句。</p><p>小老板说,签过合同之后没几天,村子里来了个陌生人,花了大价钱把这双鞋买走了。钱货两清,陌生人就提着鞋子赶紧走了。</p><hr><h1 id=\"三\"><a href=\"#三\" class=\"headerlink\" title=\"三\"></a>三</h1><p>稍微年长一些之后,我也没了年轻时的精神气,只是偶尔会和朋友到一些风景区去转转。</p><p>听说在我工作的城市附近建了个新的风景区,我赶忙约上几个狐朋狗友,打算周末跑去看看。</p><p>说是风景区,其实就是一座山头,收的门票倒也不贵。车可以开进去,沿着盘山公路一圈一圈地转,不一会儿就到了山顶。</p><p>往城里看,整个城的景色尽收眼底,让在工作中埋首了五天的我颇有些心旷神怡。</p><p>从风景区的后门出来,竟然是一个不小的镇子,不过到处都是工程机械在拆楼造楼,也没见着什么人气。</p><p>GPS仪上这一片画的是树林,没有标注路线,我和朋友兜着兜着就迷路了。</p><p>幸好,路边走过去一位老大爷,我们赶紧凑上前去问路。</p><p>一问之下,原来老大爷正准备去镇另一头的车站乘车进城,我索性带老大爷一段,麻烦他给指个路。</p><p>老大爷虽然年纪挺大,但是精神还算矍铄;他跟我们讲,其实这个小镇是跟着风景区发展起来的,结果没多久,政府就安排他们拆迁了。</p><p>说到这,老大爷的语气就有些忿忿不平。他说,风景区的路是他们镇上自己修的,投入不少,好处还没捞着就得让出去。</p><p>老大爷又说,一直在外面闯荡的狗娃,拿全身家当换回来一双破鞋,镇上的人还觉得他傻了,等路一修好,大家就都服了。</p><p>我心下一动,问老大爷,难道是一双能走出路来的鞋?</p><p>老大爷似乎挺惊讶我知道这件事,连连点头。</p><p>他接着说,拆迁的时候,狗娃因为当年把宅基地卖了,没分到钱,大家乡里乡亲,狗娃也是为镇上出力,就各家匀了些给他。</p><p>结果狗娃也是个败家的主儿,拿到手的钱很快就花完了,他老婆只好拿家里的东西出去卖钱,有一回不小心弄错了,把那双鞋也一块儿卖了,听说是卖给一个大学生了。</p><p>把老大爷送到城里,临下车时,我对老大爷说了声谢谢。老大爷呵呵地笑着:现在的年轻人真有礼貌。</p><hr><h1 id=\"四\"><a href=\"#四\" class=\"headerlink\" title=\"四\"></a>四</h1><p>拿出前几年工作的积蓄,再加上从朋友那儿借来的钱,我创办了自己的公司;几年的经营下来,利润也算不错。我想到自己年轻时候去过的那些小乡村,想到那双能走出路来的鞋,心里突然有了一个念头。</p><p>经过朋友的牵线搭桥,我联系上了一家基金会。我提出,想要捐钱给一个偏远地区的乡村修一条路。</p><p>基金会的回复很快,他们给我找了好几个项目,但是其中大部分所需的资金额远远超过我的能力所及。</p><p>在列表的最后,我终于找到一个我能够承受的项目,但这个项目和我预想的有点不太一样。从县城到这个村子的路其实已经修好了,但是一年前一场大雨引发泥石流,又把路给堵死了。</p><p>我决定和基金会的人一块儿去走一趟。</p><p>在路上,我问基金会的工作人员,这样的项目政府怎么不出面呢?</p><p>基金会的工作人员也是苦笑,有些县城根本就拨不出钱来,就算能拨出钱来,分给底下几个村去修路,也分不到多少钱。</p><p>而且我接的这个项目还有些不同,据说,这条路压根儿就不是政府修的。</p><p>到了县城,接待我们的不是县里的人,而是县城中学的领导。</p><p>基金会的工作人员说,一会儿县城中学有些师生会和我们一起过去。听说,他们是要去给一个老师扫墓。</p><p>快要到被泥石流掩埋的路段时,我们看到了路边的一方小小的坟墓。在墓碑前面,有一个玻璃盒,里面放着一双鞋。</p><p>这双鞋看起来挺眼熟,旋即我就想到:这不就是牧羊人那双能走出路来的鞋吗?</p><p>在县城中学师生的叙述中,我大概知道后来发生了什么。</p><p>那个买到这双鞋的大学生马上就要毕业了。他带着那双鞋回到家乡,成为县城中学的一名老师。</p><p>由于他年纪不大,学生们都喜欢喊他“小老师”。</p><p>县城中学的学生大多来自周围的小乡村,县城里的学生往往都是到更远的省城里去读书。因为路实在太难走,这些学生平时都住在学校,只有寒暑假才回去一趟。</p><p>在偶然之间,大学生发现了这双鞋的神奇之处。</p><p>从此之后,每到假期,他就会穿上这双鞋,跟着学生去家访,县城到各个村子的路也一条一条地通了。</p><p>去年,在从一个村子回来的路上,突然下起了大雨。他躲闪不及,被埋在了泥石流下面。</p><p>等到其它师生把他的遗体挖出来之后,他们发现,那双鞋也不能用了。</p><p>我问县城中学的一个学生,以后想要做什么。</p><p>她对我说,以后要研究鞋子,要造出像小老师那双一样能够走出路来的鞋。</p><p>我把牧羊人的故事告诉了她,然后说,这样的鞋只有一双而已。</p><p>她说,不对,还有一个人也有一双这样的鞋。</p><p>我很惊讶。</p><p>她说,小老师给我们上语文课的时候讲到,中国有一个著名的作家叫鲁迅。他肯定也有一双这样的鞋。</p><p>因为他曾经说过:其实这世上本没有路,走的人多了,也便成了路。</p><hr><p>(2014.03.03 @ 人人 @ 《杂记GS——人与路》)</p><hr>"},{"title":"有九条命的盖玻片","path":"2014/02/27/ninelives/","permalink":"http://sx349.github.io/2014/02/27/ninelives/","content":"<h1 id=\"一\"><a href=\"#一\" class=\"headerlink\" title=\"一\"></a>一</h1><p>我是一片盖玻片,薄薄的、方方的那种。</p><p>我在偶然之间发现,自己竟然有九条命。</p><p>为什么我知道自己有九条命?我也不知道,大概就跟你知道你自己只有一条命一样。</p><p>当然,当我发现自己有九条命的时候,实际上已经是“曾经”有九条命了。</p><p>我在一所初中“工作”。之所以用“工作”这个词,是因为我一直想做个有事业心的盖玻片。</p><p>盖玻片绝对是高危行业,学生每做一次实验,都会有几个兄弟“殉职”。</p><p>这里面,大部分发生在装片的时候,稍微压得用力一点,盖玻片就粉身碎骨了。</p><p>由于不是什么严谨的实验,做完之后,盖玻片和载玻片都会冲洗干净回收。这是另一个事故高发时间段。</p><p>我现在还记得发现自己有九条命的那次实验。那是一个做观察洋葱外表皮细胞实验的小男孩,微胖。</p><p>微胖的结果就是手略肥,用力略大。只不过这个“略大”已经让我不堪重负了。</p><p>我能感觉到一瞬间自己的身子碎成了两截。我想喊出声,但是我的振动频率太低了,绝对达不到20Hz。</p><p>我确实听到了我的两截身子发出的声音。在那一瞬间我有种灵魂出窍的感觉,仿佛看到两个相干波源发出的声波在我的“灵魂”前发生着干涉。</p><p>为什么我听得到自己发出的声音?我也不知道,大概就跟你听得到你自己发出的声音一样。</p><p>然后我就活过来了。</p><p>当我活着从实验台上回到仪器柜的时候,我受到了犹如英雄凯旋般的欢迎。</p><p>临近实验台的几个盖玻片和载玻片都看到了我断成两截之后又重新粘合成一块的“神迹”。</p><p>我跟他们说:我有九条命。我没说是“曾经”有,反正他们也不会计较这个。</p><p>从此之后,其它载玻片和盖玻片都叫我“九哥”。甚至老资格的三口烧瓶和布氏漏斗看到我的时候也会跟我打个招呼,不过他们管我叫“小九”。</p><p>过了几年,当这所初中终于要搬迁的时候,我大概还剩下四条或者三条命。</p><p>仪器当然是要带到新学校去的,但是盖玻片和载玻片都是便宜货,生物老师估计连看都没有看我们一眼。</p><p>我看到了我的“灾星”,那个差点把我捏死的小男孩。不过当年的熊孩子已经进化成马上要奔赴另一个城市念大学的翩翩少年了。</p><p>他今天可不是“灾星”,而是“救星”。生物老师送了他一些本就打算淘汰的实验仪器,载玻片和盖玻片自然是作为赠品附送了。</p><p>我最后看了一眼仪器柜。Ade,我在仪器柜的青葱岁月。</p><h1 id=\"二\"><a href=\"#二\" class=\"headerlink\" title=\"二\"></a>二</h1><p>生活当然没有想像的那么美好,我从仪器柜中重见天日之后不久,就又被关进了另一个大箱子。</p><p>整理好行囊的男生出发了。我和其它实验仪器一起,留在了他家的储藏室里。</p><p>这样的日子也不算多么难熬。大家会三三两两地围成圈听故事,像三口烧瓶和布氏漏斗这样的老前辈,经历的都是大场面,大家都喜欢听;广口瓶因为常常被拿去做演示实验的缘故,也常常有很多拥趸;当然,我也会不厌其烦地跟各位听众讲我当年如何一次又一次地从熊孩子的魔爪中活下来的惊险故事。</p><p>“九哥,我听牛角管大叔说过,有块载玻片也说过自己有九条命。”有个不识趣的橡胶管冒出头来。</p><p>“呵呵,你以为九条命是路边捡的,随便谁都有九条命啊?来来来,老子告诉你,当初老子知道自己有九条命的时候,那家伙,那场面……”</p><p>古人说过:“山中无甲子,寒尽不知年。”不仅山中无甲子,关在箱子里也是没有甲子的。</p><p>突然的,光线从我们头顶漏了进来。我们看到了男生和一个中年女人站在箱子边上。应该是他的妈妈吧?</p><p>男生开始把仪器一样一样地取了出来,最后挑了一块载玻片和一块盖玻片。</p><p>那块盖玻片就是我。我认得那块载玻片,大家叫他“老逗”。</p><p>其实他本来不叫这个名字,但是因为他经常插嘴说些没头没脑的话,才有了这个大号。</p><p>“老逗”就是那个说过自己曾经有九条命的载玻片,当然我一向是嗤之以鼻的。</p><p>我的事迹可是有好几块盖玻片和载玻片亲眼目睹的,除此之外,目击者还包括一根玻棒、一个烧杯和一个培养皿。</p><p>男生小心地把一小瓶水倒进培养皿。这瓶水很浑浊,漂着点点绿藻。男生开始小心地寻找起来。</p><p>我和“老逗”没有什么交流,像我这样有头有脸的人物也不愿意和这样拙劣的模仿者计较。</p><p>男生把一滴水滴在“老逗”身上,然后我就被他盖了上去。</p><p>我盖住的是一个绿色的小细胞,长得很丑,就像把毛豆掰成两半然后压平。</p><p>我不认识它,倒是“老逗”先认出来了:“亚心形扁藻。”</p><p>我很好奇,但又放不下架子去问他。还好“老逗”接着说:“我以前在一个标本室待过,所以见过。”</p><p>“老逗”好像知道我在想什么。</p><p>男生仔细端详了一会儿,不知从哪里找来一个刚刚好的玻璃盒,把我和“老逗”放了进去。</p><h1 id=\"三\"><a href=\"#三\" class=\"headerlink\" title=\"三\"></a>三</h1><p>男生又要到外地去了,他不仅带上了玻璃盒,还专门拿了一个小绸包装着。</p><p>在这期间,我和“老逗”终于迈出了交流的第一步。</p><p>准确地说,在我向“老逗”又讲了一遍自己有九条命的经历之后,他说了一句:“我也曾经有过九条命。”</p><p>他看来是不想深谈这个话题。但是打开了话匣子之后,我们的对话也不再像之前那样单调了。</p><p>我从他那里知道了各种各样的标本,而我也告诉他我见过的各种各样的熊孩子。</p><p>我感到小绸包有轻微地摇晃。应该是男生拿着绸包在急匆匆地往哪里赶去。</p><p>他停住了。应该是说了些什么话,但是在绸包里听不太清楚。</p><p>然后男生手忙脚乱地打开了绸包。</p><p>我看见他面前站着的女生,很文静,戴了一副粉红色边框的眼睛,一袭天蓝色的长裙,长发及腰;虽然只是微笑,但是很甜。</p><p>男生把玻璃盒递了过去,脸上略有点害羞的红。“这是送给你的……”</p><p>女生伸手接了过去,仔细地看着。</p><p>“这叫……”男生看女生没有说话,打算解释一下。</p><p>“曼德勃罗特集!”女生的语气带着惊喜。男生愣住了,他看到了载玻片上的那个绿藻细胞。</p><p>我也不敢相信自己的眼睛,虽然我没有眼睛。那个绿藻细胞还是大致呈现心形,但是两边延伸出去,成了它本身小一号的复制品,然后又从两边延伸出去……</p><p>“谢谢你!这是我收到过的最好的礼物。”女生还是带着微笑的表情,男生还是愣愣地站在原地。我都替他着急了。</p><p>后来发生了什么,就不用我多说了。</p><p>不过,我一直很奇怪,到底是谁把那个丑不拉几的亚心形扁藻变成了那么漂亮的曼德勃罗特集。</p><p>我猜是“老逗”搞的鬼,但是他不肯说。</p><p>有一天,他又没头没脑地打断我说话。当时我在和女生书桌上的笔盖宣传自己九条命的事迹。</p><p>“我也曾经有九条命。”他没有像以往那样停下来。“有一天,我听到一个声音跟我说:你愿意用八条命换一次奇妙的经历吗?”</p><p>“载玻片本来就只有一次生命。多出来的八条命我也用不上。于是我决定换。然后我就感觉到自己又只剩下一条命了。”</p><p>“为什么我知道自己只有一条命了?我也不知道,大概就跟你知道你自己有九条命一样。”</p><p>我大概知道在绸包里到底发生了什么。看来我的猜测还是挺准的。</p><p>“你不后悔吗?”我忍不住多嘴了一句。</p><p>“有什么好后悔的?”“老逗”这时候真的很像一个历经沧桑的老人,我感觉他的声音像是经历了几百个世纪传了过来。</p><p>“有些事情,用一次生命去经历就足够了。”</p><p>(2014.02.27 @ 人人 @ 杂记GR–有九条命的盖玻片)</p>"},{"title":"出院","path":"2014/02/24/novel01/","permalink":"http://sx349.github.io/2014/02/24/novel01/","content":"<hr><h1 id=\"一\"><a href=\"#一\" class=\"headerlink\" title=\"一\"></a>一</h1><p>ψ先生确信,自己大概是在2岁的时候被送进精神病院的。</p><p>这是间很大的精神病院,有一座小城镇那么大,有学校,有警局,有政府,甚至还有一间精神病院。</p><p>精神病院里的精神病院,住的肯定是一些非常严重的精神病人。</p><p>上小学的时候,ψ先生,不,那时候还是小ψ,知道了“精神病”这个词。在他班上有一个戴眼镜的小男孩,成天捧着一本厚厚的大英百科全书,从来不和其它小朋友玩。小朋友们喜欢叫他“小四眼”,但有的小朋友也会喊:“神经病!”</p><p>“小四眼”当然不喜欢别人叫他“神经病”:“你想说的实际上是‘精神病’吧,我没有精神病,我妈带我去检查过……”</p><p>“神经病!”其它小朋友一哄而散,只留下“小四眼”一个人。</p><p>小ψ到了初中,终于也搞清了“神经病”和“精神病”的区别。他开始觉得,自己没有精神病。</p><p>小ψ没有父母,是靠社会福利长大的,所以他也没法去了解自己的家族有没有精神病史。但是觉得自己没病的念头却越来越强烈,甚至有逐渐蔓延的趋势。有时上着课,小ψ会突然走神:老师也在这间精神病院里,老师有没有精神病呢?老师是不是也是误诊了呢?</p><p>小ψ从小学念到初中,又从初中念到高中,没有念大学就去工作了。精神病院里怎么会有大学呢?</p><hr><h1 id=\"二\"><a href=\"#二\" class=\"headerlink\" title=\"二\"></a>二</h1><p>开始工作之后,小ψ也终于成了真正的ψ先生。差不多在同一年,精神病院里接通了因特网。</p><p>ψ先生第一次发现,原来还有很多人同样觉得自己是被误诊的精神病人。每天下班回家,他就在互联网上和这些“病友”交流如何从精神病院里逃出去,他们还建立了一个组织,叫作United Fleeing-the-hospital Organization,简称UFO。</p><p>ψ先生是UFO的骨干成员,参与讨论也非常积极。他在偶然中发现,有些人在网上搞了些大动静之后,就突然销声匿迹了。他推断,这些人一定是经过重新诊断之后出院了。</p><p>他迫不及待地把这个消息告诉了UFO的其它成员,其它人对ψ先生的睿智感到无比钦佩,于是一场名为“我们要出院”的大型活动草草出炉了。活动的地点,经过个别成员提议及全体成员投票,决定放在有城楼的那个广场。</p><p>ψ先生是在去印刷“我们要出院”标语的第二天被关进看守所的,他不知道警察是怎么找到他的。他在看守所待了整整一个月,等到他回家时,UFO里已经没什么动静了。ψ先生想,其它人应该还是顺利举办了活动,都出院了。</p><p>UFO里还剩一条未读消息,是一个小故事:三个人被关进了精神病院,第一个人不停地讲“世界是圆的”,讲到第14遍的时候,护士给了他一针;第二个人把他知道的各种政治历史地理常识都说给护士听,护士也给了他一针;第三个人该吃的时候吃,该睡的时候睡,没过几天,他就被送出了精神病院。</p><p>ψ先生恍然大悟:想要出院,只要像个正常人那样生活就行了。从此之后,ψ先生的生活重新恢复了平静。</p><hr><h1 id=\"三\"><a href=\"#三\" class=\"headerlink\" title=\"三\"></a>三</h1><p>28岁的时候,事业小成的ψ先生结婚了;1年后,他又添了一个可爱的女儿。但是ψ先生从来没有忘记自己不是精神病人,自己要从这个精神病院里逃出去。</p><p>在30岁生日那天,ψ先生许了个愿:我要出院。</p><p>晚上,ψ先生突然对妻子说:“明天我们去看一下医生吧。”</p><p>妻子很奇怪:“你不舒服吗?”</p><p>ψ先生摇了摇头:“我要出院。”</p><p>妻子愕然。</p><p>第二天,还睡得有些迷迷糊糊的ψ先生跟着妻子来到了精神病院里的那间精神病院。做了一系列检查,又被问了几个问题之后,医生让ψ先生到门口等着。ψ先生隐约听见医生在和他的妻子说:“……脑部撞击……后遗症……重度妄想……”其它的都听不清了。</p><p>医生出来了,没有管ψ先生的妻子,径直领着ψ先生向病房走去。</p><p>路过一间没关门的病房,ψ先生好奇地探头进去看了一眼。门边站着一个穿着病号服的男人,一只眼睛闭着,一只眼睛睁着。</p><p>“虎子?”ψ先生认出这是UFO的一位“病友”。</p><p>“我不是虎子,我是一个开关,所以我不能跟你说话。”这个高大的汉子往背后一靠,墙上的开关又被按下去一个,病房里的灯全亮了。高大汉子赶紧把闭着的那只眼睛睁开。</p><p>“快走快走!”医生把ψ先生从这间病房里拉了出来,把他推进了另一间病房。ψ先生又看到了几张熟悉的面孔,也是UFO的“病友”们。</p><p>躺在病床上,看着周围乳白色的墙,ψ先生不由得感叹:</p><p>终于出院了。</p><hr><p>(附注:ψ,希腊字母Psi,此符号有时用于心理学及精神病学)</p><p>(2014.02.24 @ 人人 @ 《杂记GQ——出院》)</p><hr>"},{"title":"世上只有一只乌鸦","path":"2014/02/23/crow/","permalink":"http://sx349.github.io/2014/02/23/crow/","content":"<p>“世上只有一只乌鸦。”他看着窗外枯枝上密密麻麻的乌鸦说道。</p><p>我感觉自己没有跟上他的思路:“一只?”<br>“对,一只。”他指了指窗外,“这只……当然也可以是这只。无所谓,只有一只。”<br>“你在开玩笑,”虽然我知道他一定不是在开玩笑,更高的可能性是……他脑子被门夹了。</p><p>噢,我忘记说了。他是一名哲学系的研究生,但是他本科期间曾经在CERN待过半年。<br>他曾经和周围几乎每个人都开过这个玩笑:“看,那儿有一只薛定谔的猫咪。当然你看不见她是正常的,因为你一观察她就坍缩成不存在了。”<br>总之……就是这样。</p><p>“世上的确只有一只乌鸦。”他没有反驳我的鲁莽推断,“所以中国有句老话:天下乌鸦一般黑。”<br>“别卖弄了,这句话根本就不是你说的这个意思。”我很不耐烦,很想去外面捉一只,不对,是两只乌鸦给他看。<br>“我当然知道。这叫歪打正着。”他在这个正对着窗户,恰好能够让太阳晒到上半身而晒不到下半身的沙发上直起身来。“别数了,在你开始数数之前,在你的视线范围内有三十五只乌鸦——”<br>“三十六只。”我看到一只乌鸦落在了枝上。“你不是说只有一只吗?怎么突然又承认有三十五只了。”<br>“——的投影。”<br>他刚才绝对是故意让我打断他的话的。不过我没有接腔,因为我知道他一定在等人问他为什么,这大概是一切有头脑(天哪我竟然会用这个词来形容他)的人都会做的事。</p><p>但是他的身子却和整个对话一起陷入诡异的沉默之中。哦对不起,他的身子只是陷入了沙发。<br>“什么……”我刚想问,他却开口了:“投影。顾名思义。投下来的影子。”<br>我觉得自己被抢白了,但是这不妨碍我的问题:“什么的投影?”<br>“世上唯一的那只乌鸦啊。”他又躺进沙发里了,我不知道他在不在笑,但是显然这句话在嘲笑我的智商。<br>“我的大哲学家,你的那只乌鸦在哪里呢?”我觉得我的咬牙切齿已经到了恨不得把那只乌鸦煮了吃了的地步。<br>“终于问到点子上了。”他也终于决定起身了,回过身来的脸上带着一副“不错哦,小朋友你总算开窍了”的表情。“在更高一维的空间。”<br>我不喜欢他这副表情。就像我不喜欢那个更高一维的空间一样。但他已经摆出了这副表情,就好像在说:“更高一维的空间嘛……她就是存在着。”<br>“在三维空间中,影子是真正的二维物体,它缺少了厚度。当然,称之为‘物体’其实是很不准确的。”<br>“那么,一个四维的‘乌鸦’,被一个四维的‘光源’照射产生的投影,就成为了一只三维的乌鸦;被无数个‘光源’照射,就出现了无数只乌鸦。”<br>他的讲座还没开始就结束了,因为讲完这两句话他就走回沙发,重新陷了进去。</p><p>我突然觉得之前的薛定谔的猫咪真是太可爱、太人畜无害了。<br>一只不存在的猫咪当然比一只乌鸦可爱。不管怎样那也是只猫咪。希望那只猫咪喜欢吃乌鸦,吃得只剩一只就好,吃得一只不剩更好。<br>“不对!”我突然意识到一个问题,感谢我的高中生物老师……算了,其实跟她也没什么关系,进入大学以后我的生导完全是重新自学的。“肯定有人解剖过乌鸦吧?如果乌鸦真是影子的话,那么应该彻彻底底地都是黑的吧。”<br>我觉得我应该难倒他了。理由之一划一就是:他不说话了。<br>“因为衍射。”在漫长(其实不超过一分钟)的沉默之后,我差点以为是沙发说话了。当然,沙发是不可能说话的,除非世上只有一台沙发,那什么都有可能发生。<br>“在现实世界中,衍射造成了影子的模糊边缘以及泊松亮斑。所以……”他话说了一半,但我想我已经明白了他的意思。<br>“那进化学说呢?”我又抛出了一枚‘炸弹’。我觉得我瞬间的智商达到了200。<br>“为什么四维的‘乌鸦’不能进化?”他听起来有点强词夺理,因为进化绝不是一条路走到黑的。走到“黑”,好吧,这真是一个无聊而巧妙的双关。</p><p>我和我的“手下败将”都没有说话。虽然我确实觉得他是因为无话可说了,但我更觉得是他觉得我面对他的反问无话可说了。<br>“其实世上确实不是只有一只乌鸦。”他退缩了。我决定今天回去要在人人上发一条状态,然后用32个小号给自己点32个赞,来纪念我今天200+的智商。<br>“世上只有一个生命体,乌鸦也好,天鹅也好,癞蛤蟆也好。还有你。”他重新夺回了阵地,省掉了我回去建32个小号的麻烦。<br>“当然也有我。”他懒洋洋地补充道。我不知道他是不是感觉到我手上的茶杯有很大可能做一个从我的手到他的头的斜抛运动。<br>“所以你追妹子的时候打算告诉她:‘其实我们是同一个人,只是投影在这个世界上的时候变成了两个’?”我决定反击。哲学系的男生在找妹子方面的天赋简直惨不忍睹,至于工科出身的哲学系男生……惨不忍睹中的惨不忍睹。<br>“呵呵。”天生脸皮厚什么的最讨厌了。</p><p>我决定把我们的对话贴到网上。在我打完最后一句话,设置好日志分类,还没按下“发布”的时候,人人网提示收到了一条新消息。<br>我还是先按了“发布”按钮,然后打开新鲜事。我感觉见了鬼了。<br>新消息是对我新鲜出炉的日志的第一条回复。是他。<br>“也许事情更好玩一些。其实世上只有一个‘基本’粒子。比夸克更加基本的‘基本’粒子。一切都是它在自娱自乐罢了。”</p><p>(2014.02.23 @ 人人 @ 《杂记GP–世上只有一只乌鸦》)</p>"},{"title":"论马恩早期社会主义观点的局限性","path":"2014/02/05/socialism/","permalink":"http://sx349.github.io/2014/02/05/socialism/","content":"<p>原题《从尚未灭亡的资本主义看马克思和恩格斯社会主义观点的历史局限性》,马原课程期中读书报告。</p><hr><h1 id=\"概述\"><a href=\"#概述\" class=\"headerlink\" title=\"概述\"></a>概述</h1><p>无论是《共产党宣言》还是《社会主义从空想到科学的发展》,马克思和恩格斯对于资本主义的未来发展都下了如下的断言:</p><p>“资产阶级的灭亡和无产阶级的胜利是同样不可避免的。”[1]</p><p>“……应当说明资本主义生产方式的历史联系和它在一定历史时期存在的必然性,从而说明它灭亡的必然性……”[2]</p><p>但现实的观察告诉我们,资本主义不仅没有灭亡,而且仍然在西方国家中盛行;相反的,在二战过后以东欧和亚洲国家为主形成的社会主义联盟,在上世纪90年代遭遇挫败,自东欧剧变、苏联解体后,这些国家反而纷纷转为由资产阶级执掌政权。</p><p>马、恩曾经指出:“这似乎是历史发展的规律:资产阶级在欧洲任何一个国家都不能像中世纪的封建贵族那样独掌政权,至少不能长期独掌政权。”[2]他们尤其提到了法国在封建王朝和资本主义之间来回摇摆的过程,认为第三共和国“才执掌政权20年以上,可是已经显露鲜明的衰落征兆”[2]。他们关注了没有封建基础而得以顺利发展资本主义的美国,但认为对当时的美国来说,无产阶级已经在“敲门”了。</p><p>不过,法国在又经过几次反复之后,终于在1958年步入法兰西第五共和国,至今已经55年;美国在大萧条中确实借鉴了社会主义的优点(即采用罗斯福的“国家垄断资本主义”),但经过二战大发战争财之后,美国一跃而成为世界头号强国,采用的仍然是资本主义而非社会主义。</p><p>如果仔细阅读马、恩在上述两本著作中的观点,我们可以看出,他们的论断既然基于当时的社会实践,也就不可避免地局限于当时的社会发展条件;他们对于后来资本主义发展的预期,实际上是对于当时的社会生产发展路径的简单复制;他们所推得的最终结果指向一个极端化的方向,与现实的情况有着极大的偏差,也是十分正常的。</p><hr><h1 id=\"马、恩认为资本主义灭亡的原因与实际情况\"><a href=\"#马、恩认为资本主义灭亡的原因与实际情况\" class=\"headerlink\" title=\"马、恩认为资本主义灭亡的原因与实际情况\"></a>马、恩认为资本主义灭亡的原因与实际情况</h1><h2 id=\"对不同阶级的考察\"><a href=\"#对不同阶级的考察\" class=\"headerlink\" title=\"对不同阶级的考察\"></a>对不同阶级的考察</h2><p>在马、恩笔下,整个社会可以分为三个部分:无产阶级、资产阶级和其它中间阶级。以下将对这三个部分分别进行考察。</p><p>无产阶级是马、恩所考量的最重要的阶级,是作为新的社会秩序的创造者的存在。从无产阶级的角度来说,资本主义必须被推翻,因为“资产阶级……甚至不能保证自己的奴隶维持奴隶的生活……它的存在不再同社会相容了”[1]这里所谓的“奴隶”和“奴隶的生活”,指的是“雇佣劳动的平均价格是最低限度的工资,即工人为维持其工人的生活所必需的生活资料的数额……只够勉强维持他的生命的再生产。”[1]</p><p>那么,为什么工人不能要求获得自己所创造产品价值数额的收入呢?这可以找出三方面的原因:第一,“机器的改进就意味着越来越多的机器劳动者本身受到排挤,而归根到底就意味着造成一批超过资本在经营上的平均需要的、可供支配的雇佣劳动者,一支真正的产业后备军”[2]第二,“由于机器的推广和分工,无产者的劳动已经失去了任何独立的性质。”[1]这两点保证了,对于希望获得超出生活必需数额的工资的工人,资本家随时可以从产业后备军中获得所需维生的工资同样低且具有同样的工作能力的替代品;最后,“雇佣劳动……以前是一种副业,现在成了工人的唯一职业。”[2]无产阶级一旦成为无产阶级,就几乎不可能脱离这个阶层。