From 3cf2bb78c400aa774448d687a60ae1cfbdc8db16 Mon Sep 17 00:00:00 2001 From: Janet Date: Thu, 15 Dec 2016 20:48:15 -0500 Subject: [PATCH] feat: vox custom parser (#67) --- fixtures/www.vox.com/1481563623532.html | 1 + src/extractors/custom/index.js | 1 + src/extractors/custom/www.vox.com/index.js | 58 +++++++++++ .../custom/www.vox.com/index.test.js | 97 +++++++++++++++++++ 4 files changed, 157 insertions(+) create mode 100644 fixtures/www.vox.com/1481563623532.html create mode 100644 src/extractors/custom/www.vox.com/index.js create mode 100644 src/extractors/custom/www.vox.com/index.test.js diff --git a/fixtures/www.vox.com/1481563623532.html b/fixtures/www.vox.com/1481563623532.html new file mode 100644 index 000000000..b58354c21 --- /dev/null +++ b/fixtures/www.vox.com/1481563623532.html @@ -0,0 +1 @@ + Donald Trump’s tweets fit a pattern of harassment Twitter has banned before - Vox
clock menu more-arrow

Donald Trump’s tweets fit a pattern of harassment Twitter has banned before

The President-elect’s targets may endure real-life threats from hordes of his supporters.

Updated by Aja Romano
Photo by Steve Pope/Getty Images

After the election, Twitter issued a statement to Slate that it would “take action” on any accounts that violated its harassment and hate speech policies:

“The Twitter Rules prohibit violent threats, harassment, hateful conduct, and multiple account abuse, and we will take action on accounts violating those policies. ... The Twitter Rules apply to all accounts, including verified accounts.”

Do President-elect Donald Trump’s tweets fall into this category of actionable offenses? In a nutshell, yes.

It’s no secret that Trump’s statements on Twitter have been controversial, viewed by many as bullying, inflammatory, and divisive. But a new article from the Washington Post about the impact of a single Trump tweet on one of his targets — an 18-year-old female college student who questioned whether he was “a friend to women” — shows just how much damage, and potentially how much violence, can ensue from the president-elect’s chosen social media platform.

After student Lauren Batchelder, now 19, questioned Trump at a New Hampshire public forum in 2015, she was bombarded with hate, harassment, and threats from Trump supporters responding to Trump’s tweet calling her a “nasty,” “arrogant young woman” and implying she was a plant for Jeb Bush. (Both Bush’s former campaign spokesperson Tim Miller and Batchelder ridiculed this idea to the Washington Post.) This harassment has included everything from online sexual threats and doxxing (the public broadcasting of her real-life address and other personal details) to phone calls threatening violence. Over a year after the incident, Batchelder told the Post she still receives threats over social media.

While the initial tweet may have from Trump, the resulting harassment leveled at Batchelder didn’t; rather, it came from a handful of harassers among his millions of supporters. But that distinction shouldn’t matter to Twitter — and it hasn’t in the past.

Twitter has seen — and banned — this kind of behavior before

When Twitter decided to ban Breitbart star blogger Milo Yiannopoulos — a poster child for the internet-bred white supremacist movement known as the alt-right — it did not do so because Yiannopoulos was posting hate speech from his notorious account. Granted, he had been posting hate speech and harassment, like his infrequent jokes that Ghostbusters star Leslie Jones was a man — but Twitter’s responses to this kind of behavior from Yiannopoulos, a verified user, had always been temporary suspensions with eventual reinstatement.

But Yiannopoulos’s incitement of a spontaneous hate mob to rally against Jones after the Ghostbusters premiere pushed Twitter over the edge. Jones was already facing harassment from cranky alt-right members and Ghostbusters fans, rising anger that had grown steadily over the weekend of the film’s US opening. The following Monday evening, Yiannopoulos essentially threw a lit match on the growing trash pile of harassment by retweeting one of Jones’s tweets and accusing her of “playing the victim” and being unable to handle a little “hate mail.”

It was only one tweet — but it summoned hours upon hours of harassment for Jones. Over the course of that long Monday night, Yiannopoulos’s followers insulted, mocked, and harassed Jones with tweets numbering in the thousands. Hundreds of the male alt-right members and trolls harassing Jones took pains to include Yiannopoulos in replies to many of their tweets — almost as a kind of salute.

The incident was enough to finally make public outrage around Yiannopoulos’s behavior louder than the support of his fans. Twitter booted Yiannopoulos, who has since reemerged on the new alt-right-friendly social media platform Gab.