</p><p>总结这些原因,就构成了马、恩关于无产阶级最重要的两个假设:第一,私有制导致生产资料和生产力分离,出现了“自由但一无所有”的工人,而且工人将长期稳定地处于这种状态;第二,工业发展使得工人的劳动同质化,在雇佣市场上形成了买方(即雇佣方)市场。</p><p>当时的工人普遍处于这种悲惨的遭遇之中,这是不可否认的事实。但是从那时候到当今的资本主义发展,并非始终完全地维持着这种状态。</p><p>在现在的社会生产中,工资是如何决定的呢?我们可以看到,在一般的情况下,工人的工资已经包括了马克思所谓的剩余价值。实际上,如果一个工人能够创造超出他工资的价值,那么其它的企业就会付出更高的(但也不会高于其创造的价值)工资将他挖过去,随着这一过程不断进行,最终剩余价值将不再归属于企业或拥有企业的资本家,而成为工人自己的收入。这与马、恩所考虑的情况之所以不同,在于如下几个方面:</p><p>首先,在很多产业之中,产业预备军已经不复存在,在工业发展的进程中,对于能力的要求开始逐渐细分,劳动的独立性质开始回归;对于具有某方面能力的工人,不再能够轻易地找到替代品;</p><p>其次,工会发挥了重要的作用。工人之间结成联盟是马、恩所提出的重要手段,但这一手段在他们手中是发起暴力革命的组织;而现在企业中的工会,实际上更符合他们所批判的小资产阶级资本主义中工业的行会制度;当然,随着时代的不断发展,工会已经不再是熟练工人的组织,而是全体工人的组织,尤其关注改善底层工人的遭遇。工会也打破了雇佣市场上的买方市场格局,甚而有代之为卖方市场的趋势,</p><p>不能剥削剩余价值,资本家是不是就没有收入了呢?首先,资本家对工人的剥削并不是完全地被消除了,在劳动密集型企业中,底层工人仍然是受剥削的(受益于工会和政府的要求,底层工人可以获得最低工资保障,但也仅仅是维持基本的生存需要);其次,资本家本身的劳动也仍然在创造价值;第三,如果我们从剩余价值中剥离工人所创造的超出其工资的价值,是不是就什么都不剩了呢?或者说,生产资料在产生价值的过程中是不是毫无贡献呢?剩余价值理论假定了这个结论,但在当今社会,工资越向实际创造的价值靠拢,我们就越能看出,生产资料(主要是资本,也包括其它除了劳动力以外的生产要素)确实应当占据“剩余价值”的一部分;因此,资本家由于投入的资本和其它生产要素的缘故,也应当获得一部分收入。</p><p>接下来讨论中间阶级,这些马、恩看来“不稳定的手工业者和小商人群众” [2]并没有迅速地将财产出卖给大资本家和大地主,分化为一无所有的无产者;正相反,今天有越来越多获得了一定的生产资料的工人正在加入小资产阶级的行列。实际上,当人们不满足于大工业产品的单一标准化时,来自于手工业者、小工业家以及小商人的带有个性的产品,开始恢复生命力,使得这些人免于被竞争抛到无产阶级队伍里去。马、恩提出来的监工制,究其根本还是建立在存在产业预备军的基础之上,也就是标准化工业的基础之上。</p><p>如果说,在那个年代,“资产阶级抹去了一切向来受人尊崇和令人敬畏的职业的神圣光环。它把医生、律师、教士、诗人和学者变成了它出钱招雇的雇佣劳动者。” [1]那么现在,这些职业重新获得了独立的身份与价值。</p><p>最后,资本家又在做些什么呢?准确地说,恩格斯所作的预言并没有落空,“资本家的全部社会职能现在由领工薪的职员来执行了。资本家除了拿红利、持有剪息票、在各种资本家相互争夺彼此的资本的交易所中进行投机以外,再也没有任何其他的社会活动了” [2]。但我们可以看到,投机在今日的意义已经远远超出那个时代,如今在金融市场上流动的资本是实际生产过程中创造的产值的几倍甚至几十倍。但这并不是说资本流动是无意义的富人游戏,金融市场的重要作用是连接资本和劳动(正如同当年资本家的资本与工人的劳动力那样),从而创造更高的价值,至于这部分价值的归属问题已经在上面讨论过,此处不再赘述。</p><h2 id=\"对生产力发展程度的考察\"><a href=\"#对生产力发展程度的考察\" class=\"headerlink\" title=\"对生产力发展程度的考察\"></a>对生产力发展程度的考察</h2><p>生产力和生产关系相适应,才能保证社会生产的正常运行和发展。马、恩观察到,“资产阶级在它的不到一百年的阶级统治中所创造的生产力,比过去一切世代创造的全部生产力还要多,还要大。”[1]但是,“社会所拥有的生产力已经不能再促进资产阶级文明和资产阶级所有制关系的发展;相反,生产力已经强大到这种关系所不能适应的地步,它已经受到这种关系的阻碍”[i],“市场的扩张赶不上生产的扩张。”[2]这种发展导致了周期性的、因生产过剩而产生的经济危机。资本主义为了应对危机,“一方面不得不消灭大量生产力,另一方面夺取新的市场,更加彻底地利用旧的市场。”[1]但这种做法并不是马、恩认为能够治本的方法。相反,他们认为,过去的分工以及阶级划分,“是以生产的不足为基础的,它将被现代生产力的充分发展所消灭。”[2]</p><p>基于实际观察,我们可以提出对这一想法的几点挑战:</p><p>其一,最近一次金融危机,即2008年以来的次贷危机,虽然同样表现为购买力不足,但和马、恩当时的经济危机以及后来1929年的大萧条在实质上有着巨大的区别。尤以1929年的大萧条做对比,当时购买力不足的主要原因是工厂破产、工人失业导致的,对于工人的影响是间接的,而2008年的次贷危机则是之前的房产泡沫破灭使得所有人手中的财富大幅度贬值,这种影响是直接的。虽然并没有完全逃脱潜在的危机,但是资本主义已经在实际意义上脱离了周期性的生产过剩,应对危机的方式也已经有所改变。</p><p>其二,生产力的发展程度到底是否已经达到了必须要依靠生产关系的改善来继续的地步呢?承接第一点的最后,之所以说资本主义脱离了生产过剩,有着多方面的原因,其中之一就是现在市场的扩张已经赶上甚至超过了生产的扩张,而且扩张的方式就是“夺取新的市场”和“更加彻底地利用旧的市场”,不过,在扩张之中,已经增添了许多新的革命性因素,其中之一就是互联网。互联网不仅本身创造了巨大的市场,而且对于已有的几乎所有市场都提供了更加丰富的表达和发展方式,这是马、恩所无法预见的。</p><p>这不禁让我想到了马尔萨斯的人口论,马克思曾经批判地认识到,马尔萨斯的资源有限论,已经为资本主义的高速工业发展所打破,但是他却走向了另一个极端,也即对于生产力的估计过于乐观。目前,生产力和需求的发展呈现出一个水涨船高的态势,当今的科技发展对于两者的促进作用是同时的,甚至对于后者所提供的可能性更加丰富。</p><p>最终,我们也许会得出这样的结论:在可预见的未来,生产力几乎不可能达到满足需求的地步。高度发展的生产力固然可以创造出形形色色的产品来满足我们各种稀奇古怪的想法,但是再如何发展,我们也不可能在一天之中获得25个小时来满足自己的需求。我们的物质需求能够随着生产力的不断发展得到满足,但是时间的限制是无法被克服的。最终结果是,生产力的发展或将走入一个永远走不完的“死胡同”。</p><h2 id=\"对不同生产关系的考察\"><a href=\"#对不同生产关系的考察\" class=\"headerlink\" title=\"对不同生产关系的考察\"></a>对不同生产关系的考察</h2><p>在马、恩的时代,资本主义已经开始了向社会主义学习的步伐。从股份公司到托拉斯到国家占有,资本主义是否正在改变生产关系向社会主义不断靠拢呢?</p><p>的确,罗斯福的国家垄断资本主义,毫不隐讳地要求国家控制市场,但是“市场无效”对于西方经济学来说并不是一个新鲜的概念,在市场无效发生时,要求政府参与调控市场来达到最优效果,这更多地是资本主义功利性的要求,而不能简单地理解为对社会主义的模仿。</p><p>最好的一点证明是,如果如马、恩所说,“在托拉斯中,自由竞争转变为垄断,而资本主义社会的无计划生产向行将到来的社会主义社会的计划生产投降。”[2]那么如何解释目前西方各国无一例外的拥有反垄断法案和相应的监管机构?西方经济学的解释是,托拉斯的垄断相比于自由竞争是无效率的,打破托拉斯有利于提高社会总福利。因此,垄断是企业维护自身利益的行为,最终的目标是企业自身的利润,而绝非有计划地为社会福利而生产。</p><p>计划生产不能为西方所接受的,不是意识形态的问题,也不完全是如马、恩所想的对于效率的质疑(苏联和中国的计划经济时代都是工人积极性最高的时代,这一点西方各国无法否认),而是对于其根源上的可行性的质疑。完全的计划生产,既要符合生产能力的限制,也要满足社会产品的需求,光靠个人来运作显然是不行的,那么一个计划经济委员会以及数以千万计的调查工作人员,是否就能够获取完整而全面的信息了呢?从宏观的角度来看,市场经济是希望依靠局部的优化来达到市场整体的最优(但实际上市场整体还是呈现出“无政府状态”),计划经济则是自顶向下的顶层设计模式,后者将会更加符合社会利益,尤其是如果我们能够拥有充分多的信息作为决策来源。但是当视角转到微观时,时间再一次成为重要的限制条件,在较短的时间内,我们永远能够保证自己处在某个局部的交易之中,从市场经济中获益;与此同时,计划经济委员会还在依靠各种信息做出汇总,最终决策看起来遥遥无期。</p><p>也就是说,前者在微观方面,后者在宏观方面,各自具有超出另一方的优势。在现实的观察中,我们可以看到,无论是资本主义国家还是社会主义国家,在经济上的发展方式都有着将两种经济相互融合的倾向,这是一种各取所长的必然趋势。</p><hr><h1 id=\"马、恩社会主义观点的历史局限性\"><a href=\"#马、恩社会主义观点的历史局限性\" class=\"headerlink\" title=\"马、恩社会主义观点的历史局限性\"></a>马、恩社会主义观点的历史局限性</h1><p>“为了使社会主义变为科学,就必须首先把它置于现实的基础之上。”[ii]但是,所谓的“现实”既不是静态的,也不是以某种固定模式不断发展,而是经历着不断变化中的变化。而在《共产党宣言》和《社会主义从空想到科学的发展》这两部著作中,马、恩的社会主义观点都局限于当下,没有(也几乎不可能)将眼光拓展到未来的社会(尤其是资本主义)发展模式。</p><p>在马、恩撰写两部著作时,社会正处于大工业发展的鼎盛时期,工业会继续高速地以目前的状况发展是一种合理的推测:标准化的产品将替代人们的选择,个性逐渐被磨灭,资产阶级和无产阶级的分化进一步加重,由此自然而然地朝着为社会主义作准备的方向发展。但是在真实的世界中,科技出人意料地扮演了搅乱一切的角色,它所迸发出的改变世界面貌的能量是马、恩始料未及的。新科技不仅催生了新的产业,同时为旧产业增加了极高的附加值,产业细分达到了前所未有的程度,不仅通过工业的生产,同时也依靠第三产业的发展。在这个过程中,原本只是中间阶级的小商人——不进行生产,只进行“投机倒把”的一类群体——成为连接需求与生产,从而创造财富的中坚力量,而在这个过程中却并没有伴随着生产力的发展。</p><p>在对于工人的认识中,马、恩忽视了小资产阶级行会制度所能产生的效果。的确,行会制度不能从根本上改变剥削的事实,但是在无产阶级的早期实践中,就有过一些利用行会制度(而不是组织暴力革命)来改善工人生活的措施,如英国的十小时工作日法案。马、恩认为无产阶级和资产阶级的对立是一种阶级上的、不可调和的对立,但事实上这种对立正在通过利益的再分配得到一定程度上的调和,这种程度起码已经将暴力革命的必要性予以消除了。特别是现在的行业协会不再像马、恩时代那样只关注熟练工人的利益,而是像当时的国际工人联盟组织那样,尤其关心最底层工人的福利。在工会的基础上结成政党的例子,比如英国的工党,采取的正是名为“民主社会主义”,实际上是小资产阶级资本主义的价值理念。</p><p>相比于19世纪末,平等观念也越来越深入人心,即使在收入上还存在着贫富的差距,但是对于阶级的概念已经开始淡化;同时也由于原先那种将工人视为奴隶般对待的减少,整个社会更加倾向于将提供劳动视为一种公平的交易。马克思曾经认为,“行政上的改良……是在这种生产关系的基础上实行的,因而丝毫不会改变资本和雇佣劳动的关系,至多只能减少资产阶级的统治费用和简化它的财政管理。”[1]但当时的行政改良绝不可能像今天那样,将所有人放在同样的社会地位之上,而是始终将不同的阶级区分看待;平等是随着当代政治哲学的发展而兴起的新的普世价值,这是殖民浪潮兴盛的19世纪末绝不可能拥有的。</p><p>综上所述,社会生产的发展、行会制度以及平等的社会地位,是马克思、恩格斯囿于所在时代的历史局限性而不可能预见的现实状况,也反过来成为当今对于马、恩社会主义观点的最大挑战。</p><hr><h1 id=\"参考文献\"><a href=\"#参考文献\" class=\"headerlink\" title=\"参考文献\"></a>参考文献</h1><p>[1] 马克思、恩格斯《共产党宣言》<br>[2] 恩格斯《社会主义从空想到科学的发展》</p><hr><p>(2014.02.05 @ 人人 @ 《杂记GG–马原课的读书报告》)</p><hr>"},{"title":"瞎扯·2013·三","path":"2013/12/31/tweet-2013-3/","permalink":"http://sx349.github.io/2013/12/31/tweet-2013-3/","content":"<p>《瞎扯·2013·三》为《瞎扯》的第五辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>因为这是一个改变了世界的同性恋=w=<br>@邵成.torrent:<br>Hacker News上有一句话说得好:当年因此罪而含冤去世的人有数万,为何今日只赦免了图灵?<br>@陈力坤:<br>今天英国皇室正式“赦免”了阿兰·图灵同性恋之罪。1953-1954年,图灵被迫接受“同性恋治疗”,身心受到巨大创伤。于是他选了一种最浪漫的方法,咬下一口含有氰化物的苹果离开人世。对着屏幕,吃了苹果,今夜平安,值得庆贺。<br>(2013.12.24)</p><hr><p>@丁栩翔:<br>浏览了下蒋方舟的《中国纪事2013》,基本上可以判断,蒋小姐所认识的中国,就是每天睡到大中午醒来迷迷糊糊躺床上打开手机刷新下新浪微博和网易新闻客户端所看到的中国——陈水总烧了厦门公交,冀中星炸了首都机场——至于正常中国人的正常生活,和绝大多数中国的政治经济文化精英一样,她无从知晓,也没兴趣知晓。中国很大,有960万平方公里,但对有些人而言,中国只有4英寸。<br>(2013.12.24)</p><hr><p>@周孝正:<br>新闻学研究尘埃乍起,社会学研究尘埃飞扬,历史学研究尘埃落定。<br>(2013.11.02)</p><hr><p>@冯白羽:<br>必须赌博。 ——帕斯卡<br>(2013.09.30)</p><hr><p>Live as if you were to die tomorrow. Learn as if you were to live forever.<br>(2013.09.25)</p><hr><p>@范一舟ARSENAL:<br>“上尉,你是个白痴吗?!”舰队统帅大怒,“你是想告诉我们,一种没有记忆遗传,相互间用声波进行信息交流,并且是以令人难以置信的每秒1至10比特的速率进行交流的物种,能创造出5B级文明?!而且这种文明是在没有任何外部高级文明培植的情况下自行进化的?!”“但,阁下,确实如此。”“但在这种状态下,这个物种根本不可能在每代之间积累和传递知识,而这是文明进化所必需的!”“他们有一种个体,有一定数量,分布于这个种群的各个角落,这类个体充当两代生命体之间知识传递的媒介。” ——刘慈欣《乡村教师》<br>(2013.09.10)</p><hr><p>还记得所谓大学保安的三个终极哲学问题吗:你是谁?你从哪里来?你到哪里去?<br>————最佳回答是:如来者,无所从来,亦无所去,故名如来。 (来自《金刚经》,游灵隐寺北侧永福禅寺偶得)<br>(2013.09.02)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>@向鹏达:<br>今天是图灵奖获得者姚期智的生日,把这条信息转发到五个群你将高分毕业,拿到名校offer,成为著名学者,迎娶女(男)博士,迈向人生巅峰……<br>(2013.12.24)</p><hr><p>提高肥胖标准有助于减少国民肥胖率……《洋葱电影》的那个梗记忆犹新啊……<br>@蔡小伦:<br>环保部门认为重污染持续太久可以放宽污染标准。。。这减少污染的方法简直宇宙第一炫酷机智吊炸天。、<br>(2013.12.08)</p><hr><p>突然想到一个非常坑爹的问题:一个“什么都不拿就可以杀人的人”和一个“拿什么都可以杀人的人”,哪个更厉害?<br>(2013.12.03)</p><hr><p>@赵艺琛Eason:<br>“约翰·霍普金斯大学的计算机教授Peter一直奉行“以考试最高分为100%、按分布曲线依次给出最终成绩”的政策。去年秋学期,他3门课程的学生全体在期末考试弃考,以全员零分挑战这一政策。教授思考再三,称赞了学生的团结精神,给了所有人A”。<br>(2013.12.03)</p><hr><p>@郭谦益:<br>著名科学家、第四世界科学院院士涂效灰指出,正能量理论将质能守恒定律推广到了任意领域,很好地解释了实验事实,体现了中国梦的巨大能量,理因获得诺贝尔物理学奖,是比超弦理论和M理论更有希望的万有理论。他还表示,先前有许多中国科学家智慧的结晶,比如宇宙全息统一论,都没能获得诺贝尔奖,这是令人十分遗憾的。<br>(2013.10.29)</p><hr><p>这就是传说中的高级口译人员吧~<br>@向雪银:<br>“周三,在习近平进入会议室时,高管们列队相迎。他对每个人微笑致意,一带而过地说感谢他们的到来。一名口译人员介绍每一位高管的姓名,并将他们的回应翻译给习近平。”老钱作何感想大笑(WSJ倒数第二段!<a href=\"http://rrurl.cn/vnYxq0\" target=\"_blank\" rel=\"noopener\">http://rrurl.cn/vnYxq0</a> )<br>(2013.10.24)</p><hr><p>@李博:<br>微博看到一条新闻【……一些游客要登大雁塔时却发现登塔的门票价格不同,淡季20元,旺季为30元,一些游客反映这样不合理。省物价局经过调研,同意西安市大雁塔保管所取消大雁塔淡旺季门票,实行统一门票价格,具体标准为每人每次40元】<br>(2013.10.18)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>看到quasi-girlfriend,为毛我想到的是Leontief-girlfriend, Cobb-Douglas-girlfriend, Perfect-Substitution-girlfriend(这个是闹哪样= =)然后就浮现出David.D.Lee的“simple, simple, very simple”<br>@学霸那些事儿:<br>有同学聊起他的potential girlfriend,我就发明了quasi-girlfriend、semi-girlfriend、virtual girlfriend、imaginary girlfriend、pseudo-girlfriend……<br>(2013.12.24)</p><hr><p>经济学的表述就应该是I’m not well-behaved.了吧……(应该是描述效用函数的。。。)<br>@柳芸:<br>机智的我终于知道”感觉整个人都不好了”怎么翻译了:”I’m not well defined…”<br>(2013.12.14)</p><hr>"},{"title":"瞎扯·2013·二","path":"2013/06/30/tweet-2013-2/","permalink":"http://sx349.github.io/2013/06/30/tweet-2013-2/","content":"<p>《瞎扯·2013·二》为《瞎扯》的第四辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>@沈二帆渣:<br>这片土地容不下和他们有太多格格不入的地方的人,容不下超出他们想象太多的想象,容不下太多和他们的价值观相悖的价值观,容不下与他们觉得“应有的状态”相去甚远的状态。每一个梦想都应该被珍惜,而不是在简单的“你有没有考虑过“之后的颓然落幕。一个伟大的梦想家不应该是一个瞻前顾后处处小心翼翼的人,同样一个处处瞻前顾后小心翼翼的人或许是一个伟大的、有着令自己满意的生活的”生活家“,但他永远不会成为一个梦想家,也不会成为一个众人口中的”innovator“。<br>(2013.06.02)</p><hr><p>@毕业生:<br>喜欢iPad你就去买,喜欢莱卡你就去赚,想喝港式奶茶你就过口岸,想当team leader你就努力学习然后努力去工作,想去威尼斯你就攒钱去,想爱谁你就去爱,想追谁你就去追,想到了就去做,拼命努力,拼命享受,忍着守着惦记着,青春就过去了,何必用40岁的心态过20岁的年华。<br>(2013.06.02)</p><hr><p>@苏大AIESEC:<br>我对成功的理解:把自己喜欢做的事做得尽善尽美,让自己满意,不要去管别人怎么说。————周国平<br>(2013.05.28)</p><hr><p>作为一个胸无大志的人,我希望我的一生这样度过:学习、旅行、玩耍(其实学习和旅行也都是玩耍)等我周游列国累了,就到偏远的乡下弄个院子,种果树、药草、花、竹子,养猪养鸡鸭,为了表示向王羲之致敬,鹅也是少不了的,蚕就算了,怕怕。然后读书、画画、写书、练习书法、学习木工做小盒子、雕刻木头和橄榄核、刻印章、用果汁造纸、用棒针和钩针织东西、绣花、编中国结、设计饰品、酿葡萄酒和米酒、腌泡菜、做果酱、烧陶瓷……啊,人生多么美好呀!——萧秋水<br>(2013.05.22)<br>(找到了原作者的博客,特此更新版权信息:转载自:<a href=\"http://www.xiaoqiushui.com/archives/7456.html\" target=\"_blank\" rel=\"noopener\">逍遥游·萧秋水</a>)</p><hr><p>“启超读近经,渐渐能解,(亦不能尽解,解者渐多耳)观《楞伽记》,于真如生灭两门情状,似仿佛有所见,然不能透入也。大为人事所累,终久受六根驱役不能自主,日来益有堕落之惧,(日夕无一刻暇,并静坐之时而无之,靡论读经)既不能断外境,则当择外境之稍好者以重起善心,兄之闲暇望如天上也。”(三月十日《与穗卿大师书》)<br>(2013.05.07)</p><hr><p>@严旎萍: 被科学松鼠会感动翻了:“如果每个人都是一颗小星球,逝去的亲友就是身边的暗物质。我愿能再见你,我知我再见不到你。但你的引力仍在。我感激我们的光锥曾彼此重叠,而你永远改变了我的星轨。纵使再不能相见,你仍是我所在的星系未曾分崩离析的原因,是我宇宙之网的永恒组成。”<br>(2013.04.22)</p><hr><p>定理:当且仅当≥是一致的时候,有限赢利集合A上的二元关系≥可由效用函数u:A->R来表示。<br>解释:A是一个选择的集合;如果对于A中的运算(1)两个元素总能比较大小(完备性)(2)A≥B B≥C 则 A≥C (传递性) (3)增加别的元素不会改变大小比较的结果(无关选择的独立性) 则称≥是一致的;效用函数u的存在意味着总可以给A中的每个元素赋一个实数值(另有定理证明赋值不是唯一的)。<br>取自《理性的边界:博弈论与各门行为科学的统一》<br>(2013.04.19)</p><hr><p>“你们今天是法官和检察官,但你们的子孙不一定是。如果没有法律和制度的保障的话,你们的子孙也可以被冤枉,也可能徘徊在死刑的边缘。”——自由主义,最大化最小原则。<br>@周梦宇:<br><a href=\"http://rrurl.cn/r7M6cm\" target=\"_blank\" rel=\"noopener\">http://rrurl.cn/r7M6cm</a> 真希望这只是黎明前的黑暗…<br>(2013.04.04)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>@北中医资讯:<br>用“这个、这个”造句。 小学生:这个是小猫,这个是小狗。 初中生:这个是醉翁亭,这个是岳阳楼。 高中生:这个是我过目不忘的萤火,这个是我十年一个漫长的打坐 大学生:这个、这个、二两饭。<br>(2013.06.19)</p><hr><p>围观了刷下限全程的热心美国网友陶先生称:“我活了38年,这帮数学家是我见过最没有下限的人。”<br>@学术状态帝:<br>下限已经被热心网友刷到六万了。。<a href=\"http://rrurl.cn/bD5fgg\" target=\"_blank\" rel=\"noopener\">http://rrurl.cn/bD5fgg</a><br>(2013.06.17)</p><hr><p>@中国工程院作业工程苏州研究所:<br>今日袁莲花教授将在本研究所签售新书《考后三个月》,共有10余个科目可供选择,包括高等数学、大学物理、线性代数、无机化学、量子力学等。每科分为90个分册,可供广大考生使用到大学开学前。<br>(2013.06.10)</p><hr><p>@韦龙杰 Jose:<br>我来普及一下江苏高考,和别的省市一样都考语数外,但是我们第二天早上要放假,下午考英语。第三天的时候就是所谓的1+1,各个城市都是不同的,徐州考的是挖煤和挖矿石,连云港考的是海运和挖水晶,宿迁考的是小龙虾养殖和酿酒,淮安考的是治理淮河和洪泽湖,盐城考的是晒盐和饲养麋鹿丹顶鹤,南通考的是炒文蛤和分辨各种方言,泰州考的是郑板桥会话和胡氏茶庄历史,扬州考的是风月史和大运河,镇江考的是酿醋和肴肉,苏州考的是园林设计和江南早点,无锡考的是糖醋排骨和拍电影,南京考的是民国史和治理秦淮河!<br>(2013.06.09)</p><hr><p>@王子元SFer:<br>Mansion 30 清华尊享海景房。【健康】东临东区体育中心,劲享运动生活。【品位】南临至善路,观雨看海,尽显王者心境。【食道】西临闻馨园清芬园丁香园,易享鲜活滋味。【善邻】北临苏世民书院,偷窥未来领袖生活。Mansion 30, 全北京最佳海景房不要100000一平米,只要750一年啊有木有!<br>(2013.06.04)</p><hr><p>没有学霸曹操,诸葛亮能搞定?<br>@阴瓜宁:<br>每当deadline临近时就告诉自己,想想诸葛亮,面对10w支箭的project,还不是拉着鲁肃最后一晚通宵搞定?<br>(2013.06.02)</p><hr><p>@戴晨旭: 我以后一定要在清华里开一个银行,经营两种通货:抬头为“清华大学”和“清华大学教育基金会”的发票,并设立买入价和卖出价提供两种通货的兑换服务。同学可以把日常的发票存入银行,我把发票贷给有需要的同学,从中赚取发票差,其中一部分返给同学做利息。<br>(2013.05.11)</p><hr><p>@李博:<br>我:"诶?C楼超市的煎饼换成紫米面啦?"室友:"对啊。是因为快校庆了吗?都要换成紫色的。。。"<br>(2013.04.19)</p><hr><p>数学系的同学听着,不要只相信数学分析……<br>@金融学会: 彼得·林奇:不要相信数学分析。股票投资是一门艺术,而不是一门科学。对于那些受到呆板的数量分析训练的人,处处都会遇到不利因素,如果可以通过数学分析来确定选择什么样的股票的话,还不如用电脑算命。<br>(2013.04.03)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>@蘇州情書:<br>在超市排队结账,前面一外地滴姑娘等着付款。营业员阿姨用标准的普通话说“夯么郎当10元。”。。。姑娘问 : “夯么郎当什么意思?”。。。老阿姨用更标准的普通话解释:“夯么郎当就是 一沓刮子。”<br>(2013.06.13)</p><hr><p>前两天中文沟通课请的HR嘉宾提供的两条面试建议:1. 不要看他投的简历,给一张A4纸让他现场写一份简历;2. 问问是否有一项兴趣爱好或者工作坚持了三年或者以上。<br>(2013.04.14)</p><hr>"},{"title":"墓地","path":"2013/04/09/novel03/","permalink":"http://sx349.github.io/2013/04/09/novel03/","content":"<hr><p>约瑟夫·K做了一个梦。那天天气很好,K想去散散步,可当他刚刚迈出两步,就已经到了墓地。</p><p>“这是谁的墓地?”K想着,向墓地中间走去。</p><p>墓地的正中是一尊奇特的墓碑,K仔细看了一眼,墓碑上刻的是他的名字,墓碑前的土还没有翻过的痕迹。</p><p>K却并没有觉得什么奇怪,墓碑好像就应该在那里一样。他仔细地抚摸了一遍墓碑,名字是很精致地刻在上面的,石屑也被清理得很干净,没有一点残留。墓碑上没有什么特别的装饰,足以说明将被埋葬的人无足轻重。</p><p>一行黑衣人从他身后出现了,K并没有回头,但是他还是看见了,好像他脑后长着眼睛一样。四个穿黑衣的伙夫抬着一个披上白布的重家伙——看来应该就是装有K的棺材——走在人群中间。没有人说话,或者也许只是离K隔了太远了而听不见罢了。</p><p>K觉得这些人应该是在朝着他的墓碑走来,但此时那行人像是越走越远了,就在K以为那行人快要走出墓地的时候,却发现他们其实已经到了自己背后。</p><p>他们认出了K,或是没有认出K,这不太重要。没有人理睬K,K也默默地走到一边以免挡着这些人。几个男人——伙夫,以及像是临时雇佣的仆役——开始把墓碑前面的土翻开,女人开始哭。与其说是哭,倒不如说只是发出些呜呜的声音。</p><p>K终于还是耐不住了,他又走近了这些人。棺材上刻得也是他的名字,同样的精致的手艺。女人仍旧在哭,有一个最靠近K的人停手了。</p><p>“是他!他在这儿,”这个人指了指棺材,“也在这儿,”他又指了指K。</p><p>另一个人没有抬头,继续挖着坑,声音传了过来:“他想要自己来埋葬自己,还是让别人来埋葬自己?”</p><p>“把铲子递给他吧!”这是第三个人的声音,带了些玩笑的戏谑。K想找到是谁在说话,所有人都显得模模糊糊,声音也好像从很远的地方传了过来似的。</p><p>没有人再多说一句话。K也很无趣,于是决定凑到自己的棺材边上去看看,或者说去听听——因为棺材看起来是钉上了的,所以看是看不见什么的。K听到里面传来平稳的呼吸声,如此的平稳以致于有了一些催眠的功效。</p><p>坑好像挖完了,因为其他男人已经向棺材这边围了过来。女人也不再哭,只是仍在互相擦着眼泪。K不知道要不要走开,但是有一个男人——应该是之前第一个说话的人,虽然K也没有什么把握——对他说:“搭把手吧。”</p><p>K没有伸手,也没有表示反对。显然抬棺材的两边人数不一样,K还是决定上前帮一把。</p><p>他帮着抬起了棺材,里面的呼吸声变得有些急促,好像开始乱动——这不是一个好的征兆。不过由不得K多想,因为其他人已经朝坑走了过去,他也跟紧了脚步。里面的挣扎有些明显,K第一次感到有些害怕,但是他的脚仍跟着。</p><p>几个人终于停了下来,一起松手,直接把棺材扔进了坑里,尽管这是不符合常规的做法。</p><p>K感到第三个声音又在远处响起了,但是同时响起的一声尖叫使他完全没有听清,尖叫声分明来自于那个还在不断下落的棺材。</p><p>约瑟夫·K一下子坐了起来。只是个梦罢了,他这么想着。</p><p>(人文知识竞赛的续写,正好看完《城堡》,但是感觉不得要领,随手瞎写,见笑。)</p><p>(倒数第二段删去了一句话,“装睡的人不仅叫不醒,最终也往往被自己所埋葬。”)</p><hr><p>(2013.04.09 @ 人人 @ 《杂记FP——墓地》)</p><hr>"},{"title":"瞎扯·2013·一","path":"2013/03/31/tweet-2013-1/","permalink":"http://sx349.github.io/2013/03/31/tweet-2013-1/","content":"<p>《瞎扯·2013·一》为《瞎扯》的第三辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>We leave such questions to philosophers, who have thousands of years of experience in discussing the really big questions about life, the universe, and everything, not flnding any answers to any of them. (谈人工智能)<br>(2013.03.05)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>@中国工程院作业工程苏州研究所: 翻阅史料,看到了去年二月本研究所几位教授的对话,很有意思。某教授说:“小偷去莲花家,结果小偷被作业压死了。”有人补充:“小偷去叟哥家,结果被做了。”更多人加入进来:“小偷去杨总统家,结果两个人开始一起做。”最后有人说:“小偷去周以宏家,结果再也做不起来了。”<br>(2013.03.30)</p><hr><p>@学傻状态帝:<br>传闻物理竞赛能否获奖都是由天来决定的,因为——物竞天择……<br>(2013.03.10)</p><hr><p>笑寒在讨论“如果未来我成了一个瘾君子”……<br>(2013.03.05)</p><hr><p>这是在情人节怀念无性繁殖么……<br>@我们都是理科生:<br>克隆羊多莉安乐死十周年纪念。<br>(2013.02.14)</p><hr><p>他也不是拿菲尔兹奖中年纪最小的,只不过是次小的……<br>@杨宽:<br>而且他10岁时参加IMO只拿了铜牌,11岁时只拿了银牌…<br>@刘正阳Lozy:<br>陶哲轩的数学生涯也并非一帆风顺。9岁多时,他未能入选澳大利亚队,去参加国际数学奥林匹克竞赛。。。<br>(2013.01.17)</p><hr><p>adblock很赞,但是有时候(极少)会坑人:我在线背单词的时候,一到advertisement这个单词就会没有声音,我把adblock一关就可以了………………<br>(2013.01.11)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>如果是A的话应该会说“分特则”吧……<br>@田紫东:<br>苏州话18级听力:“内啊帮各个宁分特了?”“FEN 了” A:分了 B:没呢<br>(2015.03.25)</p><hr><p>@学术状态帝:<br>【真正的高帅富论文】学术期刊Gene上面有一篇文章叫“Evolutionary aspects of gobioid fishes based upon a phylogenetic analysis of mitochondrial cytochrome B genes”,该文研究虾虎鱼的线粒体细胞色素b基因。这篇文章发表于2000年,所属单位既不是大学,也不是研究院,而是一个叫“The Imperial Residence”的地方。作者有很多位:Akihito, Iwata A, Kobayashi T, Ikeo K, etc. 大家仔细看,这些生物学家中,唯独一作只有名而没有姓,这是因为,这个一作叫做明仁,他是日本天皇。<br>(2013.03.16)</p><hr><p>@张瑞祥: 《孙子算经》中曾记载这样一个著名问题:“今有物不知数,三三数之剩二,五五数之剩三,七七数之剩二,问物几何”?惊人的是,23,233,2333,23333,……,都是这个问题的解。<br>(2013.02.28)</p><hr><p>@吴语: 在苏北一带有个现象,有些人把睡觉说成“上苏州”。这要追溯到朱元璋建明后,为打击报复江南百姓支持张士诚,将45万的苏、沪、杭、嘉、湖的富户,集中在苏州阊门,全数流放驱赶至江北任其自生自灭,且规定“永世不得回原籍”。有人不惜乔装成乞丐也要回乡,但大部分人至死都无法回到家乡,于是只能把魂牵梦绕的家乡寄托在梦里。请搜索关键词:【洪武赶散】<br>(2013.02.27)</p><hr><p>上了微积分和史纲之后想到的: 其实纪传体和编年体史书就是历史的两个不同的偏导数……<br>(2013.02.25)</p><hr><p>突然发现这些都是没有拿过诺奖的。拿过诺奖的咱们都还没有学到……<br>@范卓宇: 经济学家们在比成就。魁奈:我画出了经济表!李嘉图:我有劳动价值论!萨伊:我乃供给学派开山始祖禅师!马歇尔:壮哉我剑桥学派!穆勒对马歇尔说:我承前启后,没我没你••••••斯密:我期末划了三道题重点,你们几道?众人皆叹服。角落里的马克思笑了,说:下学期见分晓!<br>(2013.01.17)</p><hr><p>应该是O(NM)啊……复杂度只考虑N是肿么回事啊!模式串长度都不用考虑了么……<br>@廖宁林:<br>信科真是大神多,今天第一次听说KMP里面进行匹配的字符串只允许有26个字母,而且算法复杂度是O(NlogN)的 ,还真是涨知识呢。<br>(2013.01.04)</p><hr>"},{"title":"瞎扯·2012","path":"2012/12/31/tweet-2012/","permalink":"http://sx349.github.io/2012/12/31/tweet-2012/","content":"<p>《瞎扯·2012》为《瞎扯》的第二辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>乔布斯出走苹果之后,创办NeXT公司,而NeXT公司的技术,也是目前苹果Mac OS X和iOS两大操作系统的核心基础。<br>蒂姆·伯纳斯·李用CERN经费买了一台NeXT电脑,设计出了最早的WWW(万维网World Wide Web)<br>————有些时候,一个人的影响不仅仅只有表面看到的这些,而是千丝万缕地渗透进了一切。<br>(2012.02.26)</p><hr><p>作者用现实主义的笔触,生动而细致地描绘了一个中产阶级子弟苦闷、彷徨的精神世界,真实地揭露了当时社会精神文明的实质。人活着除了物质生活外,还要有精神生活,而且在一个比较富裕的社会里,精神生活往往比物质生活更为重要。……物质生产发展得很快,生活水平迅速提高,中产阶级的人数也激增。但广大人民的精神生活却越来越贫乏、空虚……有些人粉饰太平,过着浑浑噩噩的日子;另有些人看不惯庸俗、虚伪的世道,想要反抗,却又缺乏光辉的理想,找不到一条光明的出路。<br>——百度百科:麦田里的守望者<br>(2012.06.23)</p><hr><p>人生的本来意义,就是从无路处走出路来;人们往往过分关注别人在有路的地方走得如何,固然走猫步比匍匐前进看起来更优雅,两者却一样地只是重蹈前人的覆辙罢了。<br>(2012.07.24)</p><hr><p>@调大愚:<br>人生就像是一匹用善与恶两种纱线交错织成的布,要不是有过失的鞭挞,我们要得意忘形了;要不是有善行的庇护,我们又要完全绝望了。 ——莎士比亚<br>(2012.11.23)</p><hr><p>疯子头人:孩子!/儿童!/为了数学!/为了数学!/大家前进!/大家抬着棺材!/棺材太沉太沉!/棺材里不止一人!/为了数学!/大家前进!/大家往一块儿走!/大家往前走!/排成一行!/抬着棺材!/鲜血染红!/谁要问我!/走火入魔!/谁要问我!/牵肠挂肚!/为了数学!/抬着棺材!前进!<br>——海子《太阳·弑》第三幕第二十四场<br>(2012.12.05)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>苹果的LOGO果然不是纪念图灵的: I designed it with a bite for scale, so people get that it was an apple not a cherry. Also it was kind of iconic about taking a bite out of an apple. Something that everyone can experience. It goes across cultures. If anybody ever had an apple he probably bitten into it and that’s what you get.<br>(对苹果图标最初设计师的专访 <a href=\"http://rrurl.cn/k257el\" target=\"_blank\" rel=\"noopener\">http://rrurl.cn/k257el</a> )<br>(2012.02.07)</p><hr><p>“林书豪也因此成为哈佛大学自1953年后首位进入NBA的球员”。我看了两遍,确认不是把1993年给打错了……<br>(2012.02.16)</p><hr><p>维基百科“2”词条 数学性质:是一个等于2的数<br>(参见 <a href=\"https://zh.wikipedia.org/w/index.php?title=2&diff=next&oldid=21896208\" target=\"_blank\" rel=\"noopener\">https://zh.wikipedia.org/w/index.php?title=2&diff=next&oldid=21896208</a> )<br>(2012.07.16)</p><hr><p>@袁钟鸣:<br>为了实现我儿子能参加苏高中1000年的梦想,我得在7年后生孩子,6年后怀孕,5年内结婚,2年内必须找到稳定的结婚对象……压力山大<br>(2012.10.19)</p><hr><p>那么清华的饭卡是南开的公司的……<br>@袁文:<br>我刚刚才知道南京地铁站里打印打折券的机器因为是东大人设计的所以东大校园一卡通也可以刷。。尼玛这都可以啊<br>(2012.12.04)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>根据IUPAC元素系统命名法,第217号元素按规则是2-1-7(bi-un-sept),最后再加上词尾-ium(代表“元素”的意思),因此其临时元素符号是——BUS。<br>(2012.01.04)</p><hr><p>一直以为coin和coincidence有着某种不可告人的秘密……但是coin词源是拉丁语的cuneus,进而演化为盎格鲁法语的coing;而coincidence是coincide的名词,coincide是co-前缀(一致的)加上-incide,-incide是 in- + cadere,词源是拉丁语cadere(坠落fall)合起来是fall into,也可以认为是 in + 后缀-cid(发生)。除了拼写恰好一样以外没有任何关系……<br>(2012.02.21)</p><hr><p>数学课上,老师说:This method applies to integers, say, 40000.<br>底下有人反对:That’s not an integer. It exceeds 32768.<br>(2012.02.24)</p><hr><p>刚看到科学松鼠会发的一个段子:“USB线插不进端口;转180度还是插不进;再转180度插进去了;USB线存在于四维空间中,证毕。”我想说,泡利早就发现了自旋量子数这个诡异的东西,±1/2的自旋量子数说明电子要自旋两周才能回到原来的位置,所以,USB线可能只是正好有个1/2的自旋量子数而已…………<br>(2012.07.27)</p><hr><p>离散课上全班和老师争论“昨天北京下雨了”是不是命题,因为“昨天”不确定……<br>(2012.09.30)</p><hr><p>@刘琳:<br>图书馆外,一学霸看着一排倒了的自行车说:这说明可导必连续,连续不一定可导……<br>(2012.10.14)</p><hr><p>@袁凯:<br>不要和经济学院的同学打架,因为他们有一只看不见的手。<br>(2012.11.01)</p><hr><p>一个绝对理性的经济学家在看电影的时候会发生什么?如果他买了一张电影票,边上有人想抢,他会恭恭敬敬的递过去,因为他觉得对方对电影票的评价比他高,他不应该造成无谓损失。而且,即使这样的抢票事件一再发生,他也会继续买电影票,因为那些都是沉没成本,而他自己对于电影的评价还是高于电影票的价格,因此买票对于自己是有收益的。伤不起的经济学家……<br>(2012.11.17)</p>"},{"title":"盗泉者说","path":"2012/06/19/welcome/","permalink":"http://sx349.github.io/2012/06/19/welcome/","content":"<p>“盗泉”一词,最早见于春秋战国时期尸佼的《尸子》一书:“(孔子)过于盗泉,渴炙而不饮,恶其名也。”随后,历朝历代的文人墨客,便称不义之财为“盗泉”,以不饮盗泉表示清廉自守,不苟取也不苟得。</p><p>据说,这眼泉水就在今山东省新泰市石莱乡道泉峪村。民国13年,有当地乡绅将“盗”字改为道德的“道”字,一直沿用至今。但这举动却被鲁迅先生在《说面子》一文中大加挞伐,他借长谷川如是闲的说法:“古之君子,恶其名而不饮,今之君子,改其名而饮之。”</p><p>因此,以“盗泉”为名的第一层含义,就是“宁当真小人,不做伪君子”。“盗泉”的水也是水,没有水喝的时候,按孔老夫子的做法就活活渴死了;按伪君子的做法,就对外宣称这是“道泉”,喝者不仅不会玷污自己的名声,而且可以得道升仙;但我宁可大口去饮,不去做什么掩饰,这才是真性情。</p><p>不过,“盗泉”在这里还有第二层意思。朱熹写过“问渠哪得清如许,为有源头活水来”,但这活水从哪儿来?山上有泉水,那么从山上引下来便是。倘若没有,那么恐怕还是得“盗”一些泉水过来搅搅自己的死水,否则就没有生气了。</p><p>如今我国的数学发展不可谓不兴盛,但是看看外界的评论,大学之前上奥数,奥数被批;大学之后上高数,高数被批。数学应当是一门充满乐趣的学科,而不是局限于一种工具或是答题模式。其实倘若看过一些试题就会发现,奥数的发展本身是向着趣味化的,但是人们对于奥数所求的太多,使得它变了味。</p><p>因此,有必要从国外“盗”一点“泉水”,来重新求得数学的本来面目,希望参与者洗刷掉功利的目的,而回归其质朴的乐趣当中。</p><p>(2012.06.19 @ 百度贴吧)</p>"},{"title":"Terraria 纯小白指南","path":"2012/01/20/terraria/","permalink":"http://sx349.github.io/2012/01/20/terraria/","content":"<p><strong>Kill all trees</strong><br><strong>砍树伐林 小白有责</strong></p><p>The first thing you need is wood. In the beginning it’s a good idea to build your house close to where you start. You will respawn in the same spot each time you die until you build a bed. One of the main frustrations of building by the spawn is that it puts you right next to the Guide NPC. He’s supposed to offer advice to new players, but he ends up wondering in and out of your home at a whim, letting zombies and flying demon eyeballs in at night.<br>你所需要的第一样东西是木头。作为新手,最好将你的房子建在你的出生点附近。因为在你建造一个床之前,你在每次死亡后都会在这个出生点复活。但是这样做有一大弊处,就是房子恰好在向导NPC边上。他本应当是给新手提供帮助的,结果却在无所事事地走进走出你的房子,并且还会在晚上把僵尸和飞着的恶魔之眼放进来。</p><p>To get wood, chop trees with the axe. It won’t look as though anything is happening for the first few swings, but keep at it. After enough hacks targeted at the same location, the entire tree will explode and shower wood down upon you. Aim at the very base, that way you won’t leave annoying stumps behind, which means you’ll get all of the wood the tree has to offer, and the space you clear will be flat and easy to build on.<br>想要得到木头,你要用斧子砍树。一开始的几下看起来什么都没有发生,但是坚持下去,当你在同一个位置砍了足够多次以后,整棵树会裂开,并且在你身边落下一大堆木材。在砍的时候务必对准树的最底部,这样你就不会在身后留下任何令人厌烦的木桩,既得到了这棵树能够提供的全部木材,同时清理出的空间也足够平坦方便建造房子。</p><p>Once you have wood, you can make a workbench. Pressing escape brings up your inventory menu. Below this, you’ll see a list of the items you can craft with the materials in your pockets. Clicking on these items a couple of times will deposit them straight into your inventory. Once you’ve got a workbench, drop it into your task bar. Once you’re back in the world, select the crafting table, with your mouse, or with one of the 1-9 hotkeys, and left click to plop it down on a flat bit of ground.<br>当你拥有木头之后,你可以建造一个工作凳。按Esc键可以显示你的物品栏,在菜单的下方,你可以看见一系列你能够用包里现有的材料制造的道具。在上面点几下就可以直接把它们放进你的物品中去。当你拥有一个工作凳之后,把它放进最上层物品栏,并在回到世界中后把它放在一块平坦的地面上。</p><p>You’ll find yourself interrupted constantly by Slimes. They hop around the surface in daylight hours, generally making a nuisance of themselves. They’re as useful as they are irritating. Bludgeon them to death with your pickaxe (it has a faster swing and is more effective than a wooden sword against annoying hopping enemies like slimes). Dead slimes drop flammable gel and money. Gel can be combined with wood to create torches, invaluable for the cave exploring we’ll be doing later on.<br>你会发现你经常被史莱姆打扰。白天,它们在地表上跳来跳去,让人感到厌烦。不过,它们尽管这么讨人厌,也还是有它们的用处的。用你的鹤嘴锄连着击打它们直到它们挂掉(鹤嘴锄攻速更快,对抗像史莱姆这样跳来跳去的敌人时比木剑更加有效)。它们挂掉之后会爆出可燃的凝胶和铜币。凝胶和木头组合可以制造火把,这对于我们之后探索洞穴来说是很宝贵的。</p><p>If you’re worried about all the mass deforestation you’re doing, you can plant acorns gathered from chopped trees. Given time, they will grow into a new forest.<br>如果你对于你的毁林行为感到担忧,你可以将砍树时收集的橡实种起来,到一定的时候,它们又会长成新的森林。</p><hr><p><strong>Build a home</strong><br><strong>金窝银窝不如木头窝</strong></p><p>If you open up your inventory while standing near your new crafting table you’ll be able to craft more advanced items. Before you start building, make sure you have lots and lots of wood. All of your starting items will be made out of the stuff. First, craft a sword and a hammer. Then, craft lots and lots of “wood wall” blocks.<br>如果你在你的工作凳边打开物品栏,你能够造出更多高级些的物品。在你开始造房子之前,先确认你有非常多非常多木材,因为你所有的初始物品都会用到它。首先,制造一把木剑和一把锤子。然后,制造一些“木头墙纸”。</p><p>A safe structure in Terraria must have a background wall. Placing wood walls on the landscape will create a wood panel backdrop. Once enclosed, this will stop monsters from spawning inside your house. You can place unprocessed wood directly onto the landscape to build walls. Create a rectangular background as big as you can, and then wall it off with wood blocks. If you build walls using dirt blocks, it won’t count as a home, and you’ll have plenty of awkward night time encounters with zombies appearing in your bedroom. Also note that your structure won’t count if its frame is narrower than 10 blocks, or shorter than eight.<br>在Terraria中,一个建筑必须有背景墙纸才是安全的。把木头墙纸铺在空白的地方,形成一个木质的矩形背景,这样做可以防止怪兽在你的房子里刷新。你可以把没有处理过的木头直接放在地上来建造墙。铺一个尽可能大的背景墙,然后用木头给它筑墙。如果你用土墙筑屋,这不会被算作一个房子,而且你会有很多个与僵尸在卧室里共舞的惊魂夜。同时也要注意,如果你的建筑如果窄于10块砖或矮于8块砖,那么也不算是房子。</p><p>Next, craft a couple of doors. Use your axe to gouge three block high gaps at each end of your structure, and plant a door in each one. Doors can be opened and closed with a quick right click. When closed, they’ll keep any creature out, unless there’s a blood moon in the sky.<br>接下来,制造几扇门。