Gab.ai

The similarities between Yiannopoulos’s behavior and Trump’s are numerous

Even setting aside the connection between Yiannopoulos’s behavior toward Jones and the alt-right’s larger role in the political climate of 2016, there are a number of similarities between Trump and Yiannopoulos. Most notable is the way each man has used Twitter as a megaphone and a weapon.

Strikingly, neither of them generally engaged in direct harassment of their opponents. Instead, both men simply tweeted out an accusation or two and let their mass of supporters do the rest. This strategy relies on the alt-right’s independently evolved tactics of trolling and harassment, tools forged in the huge multiyear harassment campaign that was Gamergate. The social media arm of the alt-right and other right-wing supporters know exactly how to swing into action quickly to launch a sustained harassment campaign against whomever their leaders direct them toward. And their leaders know exactly what the effects of their direction will be.

Earlier this week, for example, Trump sent a horde of harassers after Carrier union leader Chuck Jones with a December 7 tweet in which he called Jones “terrible” at his job. Jones is an Indiana steel union leader who previously took issue with Trump’s claims that he had saved 1,100 Carrier jobs, calling the real number closer to 730 and accusing the president-elect of lying. Jones told the Indianapolis Star the night of Trump’s tweet that his phone had since been ringing off the hook with threats from Trump fans: “Calling me names, wanting to know if I have children ... [saying] I better watch out for myself, and they know what kind of car I drive, that I better watch out for my kids.”

Again, the pattern is clear: Trump designates an enemy of the moment on Twitter, and his followers go into attack mode. They dox the target, call and invade their real-life space, and issue threats to the target and the target’s family.

This is clearly in line with the behavior for which Twitter previously banned Yiannopoulos.

Trump is galvanizing some of Twitter’s most hate-filled users

Even if Trump isn’t personally using racialized and misogynistic speech on Twitter, a huge subset of his followers are tuned in to alt-right rhetoric and social media. In January 2016, an analysis of Trump’s Twitter found that 62 percent of the people he retweeted in a given week followed more than one white supremacist Twitter account. Trump is retweeting the viewpoints of white nationalists and white supremacists, and they, in turn, are very vocally and visibly paying attention to what he has to say.

That means that every time Trump chooses to use his Twitter platform to directly target someone he’s angry with, he’s not just putting them directly in front of his 17 million Twitter users; he’s putting them directly in front of some of Twitter’s most fervently bigoted users — ones whose culture of online harassment has created an environment in which they are more likely to dehumanize their targets and detach from the harm they cause.

Trump’s casual followers may get angry and move on, but Trump’s many alt-right followers won’t. They won’t just harass Trump’s targets; they will harass them using racist, misogynistic speech and threats of violence. In fact, they’re already doing so on a regular basis. In January, a female Republican campaign strategist was subjected to an onslaught of misogynistic harassment and online threats of sexualized violence for days after Trump dropped a negative tweet about her. Later, Trump supporters would photoshop Batchelder’s face covered in semen.

Jones, a 30-year union man, was able to shrug off the threats because he had grown used to them over time. But a 19-year-old student like Batchelder, or anyone else less hardened to years of harassment, might not have as easy a time; in fact, in a 2014 Pew study, 78 to 79 percent of survey respondents described their experiences with online harassment as varying degrees of “upsetting.” (To be fair, Batchelder seems to have fared well; she created a Twitter account on Friday, where hundreds of supporters welcomed her, congratulating her for her courage in standing up to Trump and the trolls.)

The logic that targets of harassment should just deal with it and move on not only ignores the psychological reality of such harassment; it also assumes that the onus should be on Trump’s Twitter targets to just grow a thicker skin and ignore the harassment, rather than on Trump himself to be less incendiary, negative, and accusatory on social media.

But Trump, despite initially saying he might quit Twitter altogether after assuming the presidency, has instead shown every sign of using the platform as his unofficial microphone, his direct access to the public and a kind of ongoing press soapbox.

Thus, if we accept that the onus doesn’t appear to be on Trump to discourage his followers from being less hateful — or even to restrain himself from being less antagonistic and aggressive — perhaps it might be on Twitter itself.

The reality is that Twitter is unlikely to act against Trump

Of course, this all assumes that Twitter will apply its policies fairly and evenly across the platform — which, as anyone who’s tried to report harassment to the site knows, isn’t always the case.

For example, Dutch extremist leader Geert Wilders, who was just convicted in the Netherlands of inciting hate and discrimination against Dutch Moroccan citizens, still has a Twitter platform, where he continues to issue the same strain of divisive and racist rhetoric that got him convicted in the Dutch court.