用你的斧子在墙上准备安置出入口的地方凿开三块砖大小的口子,然后在每个口子上放一道门。右击一下门就可以将它打开或者关闭。当门是关闭的时候,能够把所有的生物阻挡在外面,除非空中是血月高照。</p><p>There are two ways of telling whether there’s a blood moon out. The first is the colour of the moon itself, the second will be the zombie apocalypse erupting around you. Blood moon creatures can smash through doors. Your best bet in that circumstance is to hole yourself up in a corner, swinging wildly at the forces of darkness with your best weapon.<br>有两种方式辨认是不是血月出现了。第一种是看月亮本身,另一种是看你周围暴动的僵尸。血月下的生物能够冲破门。在这种情况下,你最好的选择是躲在角落里,疯狂摇动你最好的武器来抵御邪恶力量的入侵。</p><p>If you’re building a two storey house, the best way to get between floors is to leave a gap in the floor/ceiling, and replace it with a wood platform. Pressing down when standing on a wood platform will let you drop down safely to the floorbelow.<br>如果你要盖的是双层的小楼,最好在一楼的天花板(也即二楼的地板上)留个空,然后在那里摆上木制平台。在木质平台上按下键可以让你安全地掉到下面那层。</p><p>Finally, craft a few torches, assign them to your taskbar, and plant them on thewalls with a quick left click. Voila, you have a house! Later, friendly NPCs will move into abodes, if they’re big enough, well lit, and have a table and chairs inside. Different NPCs will move in when different criteria are met. Once you have fifty silver coins, a merchant will appear. If you discover a crystal heartundergound, smash it and then activate it, a nurse will move in. Check out the Terraria wiki for more information on the different NPCs.<br>最后,制造一些火把,把它们放在你的最上层物品栏,然后鼠标左击把它们放在墙上。啊哈!你现在有一座房子啦!接下来,友好的NPC会搬进这些住所,只要它们够大、有光源、有桌子和椅子在里面。不同的NPC要搬进去需要满足不同的条件。当你有五十个银币时,商人就会出现。当你在地下发现一个水晶之心,打碎并激活它,护士就会搬进来。你可以在Terraria Wiki上找到更多关于这些不同NPC的信息。</p><p>Your next focus should be stone. It’s an extremely common material, and you’re likely to find outcrops of the stuff on the surface. Wander for a few minutes in either direction and you should find enough to let you craft a furnace.<br>你接下来的注意点是石头。这是一个极其常见的材料。你可以在地表上看见露出地面的石头。四处兜兜你就能找到足够多的石头来制造一个熔炉。</p><p>You’ll need 20 stone, 4 wood and 3 torches to make one. You can place it in the environment in the same way as a crafting bench. With a furnace in your home, you’ll be able to smelt more precious ores like iron, gold and silver.<br>你需要20个石头、4块木头、3个火把来制造一个熔炉。你可以像放置工具凳一样把它放在地上。当你家中有一个熔炉时,你就能够熔化一些更贵重的矿物,比如铁、金和银。</p><p>Now you’ve got a base of operations, the exploration can begin. The most precious ores are miles from the surface. You’ll have to head deep underground to get your paws on the best building materials. You can either wander the world in search of a natural sinkhole, or start digging within sight of some useful ore. Stone and copper are fairly common, but to begin with you’ll want to keep your eyes open for iron. It’s easy to miss because it looks quite similar to stone. Veins of iron are slightly darker and browner than stone. Here’s a picture to demonstrate.<br>现在你已经有了一个工作基地,可以开始探索了。最珍贵的矿物距离地面很远,你必须深入地下来获得最好的建筑材料。你也可以在世界中乱转来发现一些自然形成的下行洞穴,或者开始挖掘一些就在眼前的实用的矿物。石头和铜矿非常常见,但是你更要注意的是铁矿。它们很容易被忽视,因为它们看起来像石头,铁矿脉比起石头来只是稍微暗一些。下面的图片可以作为示例。</p><p>You can smelt 3 blocks iron ore into one iron bar. Iron bars can then be used tocraft many items, but an anvil and a new pickaxe are good places to start. You can make an anvil out of five iron bars at your workbench. You’ll need 12 iron bars and 3 bits of wood to craft an iron pickaxe. This will let you mine faster, speeding up the game considerably.<br>你可以把3块铁矿熔为一块铁条。铁条可以用来制造很多道具,但最好先做一把铁砧和一把新的鹤嘴锄。你可以在你的工作凳上用5块铁条来制作一个铁砧,用12块铁条和3块木头来制造一把铁质鹤嘴锄。这可以让你挖的更快,大大加快游戏的进程。</p><p>While underground, gathering iron, it’s a good idea to swipe any cobwebs you see littering the place. This will get you silk. You can combine five silk with 20wood to make a bed. Placing a bed in a valid home and then clicking on it will let you set a new spawn point. This is useful for a number of reasons. If you’reextremely far underground, and don’t want to travel back and forth between the surface, you can build a bed underground as a base of operations for mining.<br>在地下挖铁矿的时候,击打那些杂乱地遍布着的蜘蛛网是个不错的主语,它们可以给你带来丝绸。你可以用5片丝绸和20块木头来做个床。把床放在一个能够算数的房子里,然后右击它,你就能设定一个新的重生点。这很有用,尤其是当你深入地下很深而且不想在地表地下来回奔波的时候,你可以在地下放张床来作为挖掘矿产的工作基地。</p><p>Even on the surface, it’s a good idea to move away from the default spawn point. Firstly, because it’s a good way to get away from the Guide, secondly, because of goblin invasion. In Terraria, a horde of goblins will sometimes come charging across the map. No matter where your bed is situated, they will always attack the default spawn point, so planting a bed a few hundred metres in one direction or the other should let you escape the brunt of their charge.<br>即使在地表上,远离出生点也是个好主意。首先,这可以远离向导,其次则是因为哥布林入侵。在Terraria中,有时会有一群哥布林冲过整个地图。无论你的床在哪里,他们总会攻击最初的出生点,所以在出生点一百米开外造个房子可以使你躲过哥布林的冲击。</p><p>As soon as you start amassing iron, it’s a good idea to build a chest. You’ll need two iron bars and eight chunks of wood to build it, but it’ll give you somewhere to safely stow excess ore and items that you find on your travels. It’s also a good idea to deposit money in your chest whenever you can. When you die in Terraria, you lose half of the money on your character at the time. If your fortune is safely stashed at home, it won’t suffer if you, say, wander too close to the King Slime, or fall off a particularly high cliff.<br>当你开始积累铁矿的时候,造一个藏宝箱。你需要2块铁条和8块木头来建造它,但是它可以让你安全地存放在旅行中你发现的富余的矿产和道具。还要记得无论何时,最好将钱存在藏宝箱里。当你在Terraria中死亡的时候,你会丢掉你的角色所携带的一半的钱。如果你的财产都被安全的储存在家里,它们永远不会遭殃,即使你凑巧走得离史莱姆王太近了或者从一个相当高的悬崖上摔下去。</p><hr><p><strong>Arm yourself</strong><br><strong>枪杆子里出XX</strong></p><p>As you burrow deeper into the earth, you’ll encounter stronger enemies. As soon as you come up against a Mother Slime or a group of charging skeletons, you’ll probably find that your manky wooden weapon literally doesn’t cut it anymore. The next step is to craft better weapons, and, if you’ve struck lucky with enough iron or silver, a suit of armour.<br>随着你挖掘的逐渐深入,你会碰到更强的敌人。当你对抗一只史莱姆之母或者一群冲过来的骷髅的时候,你会发现劣质的木质武器简直砍不动它们。所以下一步就是制造更好的武器,甚至如果你运气够好搞到足够多的铁矿和银矿,你还可以制造一身盔甲。</p><p>There are two types of sword you can craft, a shortsword or a broadsword. A broadsword is the recommended choice if you have the eight iron bars needed to craft it. It has a slow, wide swing that can knock back low level enemies like zombies and skeletons, but is also good at taking out the underground worms that leap out of the walls. The shortsword’s quick, stabby motion is nowhere near as effective against these horrible creatures.<br>你可以制造两种剑,短剑或者宽剑。如果你有8块铁条,那么更推荐去制造一把宽剑。它摇动起来缓慢但范围更宽,能够击退像僵尸、骷髅这样低级的敌人,对着那些从墙里蹦出来的地下蠕虫发泄的时候也不错。另一方面,短剑的快速刺击对这些可怕的生物毫无作用。</p><p>Armour takes a lot of ore to craft, which makes it a luxury when starting out. If you’ve got a ton of copper, it’ll take 15 bars to craft a helmet, 20 to craft some greaves and 25 for a suit of copper chain mail. For iron armour, you’ll need 20 bars for a helmet, 25 for some greaves and 30 for body armour.<br>制造盔甲需要很多矿产,因此开始时这是件奢侈品。如果你有相当多的铜矿,用15块铜条来造一个头盔,20块来造几条护腿,25块来造一身铜质锁子甲。对于铁质盔甲而言,头盔需要20块铁条,护腿需要25块,而穿在身上的需要30块。</p><hr><p><strong>Explore</strong><br><strong>路漫漫其修远兮</strong></p><p>You’ve got a house. You’ve got armour. You’ve got a sword that even a Mother Slime would fear. Now the game begins.<br>你有了一座房子。你有了盔甲。你有了史莱姆之母都害怕的剑。现在游戏才刚刚开始。</p><p>Once you’re established, it’s time to head out into the big wide world. There are numerous different environments you can discover as you travel further from the centre of the world. One of the most common is Corrupt lands. These are pink and grey areas populated by spiky brambles and flying Soul Eaters. They are not pleasant. Corruption also has a habit of growing, infecting surrounding lands and spreading poisonous mushrooms.<br>当你安定下来后,是时候走进这个阔大无垠的世界了。从世界的中心出发,这里有无数不同的环境等着你去探索。其中最常见的是腐化之地,这是带刺的荆棘和飞行着的噬魂者所居住的粉色或灰色区域,它们并不友好。腐化之地同时有着扩散的特性,它们影响着周围的地面,同时传播有毒的蘑菇。</p><p>If you encounter Corrupt land near to your house, and want to stop its approach,dig a block or two deep and build a stone wall, or plant a sunflower. You can beat back the Corruption by digging the ground away, or by sprinkling purification powder on the ground.<br>如果你的房子附近有腐化之地,你可以挖一两个深坑并造一堵石墙,或者种一株向日葵来阻止它靠近。你也可以通过挖开地表或者撒上净化粉来使腐化之地退散。</p><p>As you explore underground, go out of your way to get to as many chests as you can. Underground chests hold special items and accessories that can’t be crafted. These include items that will let you breathe underwater, or, if you’re extremely lucky, rocket boots.<br>在探索地下的时候,偶尔偏离正路去开启尽可能多的宝箱。地下的宝箱中装有不能够制造的物品和饰品。这包括让你能够在水下呼吸的道具,甚至如果你足够幸运,你还能搞到火箭靴。</p><p>Crystal hearts are also extremely valuable. These can be found in their own cavenetworks underground. They can be smashed with a hammer and then consumed to increase your maximum life.<br>水晶之心也极其有价值。它们可以在独立的地下洞穴网络中被发现。你可以用锤子打碎它们,然后使用它们来增加你生命的最大值。</p><p>You may also encounter underground jungles as you continue to explore the depths. These contain golden jungle shrines that protect special items, but look out for swarms of giant hornets. If you burrow deep enough you will eventually encounter the underworld. This is rife with hellstone that can only be mined with a Nightmare pixkaxe. From here you can steal a hellforge, and use it to craft some of the best items in the game.<br>当你探索深入时你也会碰到地下森林。这里有金色的森林圣坛保护着特别的物品,但你得注意可能有成群的大黄蜂。如果你挖的足够深你会到达地狱,这里遍布着只能用噩梦镐才能挖掘的地狱石。你能从这儿偷到一个地狱熔炉,并用它来制造游戏中一些最好的道具。</p><p>There’s so much more to Terraria than this introduction can cover. We’ve not even mentioned floating islands,the huge boss monsters you can summon, or the enigmatic dungeon that spawns at the edge of every generated world. Hopefully you’ll now be equipped to go out explore, and discover Terraria’s best secrets yourself.<br>Terraria还有很多东西,这篇介绍不能完全涉及。我们甚至没有提到浮空岛、你能够召唤的那些大BOSS、或者是只在世界的边缘出现的神秘莫测的地下城。现在,你已经为出发探索做好了准备,亲自去发现Terraria中最神奇的秘密吧。</p><hr><p>(2012.01.20 @ 百度贴吧 @ 《【重译】Terraria 纯小白指南》)</p><hr>"},{"title":"瞎扯·2011","path":"2011/12/31/tweet-2011/","permalink":"http://sx349.github.io/2011/12/31/tweet-2011/","content":"<p>最近在清理人人,以防人人药丸……<br>《瞎扯》将甄选我发(或转发)的状态,包括段子、冷知识、摘抄等等。<br>《瞎扯·2011》为《瞎扯》的第一辑。</p><blockquote class=\"blockquote-center\">摘抄</blockquote><hr><p>【强奸民意】指统治者把自己的意愿强加在人民头上,却硬说是人民自己的意愿。<br>(<dfn title=\"2010.12.30\">2011.01.-02</dfn>)</p><hr><p>另外一个故事是出自马令《南唐书》,有:“冯延巳作谒金门曰:‘风乍起,吹皱一池春水。’中主曰:‘干卿何事?’对曰:‘未若陛下小楼吹彻玉笙寒也。’”冯延巳这个马屁拍得李璟龙颜大悦,冯延巳人品不足道,同他一起为相的孙晟骂他“谄妄险诈”,由此可得一斑。<br>(2011.05.23)</p><hr><p>——黄庭坚:春归何处?<br>——贺铸:只有春知处。<br>(2011.07.05)</p><hr><p>CCTV里放动车被拉走的时候,拉车用的是曾经遍及大江南北,现在早已销声匿迹的“绿皮车”。速度还是安全,这不应当是一个问题。<br>(2011.07.24)</p><hr><p>米兰达权利(Miranda Rights):”你有权保持沉默,不必回答任何问题。如果你开口说话,那么你所说的每一句话都会在法庭上作为呈堂证供。你有权聘请律师,并可要求在询问的过程中有律师在场。如果你请不起律师,我们将免费为你提供一位公设辩护律师,会在审判过程中的任何阶段陪同你。”<br>(2011.12.13)</p><hr><blockquote class=\"blockquote-center\">段子</blockquote><hr><p>——我:明天考完三门就只剩两门了~<br>——我妈:你不说后天考完两门就全考完了。<br>(2011.01.18)</p><hr><p>我在想能不能这样:找两三个班的人到大街上,每人拿一本导学站在路边上做,最好把路都给堵上了,而且个个都要神情肃穆紧张。边上有路人问:做导学能不能防辐射?就说:那当然能了~ 然后从边上的箱子里掏一本导学给他做。<br>(2011.03.18)</p><hr><p>语文老师:“你们才看过几个当官的就觉得他们都是贪官……”我:“一锅老鼠屎……”(我想说什么你们懂的)射长:“你一不小心说出了真话。”<br>(2011.07.01)</p><hr><p>其实只要选物理的就够了…(1kg铀就可以把苏州夷为平地了…)<br>@陈凯:<br>学化学的是核心!就要期末考试了,我们把学校炸了吧。选地理的负责测量爆破点;理科的计算炸药吨数;选化学的负责配制;体育特长生搬运;选物理的爆破;美术特长生拍摄;音乐特长生配音;文科的撰稿;英语好的翻译;选政治的安抚校领导;选历史的善后;选生物的全程保护环境。就这么定了。<br>(2011.07.03)</p><hr><p>看了一圈点名,果然只有我一个人在“房间里最重要的”这个问题上回答的是“我”……<br>(2011.07.08)</p><hr><p>@马晓栋:<br>迎新现场章泽天走过,电子、自动化学长们的眼神先是→_→ 然后是 ←_ ←,软院学长们望着背影眼神都是←_ ←,生医、人文的学长学姐都表示=_=!!….<br>(2011.08.17)</p><hr><p>@王帅国:<br>某大学教务处。老师低头看了看手中的“手动选课申请”,说,“这位同学,请不要搞什么特殊,全校所有人都在平等地抽签选课,凭什么给你一个人开后门!”“老师,我今年选的所有课选课人数都好几万,一门也没有抽中,我都快没课上了,呜呜呜…”“那好吧,小李,你过来帮这位章同学手动选一下课。”<br>(2011.08.22)</p><hr><p>@顾容容:<br>刚在看凤凰卫视直播的奥巴马在肯尼迪中心的911纪念演讲,听到他说:“我们克服了内战,克服了奴隶制,克服了共产主义。。。”我明明清楚地听到了这个单词communism,但是同声传译愣了一下,犹豫几秒钟,估计是在天人交战吧,然后硬是翻译成了“克服了奴隶制,还克服了其他东西”<br>(2011.09.12)</p><hr><p>听力考试:<br>——Hu is the chairman of the P.R.C?<br>——Who cares!<br>Quesetion:Who is the chairman of the P.R.C?<br>(2011.12.12)</p><hr><blockquote class=\"blockquote-center\">冷知识</blockquote><hr><p>苯酚考完试就不见了~ 谁拿去泡在60度的热水里溶掉了请明天连杯子一起交给孙奶奶~<br>(2011.01.18)</p><hr><p>@MakeIt:<br>我们见到的太阳是8分钟之前的太阳,见到的月亮是1.3秒之前的月亮,见到一英里以外的建筑是5微秒之前的存在,即使你在我一米之外,我见到的也是3纳米秒以前的你。我们所眼见的都是过去。<br>(2011.02.12)</p><hr><p>@杨晓飞:<br>1,1,1,1四个数算出24,标准答案是(1+1+1+1)!。但是我们伟大的中华地区暗黑3官方论坛的版主同学给出了11<<11的神级答案。<br>(2011.03.08)</p><hr><p>经过Mathematica的计算…如果某件事对于每个人发生的概率是10亿分之一且对于每个人独立,那么有99.99%的概率这件事正在发生在这世界上的至少一个人身上。<br>(2011.03.31)</p><hr><p>万一有重复数据?<br>@张颂:<br>转:早晨一女生背着一堆书进了阅览室,结果警报响了,大妈让女生看看是哪本书把警报弄响了,那女生把书倒出来,准备一本一本的测。大妈见状急了,把书分成两份,第一份过了一下,响了。又把这一份分成两份接着测,三回就找到了,大妈用鄙视的眼神看着女生,仿佛在说O(n)和O(log2n)都分不清<br>(2011.04.09)</p><hr><p>@季粲然:<br>同志们,仔细看看。今天才是我们这些学信息的男人们的光棍节。2进制8个1伤不起啊!仔细想,2047年7月31日是20个1,我们把它定为神棍节!。<br>(2011.07.31)</p><hr>"},{"title":"This day. The day of LOVE. The day for me. The day called Saint Valentine's Day","path":"2011/02/14/memoK/","permalink":"http://sx349.github.io/2011/02/14/memoK/","content":"<p>今天是情人节,结果我深陷于搬书和社区活动中不能自拔,直到下午。</p><p>有什么好过的呢?我也曾想过。</p><p>于我,这种日子本就是可有可无的。</p><p>出于射手座的天性,我素来是讨厌被什么东西束缚的。</p><p>但爱情是一种牵绊,而非约束。</p><p>在找到牵绊之后,我不得不停下脚步,拉住你的手一起走。</p><p>======================================</p><p>再过一个半小时,我将度过这一天,Saint Valentine’s Day圣瓦伦丁日。</p><p>不去追究为什么是这一天,也无所谓它叫什么名字,因为这就是the very day。</p><p>你说:为什么今天送礼物啊,明天也一样的。</p><p>我只能在边上微笑,因为我知道你绝非抱怨。</p><p>你说:你让我怎么跟家里解释啊?</p><p>面对爱情,再高的智商也要乘以零,再快的大脑也会空白。</p><p>有人说恋爱中的人都变傻了。</p><p>那是因为我在乎你,只有你,才能在我心里占据这么重要的位置。</p><p>我会愿意把开心的事都告诉你,把不开心的事告诉你,我知道你也是。</p><p>和你在一起的时候,我觉得时间变得很快。</p><p>我觉得世界只剩下这唯一的角落,周围是五彩的、流动的光,它们环绕着我们,融着我们的声音。</p><p>===================================</p><p>下一次见面,不知道要到什么时候。</p><p>但是一定不会远的。</p><p>即使不能见面,我也希望能听见你的声音。</p><p>也许这就是为什么电话里还没有说什么,时间就已经过去了很久。</p><p>你让我选方送我们的生日礼物——锚,抑或是舵。</p><p>前者,是船停止的地方;后者,是船前进的方向。</p><p>我选择了锚,而让你决定前进的方向。</p><p>而我将在旅途中寻找一个可以停憩的地方,和你一起去感受面朝大海、春暖花开。</p><p>==================================</p><p>我没有看见春天的小草,只见到了那种优雅的绿;</p><p>我没有看见夏天的骄阳,只见到了那种炽热的红;</p><p>我没有看见秋天的寒霜,只见到了那种沁人的蓝;</p><p>我没有看见冬天的冰雪,只见到了那种纯洁的白。</p><p>有时候,我连颜色都已经分辨不清了。</p><p>那是因为,我看见了你。</p><p>有了你,世界无需其他的色彩。</p><p>==================================</p><p>如果读者坚持看到了这里,一定想问,我说的是谁。</p><p>请允许我保持这最后一点的神秘。</p><p>记于2011.2.14</p><p>The Very Day.</p><p>The Saint Valentine’s Day.</p><hr><p>(2011.02.14 @ 人人 @ 杂记K–This day. The day of LOVE. The day for me. The day called Saint Valentine’s Day)</p>"},{"title":"“太仓吧8楼门”事件伪人肉搜索","path":"2011/02/10/search/","permalink":"http://sx349.github.io/2011/02/10/search/","content":"<p>起因: 太仓吧雨时同学的悼念帖被人说是炒作! (8楼 ID: Avery、ty)</p><p>流程:</p><p>STEP1:查看了这个人的个人资料,注册日期是2011-2-5,发帖只有一个(绝对是小号,但应该不是针对这件事的)</p><p>STEP2:关注他的人全部都是僵尸粉(自己百度一下什么叫僵尸粉……)</p><p>STEP3:他关注的人只有一个,ID是﹏baron丶tt (突破口!)</p><p>STEP4:查看了﹏baron丶tt的个人资料,两个ID的生日都是1994年10月5日,不可能有这么巧的吧?</p><p>由此推断 ﹏baron丶tt很有可能就是Avery、ty的大号!</p><p>STEP5:寻找联系﹏baron丶tt的方式,个人资料当中没有……</p><p>STEP6:查看﹏baron丶tt的发帖记录,寻找破绽……</p><p>结果:<br>(1)在好几张帖中 ﹏baron丶tt和ID为丶冷太阳的一个人平时好像经常一起玩~<br>(2) ﹏baron丶tt的家住在新浏河风光带附近(咱们都是好孩子,就不找上人家门上去了……)</p><p>STEP7:翻阅过程中发现了丶冷太阳的QQ号(在有一张帖中提到)</p><p>STEP8:开始秘密行动以获取 ﹏baron丶tt的联系方式</p><p>ACTION1:注册了新的QQ号,加了 丶冷太阳 好友</p><p>ACTION2:聊天!</p><p>记录如下:</p><p>Detective AP 17:44:54<br>你好!<br>(这段时间内他表示在打DOTA……)</p><p>看不破红尘。 19:39:46<br>嗯</p><p>Detective AP 19:41:03<br>首先确认下 你就是贴吧上的 冷太阳 对吧?</p><p>看不破红尘。 19:41:14<br>厄你是0 0?</p><p>Detective AP 19:42:27<br>我跟你直说了…(其实我真想委婉点) 今天太仓吧上有一个女孩子过世的贴你看见了么?</p><p>看不破红尘。 19:42:36<br>嗯看见了、</p><p>看不破红尘。 19:42:49<br>王雨时对么</p><p>Detective AP 19:45:06<br>然后是这样的~ 8楼有个人说咱们是炒作(嗯 用“咱们”这个词是因为我跟她是隔壁班同学) 然后我们就在人肉他 然后8楼那个ID是 Avery、ty 我看了他的关注 只有一个关注 就是﹏baron丶tt 而且两个ID 生日是同一天 所以我猜应该是同一个人<br>我就是想问你要一下那个人的QQ 跟他谈谈(P.S.我们苏高中的是好学生啦~ 就是觉得他这么说有点过分了~ 而且明天我们还要去追悼会 这种时候也不想吵什么的…)</p><p>看不破红尘。 19:47:10<br>好吧,那个王雨时小学和初中是一所学校,- -那时候我还记得十佳歌手的时候我和她都去的- -唱歌是蛮好听的、真的可惜人才了、</p><p>看不破红尘。 19:48:04<br>我帮你问问看好吧、</p><p>Detective AP 19:49:03<br>好的 谢谢你 非常感谢</p><p>看不破红尘。 19:49:53<br>他现在不在线,怎么称呼你? 等他上了我问过他回复你吧、</p><p>Detective AP 19:53:35<br>这样吧 麻烦你先和他确认一下是不是(因为我只是根据那几个ID的资料推测的)</p><p>看不破红尘。 19:53:56<br>嗯行、</p><p>Detective AP 19:55:11<br>如果是的话 麻烦你代我们跟他说一下<br>我们其实是真的很伤心有一个同学离开我们 也不是要炒作什么的 而且后面也有人回帖了 她家里并不缺钱 但是人去了就回不来了 这个时候认识的人都很伤心 也不是说一定要把责任都推在司机身上 但是司机起码要负比较大的责任 我们现在先让她安静地去了 希望他不要再说这种话</p><p>看不破红尘。 19:56:01<br>嗯。。好的,我帮你先确认一下好吧。</p><p>看不破红尘。 19:56:05<br>他现在不在线、</p><p>Detective AP 19:56:41<br>麻烦你了 如果是的话 把话带到就可以了 这个QQ号我以后估计就不用了~(新注册的…)</p><p>看不破红尘。 19:57:12<br>厄,话我会帮你带到。</p><p>Detective AP 19:58:06<br>再次感谢~</p><p>看不破红尘。 19:58:13<br>客气了、</p><p>Detective AP 19:59:18<br>我先闪了</p><p>看不破红尘。 19:59:34<br>嗯好的</p><p>Result: 丶冷太阳 答应在确认 ﹏baron丶tt 是发帖人后将话带到</p><p>=======================================</p><p>尽管没有最终确认,尽管没有直接找到本人,但是我相信有些事情自有公允。</p><p>雨时,有那么多人还记得你,有那么多人还想念你,有那么多人企盼你能在那边过得更好。</p><p>明天我没办法去送你,只能今天尽力将这件事做个完结。</p><hr><p>(2011.02.10 @ 人人 @ 直播——“太仓吧8楼门”事件伪人肉搜索!)</p>"},{"title":"《流星》——以未逝者之名","path":"2011/02/08/meteorid/","permalink":"http://sx349.github.io/2011/02/08/meteorid/","content":"<p>有人爱着那流星/一瞬间的美丽<br>但请不要忘记/它也曾自由/在遥远的星际<br>我将那飞翔的过往/铭刻于心<br>划过天空的光/那是不是你</p><p>我希望你是那流星/穿越荒芜与孤寂<br>即使短暂/你的美丽也融入了天地<br>我希望你不是那流星/片刻间失去踪迹<br>来不及看遍/这世间的奇迹</p><p>流星啊/请你停下脚步<br>再停留一秒/让我有时间<br>好好/看看你</p><p>——2011.2.8</p><p>以未逝者之名<br>致以最沉痛的悲哀</p><hr><p>(2011.02.08 @ 人人 @ 插播——《流星》 以未逝者之名)</p>"},{"title":"那一坨通缉令的照片~","path":"2011/01/22/memoJ/","permalink":"http://sx349.github.io/2011/01/22/memoJ/","content":"<p>之前看到有朋友发了“输入名字就能生成通缉令”的图片,作为一个富有学习意识的人,我们要充分挖掘这个东西背后的内涵……以下是研究过程:</p><p>Step 1:输入0~20,查看它们所对应的图片,发现11和20是一样的!</p><p>Step 2:输入字母A,发现A和11、20也是一样的,有什么联系呢?</p><p>Step 3:输入字母a,没发现什么,但是发现字母b和A、11、20也都是一样的!</p><p>Step 4:朋友那张输入的名字是“Echo Peng”,我又打了“Pen gEcho”,结果是一样的,也就是说和顺序无关!</p><p>猜想:ASCII码逐位相加!</p><p>为什么呐? 因为大家来看:在ASCII码中ASC(b)=98,ASC(0)+ASC(2)=98,ASC(1)+ASC(1)=98。</p><p>但是ASC(A)=65,这是为什么呢?因为它没有那么多图片!所以要再加上一个条件:取模!</p><p>假设模为K,因为98=65(mod K),所以33=0(mod K),因此K有以下可能:1、3(难道只有这么点图么?不可能的啦~)、11(这个还是有可能的,不过貌似不止这些吧?)、33。</p><p>综上,一共有33张图片循环出现,如果你输入的全是英文(包括英文标点、空格、占位符(如果你有本事输的话……)),那么就是逐位求ASCII码,然后相加,然后mod 33。</p><p>========================================</p><p>接下来看中文!</p><p>汉字 “一”的ASC码是-11589(其实是因为第一位是1被认为是负的……),但是检验下来跟上面的公式对不上……难道是公式不对吗?</p><p>错!那是因为,这种即时处理并反馈的网页用的永远是JAVASCRIPT代码,而JAVASCRIPT对于中文不是用ASC码,而是用UNICODE码,而英文字符的UNICODE码与ASCII码是一样的!</p><p>汉字“一”的UNICODE码是19968,和字母f的UNICODE码102模33同余,两个输进去出来的都是“猥琐的摄像师”,所以那个公式没错,只不过不应该用ASCII码,而应该用UNICODE码。</p><p>=========================================</p><p>综上所述,这个网页活动用的代码基本内容就是:</p><p>计算你输的每一个字符的UNICODE码,相加,然后输出mod 33后的值。</p><p>我们只需要搞清0~32对应哪张图片,然后自己就可以查着UNICODE码确认某一个名字对应哪一张通缉令啦!</p><hr><p>(2011.01.22 @ 人人 @ 杂记J–那一坨通缉令的照片~)</p>"},{"title":"有关X^X的……","path":"2011/01/21/memoI/","permalink":"http://sx349.github.io/2011/01/21/memoI/","content":"<p>这个……其实是有关数学的……</p><p>不喜勿看~</p><p>=================================</p><p>Part 1</p><p>缘起是这样的,小坦克有一天上操问我,$y=x^x$的导数是什么?</p><p>然后我华丽丽地告诉他,先两边求自然对数,然后隐函数求导,也即<br>$$\\ln y = x \\ln x$$<br>$$\\frac{y’}{y}=1+\\ln x$$<br>所以<br>$$y’=y(1+\\ln x) = x^x(1+\\ln x)$$</p><p>=================================</p><p>Part 2</p><p>几天后,我在无聊当中玩计算器,自己定了个规则:</p><p>对于一个数<br>$$x=x_1\\times 10^n+x_2\\times 10^{n-1}+x_3\\times 10^{n-2}+\\ldots+x_n\\times 10 +x_{n+1}$$</p><p>令<br>$$F(x)=x_1^{x_1}+x_2^{x_2}+x_3^{x_3}+\\ldots+x_{n+1}^{x_{n+1}}$$<br>($0^0$认为是$0$~)</p><p>这样不断地求$F(x)$竟然做了许多步都没有循环……(不排除循环了我没注意……)</p><p>我开始考虑有没有可能是不循环的,但显然不是不循环的:</p><p>(伪)证明: 显然如果一个数$x$充分大,那么每当进行一次取$F(x)$操作,$x$最多等于$9^9(\\log x+1)$。当$x>10^{10}$时,$F(x)$是小于$x$的;当$x<10^{10}$时,显然$F(x)$的可能取值是有限的(求组合数为${19 \\choose 9}=92378$个),因此超过$92378$步后根据抽屉原理,必然会重复;证毕。</p><p>最有效的证据是什么呢?当然是自守或者叫不动点的情况,也就是$F(x)=x$咯~</p><p>但是,手动构造真是太复杂了! 因为这个很没有规律…….</p><p>============================================</p><p>Part 3</p><p>此时就凸显出计算机的重要性啦~</p><p>我拍了个程序(很弱智的那种,枚举500步的状况,然后懒得写HASH了,再写了个程序判是否循环),经过孜孜不倦的努力,我终于发现了以下几组<del>3组循环,3组不动点……(P.S. 我只测试到$10^7$,还有$10^{10}-10^7$种,我电脑跑起来非得一天一夜……望有识人士相助</del>)</p><p>(1) 50119–>387423616–>17694386–>405114844–>16781367–>18517643–>17650825–>17653670–>1743551–>830078–>34378002–>17601073–>1693771–>389114260–>404244650–>50809–>404200830–>16777759–>390764443–>388291510–>420978079–>793265540–>388297225–>421801628–>33601354–>50119</p><p>(2) 404247490–>388245060–>33604500–>50091–>387423615–>17650855–>17656791–>389164014–>404244902–>387421521–>17604177–>2517543–>830081–>33554460–>53472–>826955–>404250615–>53423–>3439–>387420799–>793265567–>389167168–>421891805–>420978308–>421798751–>405848178–>51158829–>420981177–>405845053–>16787130–>18470987–>422622264–>93844–>404198244–>404198734–>405022044–>3901–>387420517–>18427715–>18427689–>421845381–>33558102–>16783525–>17653697–>389164040–>404244901–>387421518–>34381389–>420975259–>775671035–>2523563–>52968–>404247490</p><p>(3) 388247166–>34471575–>1653876–>17697224–>389114496–>791665391–>775760987–>407541658–>17654178–>18474341–>17601556–>923107–>388244064–>33601887–>34424686–>16871327–>18470991–>792441995–>1163088652–>33650902–>387470328–>35201832–>16780404–>17647928–>405891708–>421801846–>33601606–>140023–>288–>33554436–>53499–>774844386–>35248969–>791668262–>405161225–>53172–>826700–>17647419–>389114745–>405024914–>387424387–>35202088–>33557592–>388253465–>33607652–>920038–>404197736–>389114771–>405845077–>18431064–>16824413–>16824417–>17647933–>389114542–>404201375–>827212–>17600772–>2517290–>388247166</p><p>(4) 3435–>3435</p><p>(5) 1–>1</p><p>(6) 438579088–>438579088</p><p>==============================================</p><p>如果你能够坚持看到这里……</p><p>那么,就没有了……</p><hr><p>(2011.01.21 @ 人人 @ 杂记I–有关X^X的……)</p>"},{"title":"地震云什么的……","path":"2011/01/13/memoH/","permalink":"http://sx349.github.io/2011/01/13/memoH/","content":"<p>最近一直在传有地震云什么的……</p><p>搜搜搜索“地震云”第一篇,2008年12月26日《苏州天空惊现奇怪云彩》(我囧~)</p><p>百度新闻搜索第一篇,2011年1月4日《回放:江苏苏州发生地震?网友实拍地震云(图)》</p><p>谷歌新闻搜索第一篇,2011年1月13日《12号上海地震常州会地震? 看中国地震危险城市排名(图)》</p><p>更加惊悚的消息:2010年5月12日《江苏地震局称该省未来10年可能有6级以上地震》</p><p>但是现在却说(搜搜新闻搜索):2011年1月13日《江苏省地震局:近期我省陆地不会发生破坏性地震》</p><p>知识普及:(维基英文版原文)</p><p>Earthquake clouds are clouds claimed to be signs of imminent earthquakes. They have been described in antiquity: In chapter 32 of his work Brihat Samhita, Indian scholar Varahamihira (505–587) discussed a number of signs warning of earthquakes, including extraordinary clouds occurring a week before the earthquake. In modern times, a few scientists claim to have observed clouds associated with a seismic event, sometimes more than 50 days in advance of the earthquake. Some have even claimed to accurately predict earthquake occurrences by observing clouds. However, these claims have very little support in the scientific community</p><p>地震云是一种被认为是即将发生的地震的预兆。在古籍中就有描述:在印度学者Varahamihira(505-587)的著作Brihat Samhita的第三十二章中,他描述了一系列预示着地震的标志,包括地震前持续一周出现的异常的云。在现代,一些科学家声称发现了与地震有关的云现象,有时这些云出现在地震之前50多天时。有些还声称通过地震云准确预测了地震的发生。但是,这些声明在科学社会没有得到多少支持。</p><p>MCEER什么都没有……(这个不是专业研究的,是美国工程技术中心)</p><p>中国地震信息网也没有(我了个去,数据大都还是07~09年的……我根本没指望有)</p><p>查阅USGS美国地震信息网……无消息 (所有的地震预测都是美国的……)</p><p>P.S. USGS中众多的专家也承认,地震是无法预测的,所以就不要抱怨天朝政府了……</p><hr><p>(2011.01.13 @ 人人 @ 杂记H–地震云神马的…)</p>"},{"title":"反乌托邦","path":"2011/01/06/memoG/","permalink":"http://sx349.github.io/2011/01/06/memoG/","content":"<p>反乌托邦(Dystopia)(又译作“反靠乌托邦”、“敌托邦”或 “废托邦”),又作anti-utopia或cacotopia、kakotopia,根据后两者的音译,又可译作坎坷邦。与乌托邦相对,指充满丑恶与不幸之地。</p><p>——摘自《度娘百科》</p><p>最近看了世界三大反乌托邦小说《1984》、《我们》、《美丽新世界》,有些东西其实和现在的现实很像。</p><p>乌托邦(Utopia)是处于理想中的国家,而反乌托邦则充斥着极权、国家暴力、社会约束等等。但真正的反乌托邦并不是充满泥泞的,相反的,一切都处于“看起来是美好的”这样的状态之中。</p><p>这样的状态其实是很可怕的:《1984》之中,每一个英社“党员”享受着卑贱民所没有的一些奢侈品,而所谓的“内党党员”又有着比外党党员更优厚的待遇,于是他们就有了一种对比,进而为了所谓的晋升而努力,他们充满了幸福的错觉。</p><p>《我们》当中,每一个在大恩主带领下的号码感受到的是和谐、平等、集体化,这同样体现在《美丽新世界》中,两者都更多的聚焦在社会约束与个人自由的矛盾之中。</p><p>的确,真正的极权主义的反乌托邦,就在于两大矛盾:</p><p>第一,是社会约束与个人自由的矛盾,也即所谓社会感和个人主义之间的矛盾。</p><p>天朝号称国家利益与个人利益相统一,政治课上我们知道,在国家利益与个人利益相冲突的情况下,应以国家利益为重。</p><p>长此以往,我们难以否认这样的可能性:个人自由为社会化让道,因为个人自由使人暂时性的脱离社会大整体(并没有脱离社会,因为个人自由还是需要社会框架的,但是脱离了整个集体),降低了社会的发展进程,所以社会要求取消个人自由,依照社会时刻表,以充分体现一个人的劳动价值。一个人的劳动简直成为了强迫症——马克思主义的共产主义社会恐怕也是如此吧? 这就是为什么西方社会长期反对社会主义(也即共产主义的前期“症状”)的原因了吧?(其实这是显著的计划经济的样板模式,只不过苏联模式是积极性不高,而在这里是生产积极性过高……)</p><p>《1984》中,不断提到所谓的群众俱乐部,这是毫无意义的一帮人的聚会,谈论的也是老大哥的英明之类的。</p><p>《我们》中,每天必有一小时的时间,四人一组去轧马路,强制性地社会活动,而个人自由时间的两小时则被作者视为完全不必要的。</p><p>《美丽新世界》中,每周都会有一个所谓十二人合一的小组仪式,充分取消个人自由。</p><p>由此进而引发了第二大矛盾:幸福与自由的矛盾。</p><p>我们总可以看到这样的论述:“幸福和自由不可共存”、“自由无幸福”。</p><p>这是掌权者为了控制而引发的言论。我们不否认过度自由下形成的混乱,但是取消自由、取消幻想、取消个人时间,难道就是幸福了吗?</p><p>《1984》和《我们》中,掌权者解释了答案:为了控制、为了稳定、为了党的利益(最终是为了自己对权力的把握)。</p><p>(《美丽新世界》中却有了一个新的解释:我们不需要“自由”,他们不需要“自由”,他们甚至不了解“自由”。)</p><p>我想到了最近报纸上正在狂批我党一贯的“稳定压倒一切”的方针政策。</p><p>一切为了稳定。</p><p>《1984》中的英社刻意向境内开炮制造战争假相。(类似的事情没有过吗?)</p><p>《我们》中的大恩主用绿墙围起了整座城市。(类似的事情现在不存在吗?)</p><p>《美丽新世界》中用唆麻(类似毒品、致幻剂)来麻痹人们。 (类似的东西现在看不见吗?)</p><p>利用阶级不同来规范“幸福”的等级秩序,利用阶级矛盾来淡化对“自由”的追求,这就是反乌托邦的真谛。</p><p>遍观四周,我们又身处何方?</p><p>共产主义是乌托邦还是反乌托邦?</p><p>我们需要学会怀疑,需要学会分析,需要学会否定,需要学会……</p><hr><p>(2011.01.06 @ 人人 @ 杂记G–反乌托邦)</p>"},{"title":"2010之后,不是2011","path":"2011/01/01/summary2010/","permalink":"http://sx349.github.io/2011/01/01/summary2010/","content":"<p>2010之后,是长大。</p><p>我们已经或即将步入17岁,于是我们离18岁近了一步。</p><p>我们走着一条不能倒退,没有后路的成长之路。于是我们在这条路上只能期待前面是更美好的风景,却再也没有办法找到什么挽回过去的途径。</p><p>我们拥有的青春少了一年,但是我们拥有的经验与教训多了一年。我们失去了什么,我们将得到一些别的什么。我们用十六岁的毕业证换取了十七岁的录取通知书,但是我们的十六年,仍然属于我们。</p><p>这,就是长大。</p><p>2010之后,是变化。</p><p>我们的今天是有后效性的,当你在这一秒钟选择打开这片日志而不是另一篇时,你的人生轨迹已经改变了。</p><p>新年的钟声过去,那一秒之后世界将会怎样?我们将在何方?我们将会和谁在一起,说着什么,共同的笑着抑或疯着;或者也许我们呆在某一个阳光与暴风周旋的角落,品尝着寂寞与孤单,让自己的心放下一些,用眼泪润湿那些已经伤痕累累的角落,感叹别人的温暖与自己的寒冷,或是别人的寒冷与自己的温暖——乐观与悲观者如是说。</p><p>我们曾经是90后,我们现在是90后,我们将会是90后,我们不是90后,我们不仅仅是90后……</p><p>我们是自己,我们是独一无二的,我们随着自己的变化而变化,随着世界的变化而变化,随着宇宙的变化而变化,随着一切的变化与不变化而变化。</p><p>变化之后,就是新我。人不能两次踏进同一条河流,一条河流中从未有两个同样的人踏入。</p><p>2010之后,是抉择。</p><p>走出这片生养你的土地,抑或是留在这里。</p><p>选择阳关道,抑或是独木桥,也许你还可以飞过去。</p><p>放手,还是坚守,还是准备后路?</p><p>岔路之后,还是岔路,不断重复,永无尽头。</p><p>我们每时每刻,每分每秒都在面临抉择,于是我们每时每刻、每分每秒都在做出抉择。</p><p>选择去选择,还是放弃去放弃?</p><p>这不是一个命题,这是一个真理:你必须选择去选择,因为你的人生,就在选择中前进。</p><p>2010之后,不是2011。</p><p>“旧我”祝愿每一个阅读这篇日志的“新你”:</p><p>无惧长大,适应变化,直面抉择。</p><p>人,只能由自己掌控。</p><p>走自己的路,即使无路可走。</p><p>因为,路是人走出来的。</p><hr><p>(2011.01.01 @ 人人 @ 插播–非跨年非直播–2010之后,不是2011)</p>"},{"title":"我的权力观","path":"2010/12/31/memoF/","permalink":"http://sx349.github.io/2010/12/31/memoF/","content":"<p>这两天在思考,为什么我们会反感社会上的一些现象?比如官员贪污,比如医生吃回扣,比如官二代被安排工作,诸如此类。</p><p>究其原因,我得出的结论是:权力的无条件、无代价的使用。</p><p>所谓无条件,其一是指本身这个权力的实行本身是无条件的:比如说医生开药方,用药的权力是无条件的。因为信息的不对等,知识水平的不对等,病人没有确定用药到底用哪种的知识,而一般而言,医生却恰好相反,他们拥有用药的知识,如何用药他们应当比病患更懂,因此他就有了开药方的权力。</p><p>其二则是指这个权力的实行的条件本身很多,但却经常被忽略了。比如说众多被跨省追捕的案例,当地的各种官员都可以随意的“派遣”警察跨省追捕,这种权力使用的条件应该是很严格的,结果却在一张张某官员的信函中,条件自动消失了。</p><p>所谓无代价,表面上就是说金钱、地位、名誉上的损失,比如被抓或者被降职、被罚款(罚款其实是很轻的了……能用钱解决的问题都不是问题!!!!)等等。医院曝光医生吃回扣,这些医生被解聘了吗?那些违法批地、违法抓人的官员,他们被抓了吗?如果没有,那么他们这些行为就是无代价的(也有轻代价的,但是这种代价往往和无代价没什么区别,官员同级调动连代价都不算)。</p><p>而更深层次的,就是这种权力的使用是无记录(或伪记录)的。那些发动跨省追捕的人,没有人曝光的话,公众会知道吗?那些医生吃回扣,公众会知道吗?有些地方被盗,监控摄像头总是“恰好”坏了;监狱里面死人了,总是“恰好”没有监控摄像。</p><p>有人相信吗?有人能不相信吗?</p><p>没有记录(甚至有时候有记录而记录却被篡改)使得权力的使用将不会有延续性地追究 (用我们搞计算机的话来说就是“无后效性”,这一决策将不会对后面的决策进行影响)。</p><p>为什么中国的历史是断断续续的? 不言而喻。</p><hr><p>(2010.12.31 @ 人人 @ 杂记F–我的XX观(四))</p>"},{"title":"我的历史观","path":"2010/12/20/memoE/","permalink":"http://sx349.github.io/2010/12/20/memoE/","content":"<p>历史观可以分为“存在历史观”和“展望历史观”(自己分的,不要去找文献了……)。</p><p>存在历史观,顾名思义,就是针对已经存在的历史的历史观。</p><p>已经存在的历史是什么样的?人类从原始社会走来,跨入奴隶社会、封建社会、资本主义社会,随后是社会主义社会,最终目标是共产主义社会(天朝属于跳跃式发展)。</p><p>我们在这里不讨论氏族公社是什么,不讨论封建社会的历史再现性,不讨论资本主义社会的大萧条和经济危机,更不讨论具有天朝特色的社会主义。</p><p>我们只看一个问题:存在的历史是什么?</p><p>首先,存在的历史是成功者和先进者记录失败者和停止者的工具。古人云:“成王败寇。”又有人说:“历史是胜利者的历史。”因此,不可否认的,历史总是带有其主观性与误差性。</p><p>中国最严密的历史莫过于史书。</p><p>以《史记》为首的纪传体,注重以个人发展为记录顺序,因此总免不去对与个人行为与语言的臆断与猜测。这其中尤以《史记》为严重,后来的还有赵匡胤之死等等众多文学性重于史学性的东西。</p><p>以《资治通鉴》为首的编年体,注重以历史发展年限为记录顺序,这一类相对的没有文学性的东西,减弱了影响,但是还有一个不可避免的就是中国皇帝的影响,比如李世民改史、朱棣改史,不一而足。</p><p>在这其中,又要提到中国始终没有真正的《元史》、没有《辽史》(均是由于语言失传等)。这些史籍往往只能从别的书中补出(尤以《汉晋春秋》一书为例,现存的只有清朝人的补辑,散记于《三国志》等的注中,原本早已散佚)。</p><p>最后还要加上我们缺乏历史的民国&天朝时期,尤其是某些年份,1959<del>1961的已经基本上公开并且纠正,1967</del>1976有很多东西都被掩埋了,至于之后的某些恐怕就更没有人知道了。</p><p>(这其实是一个正常有奇怪的逻辑,正常在于一个执政的政党确实不希望有很多人知道它的一些不光彩的东西,但是奇怪就在于既然已经认定为反革命行动,那么为什么还要遮遮掩掩呢?毕竟我们的父母一辈对这件事也是有所耳闻的。禁言会带来什么?《六国论》早就告诉我们了:“不敢言而敢怒……戍卒叫,函谷举。楚人一炬,可怜焦土!”)</p><p>由于历史的这种主观性和误差性,进而引发了历史的不完整性。看《三国志》,既要看陈寿的原文,更要看裴松之的注,大约就是出于这个道理。</p><p>展望历史观,同样可以看出,就是个人所认为将来可能形成的历史。</p><p>有外国学者认为,世界发展将成为联邦制,世界将分为十一个联邦,我不敢苟同。</p><p>社会意识和经济形态的不同,最终导致的结果必然是对立的;对立的结果,又必然是有一方衰败。</p><p>无论那一方衰败(我们要对天朝有信心!!!),在这期间已经很难再出现新的影响政治格局的力量(不像19世纪80年代那样,美国过于注重与苏联的对抗),因为对立的本身就是彻底集团化的对立(不结盟运动基本上就是扯淡了……如果真的到了社会意识形态的对立,不结盟是不可能的;其实他们已经自己和自己结盟了么~)。因此,最终一个集团会将另一个集团吞并,形成暂时化的统一局面。</p><p>有另一个可能提前这种统一局面的诞生,就是类似于2012这样的天灾。并不是说一定会天崩地裂,形成下一个亚特兰蒂斯(其实《亚特兰蒂斯水世界》这个游戏早就有了,大概2003还是04年的时候)而是极有可能以极端天气为主,比如暴雨、暴雪、超强台风,反而不一定会有地震与火山爆发(内力作用受外界影响显然比气候受影响小),进而引发的是山体滑坡、泥石流、海啸(台风引起的),大自然的威力迫使人们必须统一起来。</p><p>最不可能的一种可能就是外星人入侵。(我从没指望能和平相处,摆出和平相处架势的外星人恐怕只是想奴役地球人)(不过我们有伟大的锑星人,他们将用导学拯救世界~)</p><p>无论哪一种情况,统一的人类帝国必将成立,其形式一开始可能能够成为共和国,但是随着时间的推移,家族将成为重要的政治力量,进而引起趋向贵族制的发展,并终将发展成为帝国,类似于罗马帝国一般,大型家族将成为统治阶级(贵族),长老会或元老会将取代共和国的议会成为权力机构。</p><p>接下来呢?必然是再次的分裂。中国人早就预言了:“天下大势,分久必合,合久必分。”古代罗马帝国分裂为东西罗马帝国,东汉分裂为魏蜀吴,查理曼帝国分裂为德意志、意大利和法兰西;近代还有苏联解体分裂为俄罗斯等十九国等等。随着利益集团的增多,整个帝国分裂出来的部分也将更多,这种分裂在前期是以明面上的控制权为体现的,就像唐朝和日本幕府时代的藩镇割据;随后就会出现实质性的分裂,就像东汉与三国鼎立之间,一州一郡的太守就能把持一方;最后就是自成一家,尤其在未来帝国时期家族作为整体进行政治斗争的时候,这种现象必将更加明显。</p><p>以上,就是我的历史观。</p><hr><p>(2010.12.20 @ 人人 @ 杂记E–我的XX观(三))</p>"},{"title":"我的人生观","path":"2010/12/18/memoD/","permalink":"http://sx349.github.io/2010/12/18/memoD/","content":"<p>今天讲讲人生观~</p><p>曹操童鞋说过:“对酒当歌,人生几何?”翻译过来就是:“眼睛一闭<del>不睁</del>一辈子就过去了,嗬哦<del>大家一起来喝酒吧</del>”</p><p>那么人的一生应当做什么呢?</p><p>我认为:人的基本目标是生存,中间目标是金钱,最终目标是财富。</p><p>人的基本目标是生存呢,那么什么是生存?</p><p>人活在这个世界上,从物质上说,与一只同等大小的猴子或者是其他什么动物没有什么太大的区别。因此,单从物质角度来说,我们的目的与其他动物是相同的。</p><p>任何一种物种,它存在的价值与意义,可以从两方面来看:</p><p>从其个体意义来看,它必须活下去,不被杀掉不被吃掉,这是个体的生存。</p><p>从其整体意义来看,它必须通过繁殖来使整个种群不断延续,使整个种群不灭绝,这是整体的生存。</p><p>综上而言,人的一生最基本的目标就是个体的生存与整体的生存的结合,也就是活下去和生育后代。</p><p>人的中间目标是金钱,这是人区别于动物的一个重要不同。</p><p>如果让你说人与动物有什么区别,你一定能举出很多种。比如,人会使用工具并创造工具(并且在此过程中深刻理解工具的作用并找出规律)、人直立行走、人具有意识等等等等。</p><p>但我要说的是这样一点:人拥有独一无二的、与一般集群生活的动物所不同的“社会”的概念。</p><p>因为社会概念的存在,人类的生活不像动物那样缺少物质性的交流。</p><p>而物质性的交流,也即我们现在常见的商品交换,最终引导我们趋向一个东西:金钱。</p><p>人的最终目标是财富,财富的含义,既包括金钱——物质财富,又包括文学、艺术、科学、历史、文化、各种人生的态度、品质、人格等等——精神财富。</p><p>有些人喜欢挣钱,他拿了很多的钱。</p><p>有些人追求科学的真理,他发现了自然的规律。</p><p>有些人舍己为人,他把跳河的人救了起来,自己死了(囧)。</p><p>这些人都实现了最终的目标,也许手段不同(或许有些恶劣),但是他们不枉此生。</p><p>综上所述,几乎每个人都遵循着这样的轨迹:</p><p>首先,他希望能够在某个地方生存下去。</p><p>如果他在某个偏僻但土地肥沃的小山村或者某一片辽阔的草原,他可能就想着长大、种地(放羊)、结婚、生孩子、孩子长大……</p><p>一定层次愿望容易被满足的人往往不会继续追求下一层次的愿望,因此他这一生也许都很难想到去挣更多的钱,尤其在这样一种比较封闭的环境下。</p><p>如果他在某个城市中长大,或者从封闭的环境中突然转移到了一个十分开放的地方,那么他的生存可能就包括在这座城市中拥有一处自己的房子,拥有一个稳定的工作,获得一份可以使自己生活下去的收入。</p><p>进而的,他必然会产生获得更多金钱的追求,也即对中间目标的追求,这是一种自下而上的发展。</p><p>但与此不同的是,对于追求人生最终目标的人来说,他们往往是不同于此的。</p><p>他们从一开始就不满足于金钱(或者说是维持生存的金钱),而是追求着各种各样的财富,而金钱是他们去进一步达成愿望的手段与途径,这就是一种自上而下的发展。</p><p>(这一点其实有待商榷,因为有些追求精神财富的人可能连金钱都不需要,他们直接跨过了中间目标而奔向最高目标。)</p><p>前一种人由于对于金钱的渴望,最终往往会成为后一种人中追求金钱的那部分,但同样有可能中途堕落而成为没有经济实力的拜金主义者。</p><p>因为他们不明白,后一种人中追求金钱的意义并不是为了金钱,而是为了使自己有意义。</p><p>这一点是很明确的:在历史的石碑上,刻勒下的往往只有推动历史与社会进步的人,以及一些资本家。资本家于其贡献是不可能被人们所铭记的,但事实却恰好相反。</p><p>“为钱而挣钱”和“为了不辜负自己的才能而挣钱”,也许从一开始就是两条完全不同的路。</p><p>我的人生应当是怎样的?我希望能够成为推动人类社会进步与发展的人,而被人们所铭记。</p><p>与大家共勉,为人生找准路。</p><p>如果你有足够的经济来源,那么让自己为后人所铭记吧!</p><p>如果你没有足够的钱,那么请在有钱后参照第一条!</p><hr><p>(2010.12.18 @ 人人 @ 杂记D–我的XX观(二))</p>"},{"title":"我的宇宙观","path":"2010/12/15/memoC/","permalink":"http://sx349.github.io/2010/12/15/memoC/","content":"<p>(P.S.以下内容不要深究 看得懂就看 看不懂就闪~)</p><p>首先让我们谈谈宇宙观,作为一名党训班成员,我是坚信马克思主义唯物论滴~</p><p>BUT!!什么是宇宙?这个是马克思没有解释的。</p><p>爆炸论显然是大家都接受的,但是爆炸之后是怎么发展的呢?</p><p>平行宇宙,这是我所认为最能够解释我们所存在的一切问题的一个根本观点。</p><p>我们知道,物质是由于夸克的运动而存在的。根据现有的科学规律(例如元电荷,例如夸克间隙),我们发现这个世界从物质上而言是不连续的。</p><p>因此,我们不难提出这样的假设:时间的发展也不是连续的,其中每一个最小部分认为是1时间单位。</p><p>我们之所以感觉到时间是连续是因为间隔实在太短,正如我们认为宏观物体是连续的只是因为我们看不到分子间、原子间、夸克间的空隙。</p><p>因此,假如我们从四维空间的角度来看待整个世界,我们完全可以认为,整个世界就是一个四维的状态表格(为什么一定要在提出时间不连续假设下提出这个表格?因为表格的下标转移是不连续的)。</p><p>其中一个维度是时间,剩下三个维度表示空间,每一个维度的大小都是我们现在不可知的。</p><p>来考虑一个简化的版本,世界是一个2x2x2x2的四维体,我们用F[0..1,0..1,0..1,0..1]来表示。</p><p>F[0]表示时间为0时的状态,也就是最初的状态,在这一点上宇宙处于最初的奇点状态,不妨标记在F[0,0,0,0]这个位置有这样一个奇点,而F[0]的剩余部分为真空。</p><p>下一刻,我们用F[1]来表示时间过了1单位的状态。根据宇宙大爆炸理论,接下来应该是物质向外扩散,因此,我们可以根据某种规则(在这里就是宇宙大爆炸理论)确定F[1]的状态。</p><p>这样,我们就“发现”了一个“宇宙”。它所遵循的规则是“宇宙大爆炸”或者更准确的说“熵自发性增(表现为随机性扩散)与非自发性减(表现为规律性收缩 聚合)的统一”。它的初始状态是给定了的一个奇点。</p><p>假设我们不限定规则,不限定初始状态,那么我们将得出无穷多个宇宙:因为每一个位置上的状态是不定的,变化范围是很大的。</p><p>由此,因为我们不能否认宇宙并不是唯一的,我们就能够彻底的肯定一点:时间旅行必然是不可能的。</p><p>首先,在同一个宇宙中,我们不可能实现沿时间轴的时间旅行,因为在整个宇宙中,能量与物质折合总量必然是守恒的,这就可以解释正物质与反物质相撞湮灭产生大量能量了。因为物质的湮灭和再生必须要伴随能量的出现和消失,如果我们实现沿时间轴的旅行,我们必须以大量的能量为代价,而且还要能够使得所到达的位置减少那么多能量。</p><p>退而求其次,我们也许能够做到替换性的时间旅行,也即我们将自己(物质)与另一个地方的能量和物质进行交换。在这一情况下,我们很有可能到达的是另一个宇宙的一部分,而这个宇宙的下一步发展规则什么的其实是不定的。</p><p>简单地说:你穿越了。</p><p>以上其实还只是一个比较常规的超脱想法,更加超脱的想法是:放弃时间轴。</p><p>我们将整个世界的每个位置上的状态作为一个下标,在我们现有的知识基础上,不妨认为这个下标在这个位置是真空为0,是九种夸克则分别为1~9。</p><p>那么,我们的整个世界就可以用一个极长无比的数字串组成,根据这个数字串我们就可以还原出一个世界。</p><p>鉴于我一开始所说的,如果认为时间是不连续的,那么,我们可以认为在t1时刻,我们正处在某一个宇宙P,这个宇宙P的状态是一个很长的串。</p><p>那么在t1+1时刻,我们将在满足某一规则的条件下,从这一个宇宙P,转到下一个宇宙Q。</p><p>我们现在的规则是什么?是物质的概率波成立的世界,是“薛定谔的猫”成立的世界,是确定性与随机性并存的世界。</p><p>因此,我们在t1时刻可能的转移方向出现了无穷多种,其中只有一种指向了我们所经历的t1+1时刻。</p><p>你无法知道其他的世界,因为你不是其他世界中的你。也就是说,在这么多种宇宙状态中,有无数个你在做着不同的事,并引发下一个不同的变化。</p><p>举一个很简单的例子,在我输入这句话的时候,我的下一步操作可以是打一个空格,也可以是打一个回车。如果我打了一个空格,那么我们可以认为,在规则作用下,我的身体中某一个部位的某个夸克发生了一个可行的移动或变化,并进而引发其他夸克的可行反应,从而打了空格。</p><p>如果我身体中同一个部位的同一个夸克发生了另一个可行的移动,那么进而引发了其他夸克的可行反应从而打了回车。</p><p>因此,宇宙的发展可以认为是跳跃的,即从一个状态跳跃到另一个状态。</p><p>将这些状态拿出来,放在一条线上,就是我们所认为的“宇宙历史 ”了。</p><p>据这么一个例子,把一副扑克牌中所有的牌认为就是所有的宇宙状态。</p><p>现在开始发牌,你依次拿到了方块5、红桃4、黑桃6、梅花6、黑桃K ……</p><p>从你个人而言,你经历了形如“方块5、红桃4、黑桃6、梅花6、黑桃K……”的宇宙历史(当然,这里的宇宙状态的转移是没有规则的)。</p><p>但对于这些宇宙状态(扑克牌)而言,如果我们把他摊开来,按照花色和大小排好,那么你拿牌的过程就可以认为是依次地从方块5跳到了红桃4 再从红桃4 跳到黑桃6。</p><p>你的宇宙历史是跳跃的一个过程,而出于习惯性的考虑,我们把这些牌拿出来当成了一条线罢了。</p><p>这个想法可以很好的解释祖母悖论等等有关时间旅行的其他问题:</p><p>祖母悖论是说,如果你回到过去杀了你的祖母,那么你是否能够存在?</p><p>其实很简单,在tn时刻的你回到了你的“宇宙历史”的tm时刻杀掉了你的祖母这个行为:</p><p>第一,在你回到tm时刻时,你强行将你的状态从当前的tn跳转到一个新的状态,也即在你的“宇宙历史”上的tm时刻的基础上增添了你这一部分的物质的一种状态,造成了一个新的(tm时刻的)状态,这已经不是你的“宇宙历史”上原本的那个tm时刻的状态了。</p><p>第二,因为你杀祖母是在新状态下的一个可行的反应,因此你从tm时刻开始开始了新的宇宙历史,直到tn时刻。</p><p>在从tm到tn的过程中,你将作为一个存在于tm时刻的物质参与整个世界的运行。</p><p>在这个观点上,时间旅行的可能性其实与之前的一种想法是一致的,也即必然要以能量换质量。</p><p>好吧 讲了这么多还只扯了一个宇宙观。欢迎探讨,不欢迎膜拜,有想法的欢迎交流~</p><hr><p>(2010.12.15 @ 人人 @ 杂记C–我的XX观(一))</p>"},{"title":"我的理想","path":"2010/12/13/memoB/","permalink":"http://sx349.github.io/2010/12/13/memoB/","content":"<p>我的理想哈~ 其实我也不知道哈~</p><p>好啦<del>耍你们的啦</del> 我当然知道我的理想是什么。</p><p>从很小很小的时候开始,大概还是牙齿漏风的时候……</p><p>什么?你说牙齿漏风的时候岁数已经很大了?那反正就是很小的时候。</p><p>那时候听说有一个“诺贝尔奖”,所以立志一定要做科学家,然后去拿诺贝尔奖……</p><p>结果后来听说中国人拿过了,不过都是美籍华人,就决定要做第一个来自中国本土的中国人拿诺贝尔奖……</p><p>结果后来发现,在中国这么混下去怎么可能拿诺贝尔奖么……于是再改,好吧,我妥协,就做第一个中国籍获诺贝尔奖的……</p><p>结果后来发现,这也已经有人做到了。所以,我决定做第一个中国籍拿诺贝尔物理或者化学奖的……</p><p>结果(这个词每次出来都铁定没好事)我注意到杨振宁获奖时貌似还没加入美国籍,是拿奖后半年才入的,所以严格说起来,他就是第一个……</p><p>啊咧……怎么办呢……我就当一个21世纪跨入新世纪的中国诺贝尔奖获得者吧……</p><p>好吧,这个理想其实太大了,大家跟着我回归现实吧~</p><p>小学读了市实小,当时去立达的两个毕业生回来做讲座,说立达如何如何的好(以及什么作业必须得用抄的神马的)于是垂涎……当时已经在想进了立达怎么混日子了<del>然后还要争取混进苏高中</del>因为说苏高中也很好滴~</p><p>后来就保送了<del>很高兴啊,没考试就进了立达。然后快开学前我爸和我一起去找了当时的副校长胡青薇老师(教数学的那个)那时候还在原来老立达的地方,就是前AL中心的地方,就是现在的施工工地撒</del>然后说如果拿满三个国一就可以签协议了哟~于是努力去准备竞赛,结果就那一年初中竞赛取消了,囧……</p><p>结果唯一没被做掉的信息学竞赛就成了唯一选择,然后看见了已经进入苏高中的几位神牛,尤其是张徐凡,全国第五啊!!后来的宋文杰神牛也没有做到啊~所以一直想和张徐凡比肩,妹的,这个任务太艰巨了……</p><p>所以后来韦斯帕克拿奖就很爽,因为张徐凡是初二才拿奖章的,我初一就拿了。计算机比赛,初二一上来就国一了,Mission 1/3 Accomplished!(结果后来他去世界联合书院了,妹的,我不想去啊~但不去就比不过他了……)</p><p>当时就开始幻想进省队,真的连理想都不算,因为差太多了,我学的东西太浅了,跟常州那帮子五年级就开始学的没法比啊。所以就一直浑浑噩噩的,今年才算真正的冲一把,因为最后一年了,明年只有比赛没有省选了。</p><p>后来计算机又一个国一,再加化学一个国一,但是没用,也没加分,去考了个省招班也只是保底。中考算小爆发吧,之前也没考过这么高,结果没爆过某熊同学……</p><p>以后会不会去当计算机工程师呢?说不定啊~其实我是希望研究基础科学,比如基础物理学、基础化学或者数论,因为我一贯坚信最本质的东西是最重要的东西。因此暂时对于大学的目标还是清华的理科吧,然后出国去读研究生,希望能够在30岁之前有点成就吧,否则的话恐怕很难在国外立足。</p><p>如果能够去卡文迪许实验室或者费米实验室,那就更好啦~不过我不喜欢核物理呃,所以费米实验室只是次选。</p><p>由此可见,昨天有人把我比作BIG BANG THEORY里面的Sheldon是错误的(比作哪一个都是错误的,因为这几个人都是量子物理学家),因为我不一定会去选择量子物理……..</p><p>苏高中历史上出过的天才太多了,间接导致后来的学生需要更高的成就才能获得称赞啊。不过24岁的哈佛终身教授蔡天西这样的奇迹,又哪是我这样的普通人能够打破的啊……</p><p>我能做的就是把教科书上诡异的HF成共价键之类的东西变成很合理的东西,搞个电正性如何?</p><p>好了,老妈催我睡觉了。晚安各位~</p><hr><p>(2010.12.13 @ 人人 @ 杂记B–我的理想)</p>"},{"title":"我与人人","path":"2010/12/12/memoA/","permalink":"http://sx349.github.io/2010/12/12/memoA/","content":"<p>好吧,其实我是很纠结的在写这个玩意儿的。</p><p>我注册人人是很久以前,大概计算机集训那会儿,去年11月的样子,就和张若愚一起在232小机房里敲着代码,然后看那个疯狂迷恋街舞的家伙在那里偷菜。那会儿已经不叫校内了吧?被忽悠着注册了一个。</p><p>那时候我还在用QQ空间,觉得人人这玩意儿有什么用呢?反正QQ上都是熟人,要找个人还不方便?那会儿已经有人(章超)整天在用饭否了,由此可见我是一个颇怀旧的人。</p><p>我刚上人人的那次,就加了个农场,然后就没有任何动力了,然后就荒废了很久,然后……就没有然后了。</p><p>有时候会来看一眼,撞见点奸情啊什么的,但是还是在空间上坚守,直到发现 空间上已经没什么人了……QQ上都是隐身,不隐身的都不鸟我(其实是因为我不鸟他们)。</p><p>正式回归是今年生日,我露了个头大家都出来了,还是这里树林子热闹啊……然后我就开始和人人勾搭上了。</p><p>我的状态总是几天一更(抑或是一天几更),日志也是不定量的,上次发的那个BLOG转载是一时兴起,后来索性就停了,最早的那篇还是从QQ空间转过来的,由此再次确定我是一个破怀旧的人。</p><p>总的来说,我本性是愿意充当一个旁观者而非参与者,因此我之前很淡定地看着尔等蛋疼……我知道QQ好友之所以默无声息的缘由了,沉默呵沉默呵,不在沉默中爆发,就在沉默中灭亡……(拽文……)</p><p>射手座的天性么,这不能怪我的~随遇而安而追求自由的我,有时候就会发现 我不过是被命运玩弄于股掌之间罢了。</p><p>在百度贴吧看了一篇《废柴道士的爆笑生活》,最经典的一句(或者说大概是这样一个意思)有一个人他找神仙改了自己的命,却最终逃不出生死轮回 阎王给出的解释是:</p><p>“其实改命也是他的命运中注定好的”</p><p>我笑<del>其实认为“命运已经注定”和“努力去追改变命运”有什么矛盾呢?轩辕剑还有三类结局捏</del>(好像是4代还是5代吧)</p><p>命运变化在选择的一刻,我选择了人人,又何尝不是人人选择了我?</p><p>庄周晓梦迷蝴蝶,蝴蝶平日是庄周。</p><p>我与人人,还是人人与我;我是谁,或者说,谁是我?</p><p>扯远了<del>回来</del></p><p>支持着我坚持上了这么久人人的还有一个,就是让我爱之而又恨之的——小游戏…</p><p>那个快乐探宝一点都不快乐…….. 我总是纠结于五行石不够用,然后发现,其实是因为我不是RMB玩家。</p><p>然后是那个宝开(POPCAP)的宝石迷阵啥啥的,大家可以去试试撒<del>还是不错的,尤其是如果你们看见我300K+的高分,我会感到很有荣耀感的(哇哈哈</del>)</p><p>人人上已经有了250+的好友,而且肯定会越来越多的。</p><p>我的QQ上只有100来个人,而且所有人都是认识的,当初我妈一直要问我这个是谁那个是谁,所以搞的我很郁闷。</p><p>现在人人就好多了,而且人人最大的好处就是:无论用什么搜索引擎都搜索不到这里——果然是符合射手天性的地方。</p><p>射手的天性是什么?第一,自由;第二,自恋;第三,有一点点幸运(这个系初中英语书上的~)</p><p>第二条来源于那个希腊神话中对着水边顾影自怜的美少年,他化为水仙花,而水仙花又是射手的守护花。</p><p>靠!我哪里自恋了!我只符合第一、第三条,偶尔、极少情况下符合第二条。</p><p>再次言归正传,现在已经有很多好友是我不认识的了啊,比如说高三的,比如说高一的,比如说朋友的朋友。</p><p>不过认识人总是好的,撞破奸情也是很让人鸡冻的~所以,这里还是我的老巢。</p><p>欢迎参观,欢迎破坏,欢迎离开。</p><p>期待我的杂记B吧!</p><hr><p>(2010.12.12 @ 人人 @ 杂记A–我与人人)</p>"},{"title":"音乐课作业","path":"2010/11/30/lyrics/","permalink":"http://sx349.github.io/2010/11/30/lyrics/","content":"<p>为了表示对李安邦同学的藐视 特贴上我的音乐课作业(歌词)</p><p>P.S. 第一首还没谱曲 有兴趣的UP<br>第二首谱了 不过欢迎交流</p><p>////////////\\诡异的分割线///////////\\</p><p>《无题》(还没想好题目啦啦啦~)</p><p>Part1<br>一夜梦阑珊 蓦然回眸看 人已黯<br>独迎寒风倚栏杆 雾尽云散<br>=======Repeat=======<br>布衾寒 为你披上锦衣衫<br>谈绵缠 静谧中月光泛<br>孤雁沉吟晚 琴弦与谁共弹 惟愿一直伴你身畔<br>=======Rep End======</p><p>Part2<br>旧地昨重还 浮华今已散 春光懒<br>双飞燕子景未染 满目纷繁<br>=======Repeat=======<br>=======Rep End======</p><p>Part3<br>回环何盘盘 歧路长漫漫 欲穷难<br>浮光掠影轻摇扇 心奚以安</p><p>=======Repeat=======<br>=======Rep End======</p><hr><p>《海》(P.S.纯属虚构 如有雷同 纯属巧合!!! 再次P.S. 请勿对号入座!!!!!!!!)</p><p>那湛蓝色烟霭 泛起一片苍白 蒙着金色的无垠大海<br>我从沙滩走来 把那心绪打开 泡沫带着那波浪上来<br>看不见的未来 是那朦胧的爱 忘不却初逢的天台<br>只是你已不再 心也随你离开 我知道人生充满意外<br>心外 天外 你是否还在<br>梦海 人海 我依然等待<br>海把痕迹掩埋 仅使留下尘埃 抹平过去相互的伤害<br>也许再难忘怀 但我不愿责怪 只留下我对着大海感慨……</p><hr>"},{"title":"有关 \"终身铭记这道题!\" 这篇文章...","path":"2010/08/14/wordcount/","permalink":"http://sx349.github.io/2010/08/14/wordcount/","content":"<p>看完 “终身铭记这道题!” 这篇文章(原文转载见本文最后)很惊讶,也太巧了点吧……于是做了个程序 把大学四级词汇和GRE的一些词汇扔进去算了一下,得到了以下单词,都是100……(很多)</p><p>其实,每个人都有不同的生活,有不同的东西使得自己的生活是100%自己所想要的。即使有些是不太高尚的东西,有些是平常的东西,有些是自己所从未在意的东西,但是就是它们,构筑了每个人百分之百的人生。</p><p>在下面这些东西当中,总有一些充斥着你的生活……</p><p>accumulate<br>vt.堆积,积累,积聚<br>vi.累积,聚积</p><p>acknowledge<br>vt.承认(…的权威);告知收到;报偿</p><p>analysis<br>n.分析,分析报告</p><p>attitude<br>n.态度,看法;姿态,姿势</p><p>awfully<br>ad.令人畏惧地</p><p>boundary<br>n.分界线,边界,界限</p><p>carpenter<br>n.木工,木匠</p><p>companion<br>n.同伴,伴侣</p><p>congress<br>n.代表大会;国会,议会</p><p>corridor<br>n.走廊,通道</p><p>culture<br>n.文化,文明;教养;培养;培养菌</p><p>delivery<br>n.投递;邮件,发送的货物;分娩;讲话方式</p><p>discipline<br>n.纪律;训练;惩罚;学科 vt.训练;惩罚</p><p>elsewhere<br>ad.在别处,到别处</p><p>excellent<br>a.优秀的,卓越的,杰出的</p><p>fountain<br>n.喷泉</p><p>hospital<br>n.医院<br>n.修理店,<英>收容所</p><p>inflation<br>n.通货膨胀;(充气而引起的)膨胀</p><p>interfere<br>vi.妨碍,冲突,干涉</p><p>lightning<br>n.闪电<br>adj.闪电般的,快速的<br>vi.打闪</p><p>ornament<br>n.装饰,装饰物<br>vt.装饰</p><p>outset<br>n.开始,开端</p><p>personal<br>adj.个人的,私人的<br>n.(报刊)私人专栏</p><p>prevent<br>v.预防,防止</p><p>primary<br>adj.主要的,初期的,根本的,初等教育的<br>n.最主要者,初选,原色,初级线圈</p><p>pursue<br>v.追捕,追求,继续从事</p><p>quarter<br>n.四分之一,一刻钟<br>v.四等分<br>adj.四分之一的<br>n.一季,25美分</p><p>restore<br>vt.恢复,修复,使复原</p><p>session<br>n.会议,会期,开庭期,学期,一段时间</p><p>socialism<br>n.社会主义</p><p>status<br>n.地位,身份,情形,状况</p><p>stress<br>n.紧张,压力<br>v.强调,着重<br>vt.强调<br>n.负载</p><p>telephone<br>n.电话<br>v.打电话</p><p>telescope<br>n.望远镜<br>v.缩短,压缩</p><p>therefore<br>adv.因此,所以</p><p>thirty<br>num.三十</p><p>towards<br>prep.向,朝,关于<br>adj.在即的,有希望的</p><p>turkey<br>n.土耳其 Turkey<br>n.火鸡,笨蛋,失败之作</p><p>useless<br>adj.无用的,无效的</p><p>variety<br>n.多样,种类,杂耍</p><p>Wednesday<br>n.星期三</p><p>whenever<br>adv.无论何时<br>conj.每当,无论何时,随时</p><p>writing<br>n.写作,著述,笔迹,作品</p><p>apropos<br>adv.适当地,恰好地,顺便地<br>adj.恰当的,适宜的<br>pref.关于</p><p>ebullient<br>adj.沸腾的,热情洋溢的</p><p>espouse<br>vt.支持,赞成,嫁娶</p><p>hirsute<br>adj.多毛的</p><p>imbroglio<br>n.纷乱,纠葛</p><p>mercurial<br>adj.善变的,活泼的,水银的,[天文]水星的<br>n.水银剂</p><p>palpitate<br>vi.悸动,心跳,发抖</p><p>plummet<br>n.铅锤,铅垂线<br>vi.垂直落下,暴跌</p><p>profuse<br>adj.很多的,丰富的,浪费的</p><p>renovate<br>vt.更新,革新,刷新</p><p>reprobate<br>n.堕落的人,恶棍,被神遗弃的人<br>adj.堕落的,邪恶的<br>v.斥责,非难</p><p>salacious<br>adj.好色的,淫荡的</p><p>subsume<br>vt.包含,包容,归类</p><p>surcharge<br>n.装载过多,额外费</p><p>temperance<br>n.(行为,饮食)节制,禁酒</p><hr><h2 id=\"原文转载\"><a href=\"#原文转载\" class=\"headerlink\" title=\"原文转载:\"></a>原文转载:</h2><p>如果令 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z 分别等于百分之<br>1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26</p><p>那么</p><p>Hard work (努力工作)<br>H+A+R+D+W+O+R+K=8+1+18+4+23+15+18+11=98%</p><p>Knowledge(知识)<br>K+N+O+W+L+E+D+G+E=11+14+15+23+12+5+4+7+5=96%</p><p>Love(爱情)<br>L+O+V+E=12+15+22+5=54%</p><p>Luck(好运)<br>L+U+C+K=12+21+3+11=47%</p><p>看出来没?这些我们通常认为重要的东西往往并不是最重要的。</p><p>那么,什么可以生活得圆满?</p><p>是Money(金钱)吗? …</p><p>不! M+O+N+E+Y=13+15+14+5+25=72%</p><p>是Leadership(领导能力)吗? …</p><p>不! L+E+A+D+E+R+S+H+I+P=12+5+1+4+5+18+19+9+16=89%</p><p>金钱,权力都不能使圆满,什么能使我们的生活变成100%的圆满呢?</p><p>每个问题都有其解决之道,只要你把目光放得远一点!</p><p>其实,真正能使我们生活圆满的东西就在我们自己身上!</p><p>ATTITUDE(心态)</p><p>A+T+T+I+T+U+D+E=1+20+20+9+20+21+4+5=100%</p><p>我们对待工作、学习的态度能够使我们的生活达到100%的圆满!</p><p>你用什么态度去看待世界,你就会得到什么样的世界。</p>"},{"title":"兔子,一路走好","path":"2009/07/23/mourn/","permalink":"http://sx349.github.io/2009/07/23/mourn/","content":"<p>兔子走的时候一定是含着笑的 因为她知道我们还想着她<br>兔子走的时候一定是流着眼泪的 因为她也始终想着我们</p><p>寻找牵绊 这是兔子的网名 也是她的追求<br>我记得那个中午 你花了整整一个中午 细细地啃完了一个小小的苹果 你的身旁 是你亲爱的猫<br>我记得那些日子 我花了整整一个中午 不 是不知多久 才相信那个噩耗</p><p>但是 你还是走了 走在了这个五百年一遇的日全食之前<br>是的 太阳也为你悲哀 用黑暗来表示沉默 那无边的黑暗中闪烁的一丝光芒 贝利珠 钻石环 也许是你洒向我们最后的光辉 让上天也为之动容了</p><p>是谁在彼岸等你/那明亮的灯已经亮起/天堂也许很美丽/但我知道这不是你的本意<br>也许你还在犹豫/想要留下最后一分的回忆/时间的钟声已经逼近/它从不管你愿不愿意<br>你的眼泪也许早已流尽/你的心却从来没有放弃/青春的年华为何要逝去/你不愿再次孤寂<br>兔子,一路走好/我们将心留在你那里/跟随着你/就像纯洁的白鸽在我们之间传递信息<br>我们将陪伴你/在心里,在梦里,在每一个瞬间里,在这个世界的每一个存在里<br>我想放下笔/重拾那可笑的怀疑/也许多一分的坚定不移/就能重新修改我们的记忆<br>来自天堂的讯息/总希望它可望而不可即/多么希望有一台时光机/让一切能够重启<br>天空也蒙上一层阴翳/太阳用沉默替代笑意/过去的已经过去/但一切都将使我们珍惜<br>不用担心,也不用怀疑/你,永远在我们的心里</p><p>兔子,一路走好……</p><hr><p>(2009.07.23 @ QQ空间 @ 《兔子,一路走好》)</p><hr>"},{"title":"箭侠","path":"2009/02/06/arrow/","permalink":"http://sx349.github.io/2009/02/06/arrow/","content":"<p>题记:弓箭是我最中意的武器,因为它没有距离的限制,再强的武林高手,如果没有防备,同样会被暗处的弓箭手瞬秒……</p><p>“箭侠”这个名号,是眼前这个着黑衣的男子在其他人眼里的名字。这个名字太有名了,因而他的真名已经被人忘却了。</p><p>“徒儿,你叫我到这里,是想干什么?”黑衣男子一面眺望着悬崖对岸的峰顶,一边漫不经心的问着身旁那个穿这粗陋布衣,却拿着一把鲜艳光丽的射阳弓的小伙子:“你再不说,我可就先走了。”</p><p>“师傅,其实我今天也没什么重要的事情。”这个名叫肖飞的小伙子走离了几步,拉开了弓,虚放了一把,接着道:“我想与你了结一个恩怨。”</p><p>箭侠转过头来,瞥了一眼,随即又转了回去:“哦?有意思。”随即又不说话了。四周陷入一片静寂。</p><p>“嗖!”一声箭响。紧接着又是一声,不过这一回是一声惨叫:“啊!”峰顶上一个人影倒了下去,人影后是一片脚步声。</p><p>“今天这个猎物够了。”箭侠不紧不慢。“现在该谈你的事了,肖飞。你的那个恩怨,看来和我有关哪。”</p><p>肖飞突然有些紧张,但他还是鼓起了勇气,道:“师傅,你还记得十五年前肖家箭第六代传人肖遥这个人么?”箭侠微微点了点头,仍没有转过头来:“这人与我对射二十五箭,最后一箭时他躲闪之时磕倒在路边的石块上。倘若没有那块石头,胜负难料啊。”</p><p>“那你知不知道,肖家箭第七代的传人是谁?”肖飞紧接着问道。“怎的?莫非……是你?”箭侠竟然笑了几声:“我早该想到了。肖遥,肖飞,怪不得当年初受你为徒时,你已经如此有天赋了。”</p><p>“既然你已知道了,那我可要替家父报仇了。”肖飞从箭囊里取了一支箭,搭在弓上,拉紧了弓。</p><p>“犹豫不决,你还是这个样子。”箭侠转过了身:“我就让你射一箭,你能射中我吗?”</p><p>肖飞抿紧了嘴唇,师傅说一不二,倘若这么两三步的距离扔不中,那以后恐怕再没有这个机会了。何况,箭侠之冷血他也见识了多次了。他瞄准了胸口,放开了弦。</p><p>他以为箭侠会转身用臂上的钢甲挡住箭,手中已经握住一把箭簇,要冲上去致其于死地。但箭却中了,正好插在了箭侠的胸口。</p><p>“能死在箭下,也不枉了我箭侠的名声。我也该收手了,以后,箭侠的名号就归你了……”箭侠依旧面不改色。</p><p>“你为什么不挡住那只箭!”肖飞冲了上去,却被箭侠推开了。“我理解你的感受。知道我为什么会与你父亲对射那么多箭,一定要置他于死地吗?你的父亲,他,是我的杀父仇人。”箭侠笑了笑,脸色却越来越苍白了:“幸好我没有一个儿子,也就不会有人来找你报仇了。我们的恩怨也就此了结了吧。”随即,他转过身,向峰顶那夕阳又看了一眼,纵身跃了下去。</p><p>“师傅!”肖飞无助的喊了出来,跪倒在地上。一抹夕阳的余晖照了下来,照出了他眼中的一滴晶莹的闪光……</p><p>后记:</p><p>这一篇文章很短,我在电脑上敲了不到半个小时。但是我自己心里很震撼。</p><p>仇恨和箭很相似。没有距离的限制,甚至没有时间的限制,可以几代、几十代那么传下去。箭侠的选择,无疑是一种最好的选择,但却是一种最悲剧的选择。</p><p>肖飞的复仇成功了,他确实为父亲报了仇。但他得到了什么?如果箭侠还有后嗣,那么他得到的,就只有下一辈的新的仇恨;而现在,箭侠让他重新看到了光明。</p><p>那一滴晶莹的闪光,无论何时何地,都值得我们去仔细的回味,感受其中的滋味……</p><hr><p>(2009.02.06 @ 百度贴吧 @ 《【二月征文】箭侠》)</p><hr>"},{"title":"鸳鸯双棍·云淡·风清","path":"2009/02/03/rods/","permalink":"http://sx349.github.io/2009/02/03/rods/","content":"<p>鸳鸯双棍·云淡·风轻</p><p>题记:鸳鸯双棍,乃是中原失传已久之兵器。双棍者,一长一短,所谓一分长,一分强,一分短,一分险,这长短双棍便是将两者优势互补,长棍护住门户,使人攻不进来,短棍便趁机冲出,夺人要害。此鸳鸯双棍虽然名传江湖,但有名者不过寥寥,若重长轻短,则犀利不足,防备有余,退可守门路,但也攻不进对手。但若重短轻长,则又过于危险,短棍直接出手,若是对抗长兵器,则进攻无门,若是对抗短兵器,又威力不足。须是长短兼备,则进可攻,退可守,能使出这般功夫的,在当今天下也难有敌手……</p><p>夜色阴沉,兵器铺后的炼炉中却依旧燃着熊熊的火焰。火光映照着一副刚毅的脸庞,棱角分明,眼神中闪耀着斩钉截铁的光芒,紧紧地盯着手中煅烧着的两条寒铁棍。这两条棍一长一短,两端各镶着一圈银边,冰冷的月光下,两道寒光向四周漫射开来。</p><p>这是铁匠刃鸣的封刀之作,却是他为自己煅的头一把兵器。他估摸着火候足了,将棍从炉中猛地一抽,旋即插到身旁的水缸中,只听倏忽一声,已经把寒铁棍抽了出来。和其他铁匠的“淬火”不同,刃鸣这一式“水火交融”是他这十几年煅铁技术的融合,单纯的“淬火”技术,兵刃的韧性虽强,却没有那般硬度。刃鸣看着手中的寒铁棍,嘴角不由得露出一丝微笑:果然是好兵器!刃鸣眼中泛出了隐隐的血丝,他将两条棍挂在兵器架上,倒头睡在了炼炉旁。</p><p>鸡鸣三遍,天空中一道白云被初升的红日映照出一条金边。刃鸣站起身来,兵器架上的寒铁棍放出闪闪亮光,映照在他眼中。</p><p>“刃鸣,你起了。”门帘被挑了开来,走进来一个女子,鬓间插着一支鎏金飞凤簪子,身着一身玄青色衣裙,走到了刃鸣身边。刃鸣回头一看,不由得嘴角抹出一道笑意:“云嫣,是你啊。正好,来看看我的这两条棍。”刃鸣取下兵器架上的寒铁棍,给云嫣细看。云嫣拿过棍,细细地看了一回,不由得赞道:“好兵器!你倒想好起什么名了么?”刃鸣笑道:“昨夜刚炼出来的,你就这么催着我?你来想一个名字吧!”云嫣细想了想,对着刃鸣摊出双手:“你拿纸笔来,我给你写着,你就照着刻在棍上吧!”不一时,刃鸣拿过纸笔,云嫣随即在纸上写下“云淡”、“风轻”两行字。“怎样?还不错吧?”云嫣转过头问刃鸣,却不见了人影。云嫣不由吃了一惊,撇下纸笔,四下里打量一番,炼炉旁确实不见刃鸣,便急急地挑起门帘,正待要喊刃鸣的名字,刃鸣却从门里走了出来,脸色颇有些苍白。“我写个字的工夫,你到哪里去了?害得我吓了一跳。”刃鸣将手中拿着的信纸揉成一团,强装出笑容道:“出去溜达一圈罢了,你又奈我何?”“你……”云嫣眼角已然看见他手中的信纸,也看出他脸色不对,待要问时,听他口气也不愿说,只得改口道:“你看看我写的这两个名字,怎样?”随即递过手里的两帖字,给刃鸣细看。刃鸣接过字,也不细看,对着云嫣莞尔一笑:“既是你写的,定是好字,何必再看呢?”云嫣假装生起气来,袖中两道分水峨嵋刺随即透了出来,向刃鸣刺了过去:“你这张嘴倒是不错,让我来练练手,把你这张嘴给割下来,看你还怎么说!”刃鸣料得她只是开个玩笑罢了,随手拿起两条寒铁棍,架住了峨嵋刺,道:“你先让我刻了这字,再来向我讨教吧!”随即走过熔炉,将手上揉成一团的信纸向火焰中一扔,到桌边仔细地刻起字来。</p><p>不出半个时辰,刃鸣放下刻刀,唤云嫣来看。只见长短棍上各刻了“云淡”、“风轻”四字,长棍上一片游云绕棍而行,短棍上则刻着两道清风交互相错,确是不可多得的好兵器。刃鸣看着手中这两条棍,眼中却流露出担心的神情。云嫣在边上看得真切,道:“刃鸣,你……怎么了?”刃鸣慢慢地站起身,摇了摇头,转身便要走,显是不愿多说一句。云嫣追上几步,拉住了他:“刃鸣,你从刚才回来脸色就有些不对,你……有什么事你就说出来啊!”“够了!你让我自己静一静!”刃鸣将手中的寒铁棍狠狠往桌上一敲,将石桌也敲下了一块。云嫣被他吓了一跳,也不敢多一句话,只得匆匆地走开了去。走过门帘时,云嫣又回头望了一眼,心里颇有些放心不下,但又不敢再开口,只得轻轻地放下门帘,留下刃鸣一人站在院中。 炼炉中,那张揉成一团的信纸早已烧成灰烬。</p><p>刃鸣适才是听到了门外有些异动,才挑起门帘到外头去看。在兵器铺的匾额上,被人用一把锋利的匕首钉住了这一张信纸。这匕首曾经的主人,正是刃鸣,这一把仿“鱼肠”剑而制得匕首,是刃鸣祖传的手艺,从不外传。如今竟有人拿他的兵器来挑衅,刃鸣心中先有些惊诧了。</p><p>刃鸣拔下匕首,取下信纸,先抬眼看了看信末,看这送信人倒是何方神圣。一看之下,刃鸣心中的惊诧不由得又增了三分。此人不是别人,正是当年街头比武时与刃鸣大战一场,最后被刃鸣用匕首将其右手砍断的小混混何势坤,刃鸣手中这把匕首,正是当年刃鸣得胜所用的一把。所谓冤家路窄,这何势坤如今已是本地父母官,“不怕官,只怕管”,如今在这个地头上,何势坤已是一手遮天,他送来的信,真是吉凶难测。</p><p>刃鸣忙翻开信纸,粗看了一遍,不由得怒从心生。何势坤仗势欺人,本已在刃鸣的意料之中,这回,他却要刃鸣在三十日内造精良生铁剑、青铜剑各三百把,倘若完不成,或是冶炼不精,却不是按着律法缴些罚金,而是“另有发落”。这两样合起来一千把剑,光铜铁就各要上千斤,一月之内,这原料上不能全,更何况精良武器,一日能造十把已然很难,现在却要一日二十把剑,明摆着要让刃鸣难堪,就是最后一句“另有发落”,也让刃鸣觉得此事不妙。因而刻好寒铁棍之时,他才会面露难色,还对云嫣发了一通脾气。</p><p>刃鸣本是个热血汉子,现在却遭小人如此侮辱,心中不由得不忿。但何势坤虽官位极低,却也是朝廷命官,若是意气用事,若怪罪下来,也定是凶多吉少。无奈,刃鸣只得速去进生铁、青铜,专心铸剑,整日也不出屋。云嫣几次登门,都被拒在了门外,她虽是心里着急,但却不晓得刃鸣为何事而烦恼,听见院中传来锻造之声,心里有些莫名的恐惧。</p><p>三十日之期迫近,刃鸣虽精疲力尽,也只造了一半之数,剩下的一半,连生铁和青铜都还未到,眼下是定然完不成了。他本想将事情告诉云嫣,这几日却又总不见她人影,只得作罢。</p><p>第三十日清晨,睡在炼炉旁的刃鸣就被砸门声惊醒。一看日头还未升起,天色尚昏,刃鸣也不知是何人,只得一骨碌爬起来,挑起门帘到兵器铺中去看。门已经被砸了开来,带头的正是本县掌刑的通判,两边排好了几个士卒,都是一副没精打采的模样。通判道:“刃鸣,命你造的六百把剑可曾造好?”刃鸣知道此人要来难他,却束手无策,只得据实禀报:“因为生铁、青铜尚未到全,所以如今还差一些,望宽限几日。”虽这么说,他心里自知,就是再宽限一个月,这六百把剑恐怕也难以交出。通判脸色一变,喝道:“依本县新律,故意拖延、虚言欺上者,皆以重刑,将他拿下!”两边士卒有些人在犯迷糊,听通判一声令下,还没缓过劲来,待明白过来要上前捉刃鸣之时,他早已从兵器架上拿起双棍,大喝一声,将众人震住:“我怎得故意拖延、虚言欺上?”通判早已吓得两腿发抖,嘴里吞吞吐吐:“这是何……何势坤他……他要我们来捉……捉你,你……你……”话没说完,一下子坐倒在地上。刃鸣道:“这狗官!你竟还敢于他助纣为虐、为虎作伥,待我先杀了你,让他认一认!”随即掏出适才的匕首,轻轻一抛,正刺在这通判胸口,登时就只有出得气了。众士卒还有些半昏睡的,看边上人都惊走了,立时醒了三分,看这架势,不敢久留,也随着别人一道往外奔走。</p><p>刃鸣杀了这通判,心里的怒气但稍放出了些,一想:边上这些士卒定是回去报告何势坤了,我不若就杀过去,拿了他狗命!此时他也不想事后如何,脑子里盘算着全是怎么对付何势坤,一脚踏出去,倒踏在那通判的身上。他仔细一看,原来这通判本已被吓了三分,见一把匕首飞来,还没刺到,就已经被吓死了。刃鸣用匕首割下通判的首级,提着人头径直向县衙冲去。</p><p>话说那些士卒先一步冲回县衙,将事情都与何势坤说了。何势坤早已料到刃鸣会杀过来,道:“不要慌了阵脚。就是刃鸣来了,我也自有办法对付。”随即吩咐,十个亲卫高手伏于衙内,听到竹签落地,便杀出来;其余人等在见刃鸣进县衙后,围住县衙,来个瓮中捉鳖。</p><p>刃鸣一路走过,远远已望见县衙旁有人影闪动,料得这是何势坤的埋伏,可笑这些士卒不谙兵法,全被刃鸣看见了。他心想:先杀了这些人,也好给自己留条后路。便提棍冲上前去,也不入县衙,先绕县衙转了一圈,手中两条寒铁棍舞得如两条银龙,将这些士卒切瓜砍菜般先杀了个精光。何势坤听到外面响起一片杀声,心里也明白了半分,但此时若冲出去,便坏了他的大事,只得按兵不动,待刃鸣杀进来,再作计议。</p><p>等不多时,只见门外一个沾满血污,圆滚滚的东西飞了进来,直打向何势坤。何势坤侧身一让,飞进来的正是那通判的人头。刃鸣走进县衙,喝道:“何势坤,当年我砍你右手,按照武林规矩,死伤由命,本与我无干。但我不愿看你如此凄惨,又将你去医馆,治你的手伤,已是仁至义尽。今日,你仍要报这一剑之仇,是何道理!”何势坤冷笑一声:“一剑之仇?说得好听!我当年自断了这右手,再也无法在人中混下去。不是我当年识得少些字,略懂一些兵法,跟着一个山大王发家,我早已暴尸荒野了。后来那山大王被我设计害死,手下的喽罗被我尽皆收编了,再花钱买了这个县官,就是为了让你尝尝这种滋味!”说罢,左手拔出一把长剑,剑面上泛着菱形花纹,虽已古旧,却仍是锋利无比,肩背上隐约可见“越王”二字,正是闻名天下的越王勾践自作用剑之一。何势坤道:“这把越王剑锋利无比,要是看了你的右手,倒是污了它了。这样吧,我先不与你动手,先来给你试试这剑如何?”</p><p>说罢,何势坤从背后掏出一把分水峨嵋刺,扔在地上,对刃鸣道:“你认得这东西么?”刃鸣心头一惊,还未说话,何势坤手一扬:“把人带上来。”侧门旁两人抬着一个青衣女子走了进来,将那女子放在何势坤身旁。“这女子长得确实有些姿色,武功也不差,这分水峨嵋刺也让我们吃了不少苦头,不过还是被我手下点穴高手给封住穴道了。她叫云嫣,不错吧?”何势坤语气平平,刃鸣却无心去顾他,道:“你把她抓来,是想要挟我?你这狗贼!”随即飞身上前,要去抢人。何势坤将身后的竹签往地上一掷,随即拖着云嫣向后连退几步,伏在帐外的几人已然冲了上来,将刃鸣围住。何势坤左手持剑,剑尖直指在云嫣的右手腕处,道:“当年你让我受了断右手之苦,我今天也要让她试试!”这十人说是高手,其实也就是会些各家武功的小混混罢了,但刃鸣在十人的围攻下身形不乱已是极限,稍有不慎,就可能被其中一人伤到,也只能眼睁睁地看着何势坤挥刀砍了下去。</p><p>眼看手起刀落,云嫣却向旁边一让,躲过了这一刀:原来自从她被捉起,她一直试着用内力冲开被点的穴道,适才何势坤要砍其右手,情急之下,穴道被强硬打通,立刻身形一闪,逃了开去。何势坤见一剑不着,立时有些慌了,云嫣乘机抢过地上的分水峨嵋刺,顶在何势坤太阳穴处,喝道:“再不停手,我就取了他的狗命!”</p><p>那十名亲卫在这一惊之下,身形已乱。刃鸣看见云嫣躲过一劫,早已安下心来,看那十名亲卫此时无心再战,索性放开了手,两棍先取了两人性命。剩下几人待缓过神来,手中武器已被打落,在双棍的进攻下,毫无招架之力。那个点穴高手虽不用武器,但此时阵势已乱,没有其他人掩护,他也没法浑水摸鱼,索性先逃了开去,向云嫣下手。云嫣看他欺得近了,峨嵋刺依旧顶住何势坤,左脚猛的一踢何势坤左手,那把越王剑脱手而出,将冲来那人一剑洞穿。剩下的几个待要捡起兵器,被刃鸣一棍一个,都逃不过一死。抬云嫣来的两人此时早已不见了踪影,早已脚底抹油,一阵风似地溜了。</p><p>何势坤此时势单力薄,手无寸铁:本想着天衣无缝的计划,却不料被云嫣和刃鸣联手,他这一招不慎,全盘皆输。此时他只得长叹一声,道:“如今要杀要剐,悉听尊便。云嫣姑娘,念在我这几日在县衙里让你好吃好喝,不曾难为你,你就替我求个情便罢。”刃鸣道:“本来要取了你这条狗命,又怕别人说我理亏才杀人灭口。如今先砍了你的左手,让你不能再祸害乡里。”抄起地上的越王剑,寒光一闪,何势坤的左手已整个掉在地上。云嫣放开了何势坤,撕下那几个亲卫的布衣,给他稍微包了下伤口,随即将他一扔,扔出了县衙外。两个提早就开溜的人,见何势坤这副惨状,赶紧抬着他奔向医馆。刃鸣此时已泄了愤,便转头去看云嫣,却见她倒在地上,手臂红紫,面色发青,口中不住地呻吟。刃鸣赶紧跪倒在她身边,问道:“云嫣,你怎么了?”云嫣提起一口气,轻声说道:“刚才冲破穴道,是靠强硬之力。如今气血乱涌,脉络混乱,几条主脉已经岔了,恐怕……恐怕……”话没说完,云嫣口角已流出一丝鲜血。刃鸣赶紧抱起云嫣,正要向医馆奔去,却听云嫣在他怀中微弱地说道:“你……不要去了。听我说……我知道……你…一直很……很喜欢我……你却不说……我要……要告诉你……我……我爱……”最后一个字上未说出口,已经两手一垂,头歪在了刃鸣胸口。</p><p>刃鸣止住了脚步,强忍住自己的泪水。云嫣手上的红紫色慢慢退去,又露出雪白莹洁的肌肤。刃鸣抚摸着云嫣此时毫无血色的脸,只有一种冰冷的感觉。</p><p>刃鸣将云嫣葬在了她曾盼望着和他一同去的地方。他在她的屋里找到了一封写给他的信,信上只有一行字:“鸳鸯双棍·云淡·风轻 爱你的云嫣”</p><p>武林江湖从此流传着一种叫鸳鸯双棍的兵器,人们只闻其名,未见其详,只能胡乱猜测,仿造品也是各式各样。有人说,真正的鸳鸯双棍已经同云嫣一起埋葬了;也有人说,铁匠刃鸣将他的鸳鸯双棍重新铸造,成了又一件绝世兵器;谁对谁错,恐怕只有鸳鸯双棍的主人自己知道……</p><p>后记:</p><p>写着写着还是有些落入俗套,最后云嫣的死却是不可避免的。本来想让何势坤杀了云嫣,不过太麻烦拉,何势坤他只有一只手了,他还要跟刃鸣干一架,这个……还是算了。</p><p>其中打的场面不少,但是写的也很简单,因为我这个人比较讨厌些打斗的场面,大家就只用兵器而已,又不是比内功,你来个降龙十八掌,我来个乾坤大挪移,这种还好点。如果要打,还要去想招式的名字,这个。。。也算了。</p><p>关于名字的问题,大家可以帮我想想那个坏蛋的名字能不能改改?我总觉得这名儿太奇怪了。不过一时半会儿也想不出两个字的特别感觉坏的名字来。这个何势坤也只是取了“地势坤,君子以厚德载物”(怎么成君子了阿<del>)的意思。恩这个名字取得很差阿</del>至于刃鸣这个名字,可以去看巴金的《家》,里面有个高觉慧,笔名叫做刃鸣,我借来用下。至于云嫣,是一本比较科幻带一点点H的书 叫做《银河英雄传》还是差不多这个意思的书 里面的主人公叫天石吧好象,然后书里面有一个大家族 姓云的 其中四女儿就叫云嫣 是很有名的魔法师(怎么又搞到魔法师了阿<del>)。OK 名字就是这么来的(我感受到很多作者对我剽窃他们的人物名称感到不忿</del>听见臭鸡蛋、烂番茄飞来的声音了)</p><p>在我被臭鸡蛋、烂番茄埋了之前再说一句:感谢吧主霍七(这是我唯一一篇坚持两天创作完的文章,因为吧主说我写得不错 hoho)还有所有已经看过或将要看的人<del>~ (感谢所有TV就不用了</del>)</p><hr><p>(2009.02.03 @ 百度贴吧 @ 《【二月征文】鸳鸯双棍·云淡·风清》)</p><hr>"},{"title":"点名答题(三)","path":"2009/01/31/questions3/","permalink":"http://sx349.github.io/2009/01/31/questions3/","content":"<p>1、是谁传給你这份问卷的: YQQ</p><p>2.你們认识多久呢: 接近三年啦~~</p><p>3.你觉得ta对你來说重要: 米什么感觉~~</p><p>4.你与ta的关系: 同班同学.</p><p>5.请问ta的兴趣是: 和Z**/G**……</p><p>6.你觉得ta的性格如何: 相当疯狂阿……</p><p>7.ta在你心目中是几分: 算了 就给个92.4935083095038930948…… 分吧!</p><p>8.睡覺前第一件事: 最后一件事吧?闭眼</p><p>9.起床前第一件事: 睁眼</p><p>10.你喜欢的季节: 冬天</p><p>11.你打工吗: 哪方面的?</p><p>12.打工次數: 不清楚俄~~</p><p>13.想去的国家: 美国、英国、中国(这个已经去过很多次了)……</p><p>14.你讨厌什麽样的个性:那种说不清楚的个性……</p><p>15.你会抽烟吗: 不会。</p><p>16.你会喝酒嗎:红酒+雪碧 或者 冬酿酒 或者 一两杯啤酒</p><p>17.你常哭嗎: 小时候哭 现在泪腺越来越不发达……快要退化了</p><p>18.你常笑嗎: 用来笑的那几块肌肉很发达 hoho</p><p>19.你喜欢去哪玩: 呆家里玩电脑不错</p><p>20.去玩时喜欢一個人去吗: 有人陪我么?</p><p>21.是假日時你都睡到几点: 8点半</p><p>22.今天的天气怎么样:没仔细看 呆家里了</p><p>23.朋友和情人你會选: 没有情人 只能选朋友……</p><p>24.机会和命运你会选择:命运吧</p><p>25.你很自恋吗: 射手座的幸运花是水仙 水仙的花语是自恋 我是射手座的……你看着办吧</p><p>26.这问卷多不多: 还好拉</p><p>27.要怎样才能让自己好过一点: 做白日梦</p><p>28.喜欢吃冰吗: 后面加个淇凌我很喜欢</p><p>29.现在幸福吗: 很幸福</p><p>30.最在乎哪几个朋友: **/**/***/***……你猜吧(虫食算)</p><p>31.房间里最重要的东西是什么: 我自己(如果我不在房间里 那么电脑最重要)</p><p>32.常夢到什麽: 梦到在学校里……做各种各样的事情(包括不可能在学校里做的)</p><p>33.男人精神出轨要不要原谅他: 这个好像不管我的事……</p><p>34.认为人生的意义是:为别人而活着</p><p>35.听什么歌一定會流泪: 边上有人喷胡椒粉的话 听每一首都会</p><p>36.如果沒有朋友你會怎麽做: 恩 去找两个。。。</p><p>OK 我就不点名了 你们自己想答得自己拷过去好了 That’s all</p><hr><p>(2009.01.31 @ QQ空间 @ 《点名》)</p><hr>"},{"title":"年终总结","path":"2009/01/03/classmates/","permalink":"http://sx349.github.io/2009/01/03/classmates/","content":"<p>2009年1月2日 窗外寒风凛冽<br>又是一年过去了<br>看见很多人在写年终总结<br>自己也不由得想写点什么。。。</p><p>林怡雯 第一个蹦出来的名字 也许也是最不能够不面对的名字 谢谢你的贺卡 礼物 以及一切<br>我不会忘了你的话的 我们仍是朋友<br>然后么 按学号来比较好</p><p>1 史雯 恩 感觉你很外向 很shiny 有那么一小点点自恋(不要紧滴 你确实也很漂亮<del>偶两个表姐都这么说)没有祝你生日快乐 现在算补上一个</del></p><p>2 叶青青 “文字女巫” 讲的一点也不错哦 还是很建议你把你的文章发表下 赚一笔外快也不错;另附:(1)《岳阳楼记》讲得很不错(2)金怡一点也不像赵文翰(3)如果我手边有衣服还想拿来套你的头<del>hoho</del></p><p>3 刘婧 赵文翰的确不错 除了花心了一点。。。其他没什么了</p><p>4 孙一方 & 5 朱能 你们两个么 我MS记得应该称呼为“公主殿下”吧?(“神秘星球有史以来最不像公主的公主”)Enjoy Life 是没错 但是每天都冲回家看动画片就有点。。。另附:希望两位学大提琴在新的一年有所成绩!!!</p><p>6 吴歆闻 这位大姐……尽管是唯一一个小学同学 但MS没什么好说的</p><p>7 张哲言 你可以考虑像史雯一样笑个不停 不过你MS还没达到她那种境界。。。</p><p>8 张晨 我还记得你小子上次给我的“班会活动详案” 我想说的是 详案最起码你给我写满1000字再说。。。</p><p>9 陆申雪 Bless you a happy life with Jackson(这名你知道不?)! (看不懂就算了,看得懂的话……就当我这句话没说)</p><p>10 陈一 可爱的绵羊。。。你下次还是不要再顺从地说“你错了”以及类似的语句了 偶开始有负罪感了。。。</p><p>11 陈依研 下次不要再跟我说“戳人是打招呼”之类的话了 否则 我不排除每天和你狂“打招呼”的</p><p>12 陈歆怡 看了团校的照片后确定……“狸猫”的称呼是不为过的。。。不止你希望09年F1 总冠军是谁?不管是谁 祝愿他能够拿总冠军</p><p>13 陈澄 今年不用去打针了?奇怪ing 其他没什么好说了</p><p>14 。。。跳过</p><p>15 罗一卿 你和史雯属于差不多类型的。。。详见史雯那条。。。另附:(1)你下次看我的时候眼神能不能不那样冷额 偶有一种被冰冻的感觉。。。Freezing (2)高宇洁 这人 满真挚的 其他 我不便多讲</p><p>16 金颖 恩 这个不说了 平时说得够多了 另附:还有毒没有 ?</p><p>17 温馨 “三把刀”现在一把刀也不带了。。。我想借把美工刀也没人借了。。今年说话声音已经越来越响了 加油!</p><p>18 董思颖 sing这个称呼真的很。。。搞笑。。。另附:你哥和你嫂子怎么样了(这句我特地不在陈一那条里说。。。否则又被拷的危险。。。)</p><p>女生到此为止。。。感觉我一个比一个说得少。。。</p><hr><p>2009年1月3日 天气一般 和心情一般惨淡<br>今天补上男生的 诸位不要介意</p><p>19 万晓迪 恩 突然失语了。。。感觉没什么好说的撒 打球好好打 你中投很准 没了</p><p>20 付玉杰 参见 ‘9 陆申雪’ 其他么 没了 大家对你的总结已经很多了 hoho</p><p>21 石煜煌 你指甲该剪了 还有 下次打百战的时候 不要跳死 一般你跳死一只 最后你在的那队都会输</p><p>22 刘承佳 团子 希望你在过春节的时候当心点 不要被人误吃了~~笑</p><p>23 吴敬文 好好学习你父母。。。说实话 你可以考得再好一点的~~</p><p>24 宋羿辰 你的雕工不错哦 下次可以考虑再雕一点(内容么 不要跟上次一样就行了) 踢球可以少踢踢了 没了</p><p>25 张尧杰 下次不要老是惦记我的手机了 可以去下棋么<del>手机我自己玩好了 不用你们帮忙撒</del></p><p>26 邹启辰 邹狗很厉害 手下的小弟无数俄~~能叫它们今年少叫叫么?</p><p>27 陈亦新 麻烦你去打听一下 你在台湾那两位同胞(团团、圆圆)过得好不好?</p><p>28 周子彦 希望你今年能够重登物理课代表宝座~</p><p>29 周尧 建议你good good study day day up 其他 米了</p><p>30 周看 来往较少 说不出啥了 。。。(又一次失语)</p><p>31 金康文 FISH 不要真像鱼一样 只有7秒钟记性。。。好好学习了得~</p><p>32 胡启清 你快回来<del>唉</del>唉~~(回音) 快点回来哦 我们都在等你哦</p><p>33 贺俊文 始终不敢相信 你竟然真的有个弟弟叫贺俊武。。。是不是金城武有个哥哥叫金城文?</p><p>34 赵文翰 不要太花心拉 韦小宝虽然花心 但起码对每个老婆都很好 你应该自愧不如才对~~照你这个样子 大概最终结果会跟段正淳差不多</p><p>35 钮正安 在线小说可以少看点啦~~ 到网上下载了看好了 手机在线看太费钱了 其他 米了</p><p>36 。。。还是说两句吧 希望自己能够稍微 thin 一点点。。。</p><p>37 顾旸旸 大肠还是很乐天俄 加油加油</p><p>OK 说完了(又是越说越少。。。)<br>总而言之 希望大家在今年能够超水平发挥 考上理想的高中~~</p><hr><p>(2009.01.02-03 @ QQ空间 @ 《1.2 偶尔路过》 & 《1.3 男生的 补上》)</p><hr>"},{"title":"抄作业风波","path":"2008/12/05/homework/","permalink":"http://sx349.github.io/2008/12/05/homework/","content":"<h1 id=\"一\"><a href=\"#一\" class=\"headerlink\" title=\"一\"></a>一</h1><p>卓的班长职位是好不容易才算安定下来的。<br>卓在班里不是个好学生,平时上课偶尔也会和老师吵嘴,成绩也忽上忽下。期中考试时,数学老师喜欢做一张折线统计图,把每次小测验的名次排出来。卓和其他同学一样,喜欢把它叫做“心电图”。那几个成绩名列前茅的人,“心电图”上总是波澜不惊的浮在最高处,好像那些没有生命迹象的人。卓的“心电图”,就比较像正常人的心跳规律,有上有下。<br>每个学期开始,班上总归会改选班委。卓是在好友影的精神支持下,才大着胆子来竞选班长,最后的结果有点出人意料,卓当选了。这倒不是因为卓的人缘比较好——相反,卓平时喜欢独来独往,要说交的朋友,也只能算上影这一个——而是因为没有人来跟他抢。这倒也不奇怪,升入初三,大家都忙着准备中考,还有谁有心思去当个班长?<br>卓走马上任以后,起初也是相安无事。自修课时,班里几个调皮捣蛋的家伙,也出奇地安静,仿佛每说一句话,中考时就会丢一分似的。<br>要不是后来发生一件事,卓的班长位子会安安稳稳的坐到学期结束。<br>有一天自修课,班主任——英语老师正好在另一个班有课。压抑了很久的几个家伙,竟然偷偷地拿了个篮球,从坐在门口的卓身边溜出去打球。没料想,篮球场正好在教学楼边,班主任不经意地一瞥,就发现了这几个鬼鬼祟祟的家伙。一下课,他们就被逮了个正着。<br>班主任马上开始了审查风暴。卓坐在门口,又是班长,显然脱不了干系。班主任本着宽大处理的原则,最终还是让卓戴罪立功。</p><hr><h1 id=\"二\"><a href=\"#二\" class=\"headerlink\" title=\"二\"></a>二</h1><p>卓总算还是保住了班长的职位,对班主任也不敢有什么怨言了。但是影却对班主任有些微词。<br>影是个叛逆的女孩。她一直嚷嚷着和她老妈势不两立,向卓抱怨某个副科老师对她横加指责。卓很喜欢听影的抱怨,因为这时的影总是一副气鼓鼓的可爱样子。<br>自从自修课事件发生后,影总是替卓感到有些不平的。卓安慰她,既然没有什么大损失,也就算了。<br>但是卓心里其实并不是这么想的,他总想找一件大事,来证明自己能够胜任这个班长职位。<br>这个机会,很快还是被卓发现了。<br>初三最大的特点之一,就是各科的作业很多,而且多的不计其数。不计其数的概念,简单的说,就是肯定会有人在家里做不完,要带到学校里做的。卓虽然成绩不好,但是作业还算认真,赶工赶到12点写完作业是常有的事,早上过来把作业一交了事。但是班里那几个倒数就不行了,基本上每天都要抄上一两门,要是前一天还有什么“特别活动”,那就基本上每本本子都是空着就赶来学校了。这样的抄作业大军每天都要到班主任来校以后才会逐渐停止工作。<br>卓没有把自己的计划告诉任何人,他要来一次突然袭击。</p><hr><h1 id=\"三\"><a href=\"#三\" class=\"headerlink\" title=\"三\"></a>三</h1><p>以前,卓的手机是一直是老爸淘汰下来的旧款式,连个拍照的功能也没有。上了初三,卓软磨硬缠,总算说服老爸买了一个新的手机,拍照的像素比自家的数码相机还高。这一天,新手机就要派上用场了。<br>卓早早地来到学校,果然,几个“抄客”已经开始工作了。卓拿出手机,打开了拍照功能,装作不经意地从“抄客”们身边经过。几圈兜下来,卓手上已经握有不少的“罪证”,不过,卓心里还是有些忐忑不安:没被人看到吧?<br>影突然从卓身后拍了他一把,吓得卓浑身一颤,赶紧回过头来。影奇怪地看了卓一眼,把手一摊:“把数学作业交出来。我不会做。”卓无可奈何地摸出数学作业本,交在了影手上。影心满意足地回了座位,卓心里不由得有些犹豫:要不要拍下来?<br>晨跑的音乐响了,班主任也踏着音乐走进教室,此时,众“抄客”们也已经完成了工作,数学本也被丢回了卓的桌子上。卓叹了口气,把作业本又丢到了组长桌上。</p><hr><h1 id=\"四\"><a href=\"#四\" class=\"headerlink\" title=\"四\"></a>四</h1><p>中午第四节课一上完,教室里瞬间就成了空壳,只有卓一个人没有去吃饭。他掏出手机,准备先把这些罪证交给班主任。<br>说实话,卓并不是喜欢打小报告的人。从小学开始,卓就一直是被打小报告的对象。不过,这次为了自己的颜面问题,卓还是毅然决然地走向了英语办公室。<br>还好,这一天中午没有人重默,也没有人要来背书,这样,就不可能有人把这件事情泄露出去。<br>班主任把照片一张张看了一遍,然后还把各个被抓到的“抄客”名字记了下来。自然,这里面没有影的名字。<br>卓从英语办公室走出来时,心里有一种莫名的快感。他没有把手机里的照片删掉,他突然有一种奇想,想把这些照片印出来,作为自己班长生涯的第一件大事纪念。<br>中午还没过去,去拿作业的英语课代表就在黑板上抄下了一排名字,并附了一句话:“以上同学请在放学前到英语办公室报到,过期后果自负。”<br>被点名的几个人此时还有些摸不着头脑,只得硬着头皮上楼了。教室里议论纷纷的人群中,只有卓自己心知肚明。</p><hr><h1 id=\"五\"><a href=\"#五\" class=\"headerlink\" title=\"五\"></a>五</h1><p>下午四节课后,被点到的人都已经到办公室喝过茶了。影凑过来问卓:“你知不知道他们怎么会被叫上去啊?”<br>卓的脸一下子有些发烫,赶忙说:“不知道啊。怎么回事?”<br>影没有发现卓脸色的变化,接着说道:“听说,被交上去的几个人今天早上抄作业被发现了。而且不仅仅是知道抄作业,还知道坐在哪儿抄,连抄什么都知道。真是天晓得怎么回事!”<br>卓支支吾吾地应了几声,继续做着作业。旁边被点到的几个“抄客”此时都是病怏怏的,回去恐怕免不了一顿批。<br>影把书包整理好,一边走出教室,一边招呼卓:“你快点啊,走了。”卓赶忙把作业胡乱地往书包里一塞,拎着书包冲了出去。<br>到了车库,卓把影拉到一个角落,说:“给你看点东西。”随即打开手机,把那几张“罪证”给影看了一遍。<br>“是你……”影的话说到一半,就被卓给掐断了:“喂喂,轻点,你要害死我啊。”影连忙压低了声音:“原来是你做的。”<br>卓看周围没有自己班上同学,对影说道:“今天早上我偷偷拍的。我可没害你啊。”影吐了吐舌头,道:“哼,看你也不会害我。幸好早上你没大义灭亲,不然今晚回家我就死定了。”卓笑道:“死不了。要死我来救你。”影撇了撇嘴:“就会耍贫嘴。谁要你来救啊?你还当你是英雄救美啊?”说完,影自己也不好意思了。<br>卓想了想,又对影说:“你说,会不会有人知道这事?”影说:“不晓得。应该没人知道吧。”卓还是有些不知所措,心里也没有底,只得走一步算一步了。影看着卓在那儿发呆,拉了他一把,说:“走啦,呆子。”卓又叹了一口气,把照片删了个精光。</p><hr><h1 id=\"六\"><a href=\"#六\" class=\"headerlink\" title=\"六\"></a>六</h1><p>第二天,众“抄客”们果然收敛了不少,等到卓到了教室,各位“抄客”已经各回各位。卓心里暗喜:这一轮偷袭果然收到了成效。<br>然而事与愿违。俗话说得好:上有政策,下有对策。卓一大早到了教室,就看见教室里几乎空空如也,连个人影也没有。后面闪进来一个人影,卓定睛一看,原来是影。<br>“喂,你怎么站在门口,吓了我一跳。”影对卓抱怨了一声,卓有些摸不着头脑:“人呢?”<br>“转移阵地拉。都到食堂去抄了。”影把手上的作业本往组长桌上一丢,转头拉着卓到了教室外面,低声说:“要不是你那天拍了照,谁想到那种地方去抄作业啊?光线又不好,伤眼睛啊~”卓哭笑不得:我也没叫你继续抄啊……<br>班主任很快也从别的老师那儿得到了风声,隔天,就在食堂门口把几个来不及撤退的“抄客”给截住了,免不了又是一通电话打回去。几个人又一次垂头丧气地回到了教室,又引起了一片议论。卓偷偷地看影,发现影也在偷偷看他,两人对了下眼神,心里各自会意:卓的这把火,果然烧大了。</p><hr><h1 id=\"七\"><a href=\"#七\" class=\"headerlink\" title=\"七\"></a>七</h1><p>这一次“反抄”行动总算是告一段落。卓感觉自己的这次行动成效还算颇丰,倒是影在边上给他泼冷水:“做这种事,你不损阴德啊你……”卓顶了回去:“说不定他们到时候中考高了个十分,还对我万分感谢呢!”影装出一副汗颜的样子:“你这个人……脸皮还真厚啊……”<br>时隔不久,众“抄客”们又开始蠢蠢欲动。这一次,众人可都是有备而来:一人抄,一人放哨。卓就算再神通广大,也不可能再进行他的“反抄行动”二号了。看着又一次向他桌子边上走来的影,卓慢悠悠地掏出数学作业本,最后下了个结论:这抄作业风波,恐怕永远没有结束之日了。</p><p>写于2008年12月5日 抄作业风波一周后</p><hr><p>(2010.02.16 @ QQ空间 @ 《抄作业风波》)</p><hr>"},{"title":"点名答题(二)","path":"2008/07/31/questions2/","permalink":"http://sx349.github.io/2008/07/31/questions2/","content":"<p>可恨的YVIY 我将继续追杀你~~hoho</p><p>A、<br>被點到名字旳轉載该文章,再點修改,要在自己旳空間里寫下自己旳答案,然後去掉1個你最不喜歡旳問題再加上1個你的問題,仍然組成19個問題,傳達其他10個人,列出其他10個需要回答問題旳人旳名字,還要通知對方.被點名者不得拒絕回答問題,完成游戲旳人將會永遠得到大家旳祝福.<br>B、<br>这10个人要在自己的Qzone里注明是从哪里接到的,并且再传给其他10个人,让游戏继续下去,不得回传.点到名字的人将会得到大家的祝福,并且所有美好的愿望都会在不久的将来实现.</p><p>1.永远的距离有多远?<br>至死不渝</p><p>2、半夜睡罘着怎麽辦?<br>睁眼看天</p><p>3、希望新學期旳生活怎樣.<br>继续加油啊</p><p>4、滿足于現在自己旳生活嗎?<br>不满足</p><p>5、用什么颜色来形容自己?<br>黑色 (黑马)</p><p>6、最大的愿望是?<br>点我名的人付点钱啊</p><p>7、最快樂旳事情昰什麽?<br>考得好的时候</p><p>8、幸福嗎?<br>是又不是</p><p>9、孤獨旳時候,怎麽給自己樂子<br>冥想 俗称发呆</p><p>10、心痛的时候怎么让自己快乐点?<br>同上</p><p>11、這個世界昰罘昰很美好?<br>也许吧</p><p>12、冇一個大假期,你怎麽過?<br>是有还是没有?</p><p>13、你覺得可以相信对你好的人嗎?<br>看情况吧</p><p>14、別人拒絕你,你會放弃嗎?<br>视情况而定 应该放弃就会放弃</p><p>15、爱睡觉吗?<br>很喜欢</p><p>16、下輩子想做什么?<br>啤酒盖子(比尔盖茨)</p><p>17、最讨厌什么?<br>点我名的那位同志<del>555</del></p><p>18、(此题我换一下 因为我不懂) 什么是人?<br>很有理性的一种动物</p><p>19、现在脑子里在想什么?<br>我到底要虐谁呢?</p><p>偶像各位同志一样拉 就不再虐人拉 hoho<br>大家心里的大石头就放一下吧</p><p>YVIY 等着我追杀吧<br>ENd</p><hr><p>(2008.07.31 @ QQ空间 @ 《7.31为什么是我~~》)</p><hr>"},{"title":"点名答题(一)","path":"2008/07/27/questions1/","permalink":"http://sx349.github.io/2008/07/27/questions1/","content":"<p>2008年7月27日 天气不管<br>被YVIY点名了<br>那就恭敬不如从命</p><p>1.对自己最想说的话?<br>“你是天才!!!”hohohoho~~</p><p>2.什么时候你会感到孤单<br>偶现在就很孤单<del>谁来陪我</del></p><p>3.如果有很多钱你怎么花?<br>这个 我想把微软包下来~~</p><p>4最近最想实现的愿望?<br>把问题3实现</p><p>5.有压力了怎么缓解?<br>打游戏</p><p>6.最受不了自己哪个缺点?<br>粗心大意~~</p><p>7.最不喜欢的事情?<br>逛街 尤其是陪老妈逛街 因为她每次都要给我买衣服 而我又很不喜欢试衣服</p><p>8.怎样应对伤心难过的事情?<br>写日志 玩游戏 发呆</p><p>9.说出点你名的人的3个优点<br>写作很好 蘑菇头很好 人不怎么好(喂喂,这也算优点的哦)</p><p>10.对点你名的人的第一印象是怎样<br>蘑菇头啊~~</p><p>11.你最希望从朋友(不包括爱人)那里得到的是什么?<br>Money</p><p>12.什么样的人会让你视为知己?<br>我没有知己 人一定要留底 就算是装也要装出一幅城府很深的样子</p><p>13.最值得回忆的事是什么?<br>“三·二七春游事变”</p><p>14.最经常去的娱乐场所是哪?<br>学校 hoho</p><p>15你最在乎的事情<br>曾经有过 现在没了 失落ing</p><p>16.未来三年有什么打算?<br>苏高中国际班 and 信息学省队</p><p>17.如果现在可以让你随心所欲去旅行,你想去哪?<br>木星 jupiter 是偶射手座的守护星</p><p>18.你希望你的伴侣(OR未来的伴侣)会不会做饭?<br>会 因为偶不太会做饭(当然 ,偶可以学)</p><p>19.你是一个会妥协的人吗?<br>不会 从我每次考试大部分都会跟老师争一分两分可以看出</p><p>20.对朋友的祝福<br>友谊天长地久</p><p>21.希望朋友怎么对你?<br>只要大家都开心就好</p><p>22.你理想的伴侣应该具备什么样的品质?<br>活泼开朗一点的(不过不要泼辣的) 然后最好是个才女(这个要求高了点)如果可能的话最好是个美女(这个要求更高了,所以基本不会考虑)</p><p>23.你最讨厌的人,你会怎样去对待?<br>我会在每一次见到他时狠狠地调侃他一顿(比如叶开元 我每次都叫他叶开花 ,再附加问他一句:你什么时候结果阿?)</p><p>24.爱情中最重要的是什么?<br>彼此的关心与爱护</p><p>25.最讨厌什么?<br>有很多碎骨头的糖醋排骨~~!!!!</p><p>26.做这些题是什么心情?<br>乱七八糟(因为题出得乱七八糟)</p><p>27.会对你喜欢的人许一辈子的诺言吗?<br>会与不会有什么区别吗!!!就算我许下几千年的诺言 最后的结果呢!!!</p><p>28.经济条件是你选择爱人的必要条件吗?<br>不是 绝对不是</p><p>29.朋友是什麽?<br>付诸感情的工具</p><p>30.你现在最害怕的事是什麽?<br>成绩下降</p><p>31.最喜欢什麽样的生活方式?<br>每一天都是放假(不上课的放假)</p><p>32.你觉得住大房子,开好车很重要麽?这算是你的目标之一吗?<br>一般 如果一生能够有些成就 比有很多钱更加有意义</p><p>33.对点你名的人想说的一句话<br>你小子 等着 我已经派人去追杀你了 有没有感受到一股隐隐约约的杀气?</p><p>34.最开心的时候会和谁一起分享?<br>希望跟别人分享 但是总觉得别人把我的快乐当成过眼云烟</p><p>35.希望自己的婚礼是中式的还是西式的?<br>西式的 中国传统婚礼太过繁琐了</p><p>36.最喜欢的浪漫事情是什么?<br>和相爱的人手拉手在一个比较romantic的地方走(期待Ing)</p><p>37.你对现在的生活有什么感觉?<br>不爽 相当不爽 但又不得不过</p><p>38.现在最期待的一件事?<br>有一台新电脑 内存1GB的 硬盘1TB的 四核处理器 二十英寸液晶屏(手提也可) 无线上网 智能感应鼠标、键盘(不知道市场上有没有这种电脑卖)</p><p>39.你最难忘的出去玩时哪一次?<br>安吉百草园 我已经不记得那次怎么玩的了 但是还是很难忘啊</p><p>40.最近一次大哭是因为什么?<br>最近只有小哭 大哭要追溯到婴儿时代</p><p>41.一生中什么最重要?<br>事业与家庭</p><p>42.被点名是什么心情?<br>空虚</p><p>43.打过人吗?<br>废话</p><p>44.希望自己多大结婚?<br>在碰到真爱的人后 无所谓年龄</p><p>45.你在过去的一年中变了吗?<br>重新做人(大家不要想歪拉)</p><p>46.你的生活态度?<br>大家都是出来混的~~</p><p>47.会不会用真心对待每一个朋友?(真心话)<br>要看是什么朋友 酒肉朋友需要用真心吗?真正的朋友就是你不用真心对待他他也会真心对待你的那种</p><p>48.你的学号是多少?<br>???50(小学)—〉36(现在) 问这个干嘛</p><p>49.此时心中想着哪个人?<br>心里很乱 想着很多人 包括比尔盖茨阿 爱因斯坦阿 爱迪生阿之类的</p><p>50.希望对被点名的人说什么话?<br>施主莫怪 贫僧这厢有礼了</p><p>然后是点名时间 好像很多人都被YVIY给点了 偶就点自己同学拉(希望她们看得见)<br>未完成<br><del>阿桃</del><br>就两位吧 偶就不虐人了</p><p>ENd</p><hr><p>(2008.07.27 @ QQ空间 @ 《7.27点名答题?》)</p><hr>"}],"categories":[],"tags":[]}