This is incendiary, divisive Islamophobic speech that saw Wilders indicted in a national court — but it’s also pretty close to the kind of rhetoric Trump used throughout his campaign circuit. And if Twitter has no incentive to ban Wilders, even with the rule of law working in its favor, it’s unlikely to ban Trump for tweets that so far have been less direct, even if there is ample evidence that they are fueling harassment.

After the election, Twitter did purge a huge number of alt-right accounts, arguably in response to criticism it received for giving the alt-right a mouthpiece on the site to begin with. But extending that action to purging the president-elect himself is a different matter. On December 6, the day before Trump attacked Chuck Jones on Twitter, Twitter CEO Jack Dorsey told Recode editor Kara Swisher at Recode’s conference that he thought Trump’s Twitter presence was “interesting”:

[H]aving the president-elect on our service — using it as a direct line of communication — allows everyone to see what’s on his mind in the moment. I think that’s interesting. I think it’s fascinating. I haven’t seen that before.

Dorsey’s attitude here indicates that his company is highly unlikely to take action against Trump for inciting onslaughts of virulent harassment and hate speech toward his Twitter targets. That doesn’t change the fact that the company has done so in similar situations in the past.

One thing that could arguably change the situation is public outrage — the kind Yiannopoulos sparked when he and his followers went after a popular comedian. But public outrage has so far done little to quell Trump, while his number of followers is only growing.

And as Trump’s first anti-Batchelder Twitter user was quick to point out, it will be harder than ever for Trump’s targets to avoid the scrutiny of Trump’s supporters — because “now we have social media.”

Update: On December 10, Twitter reinstated and verified the formerly banned account of white supremacist alt-right leader Richard Spencer. Spencer, who recently spearheaded a white nationalist convention where several attendees performed Nazi salutes in support of the President-elect, had previously been banned from Twitter in its November purge of many alt-right accounts. The move also comes after Twitter’s November rollout of advanced reporting tools for users battling harassment, which was accompanied by a blog post reaffirming the site’s commitment to its “hateful conduct” policy.

Correction: This article originally linked to a Twitter presence claiming to be Chuck Jones, created December 9. That Twitter account has since been suspended.

Was this article helpful?

Next Up In Culture

Get Vox in your inbox

By signing up you agree to our terms of use.
This Article has a component height of 44. The sidebar size is long.

The Latest

The presidential recount, explained

Updated by Tara Golshan

The Jill Stein–led effort is incredibly unlikely to change the election’s outcome.

Mitch McConnell just backed a bipartisan investigation into whether Russia helped Trump win

Updated by Jeff Stein

"The Russians are not our friends," the Senate majority leader said.

The 2017 Golden Globe nominations are here. Read the complete list.

Updated by Todd VanDerWerff and Alissa Wilkinson

The best in film and TV are honored at the January 8 ceremony

Methane levels in the atmosphere are now rising at their fastest pace in decades

Updated by Brad Plumer

It’s time to take seriously the second most important cause of global warming.

What “stand your ground” laws actually do

Updated by German Lopez

The laws have come up after controversial shootings. But they’re widely misunderstood.

Trump’s DHS pick is a retired general who warns Latin America could pose an “existential” threat to the US

Updated by Dara Lind

But he doesn’t exactly mean it the same way Trump does.

\ No newline at end of file diff --git a/src/extractors/custom/index.js b/src/extractors/custom/index.js index 658496358..7a9698c71 100644 --- a/src/extractors/custom/index.js +++ b/src/extractors/custom/index.js @@ -29,3 +29,4 @@ export * from './www.theguardian.com'; export * from './www.sbnation.com'; export * from './www.bloomberg.com'; export * from './www.bustle.com'; +export * from './www.vox.com'; diff --git a/src/extractors/custom/www.vox.com/index.js b/src/extractors/custom/www.vox.com/index.js new file mode 100644 index 000000000..7c54bf4f5 --- /dev/null +++ b/src/extractors/custom/www.vox.com/index.js @@ -0,0 +1,58 @@ +export const WwwVoxComExtractor = { + domain: 'www.vox.com', + + title: { + selectors: [ + 'h1.c-page-title', + ], + }, + + author: { + selectors: [ + ['meta[name="author"]', 'value'], + ], + }, + + date_published: { + selectors: [ + ['meta[name="article:published_time"]', 'value'], + ], + }, + + dek: { + selectors: [ + '.p-dek', + ], + }, + + lead_image_url: { + selectors: [ + ['meta[name="og:image"]', 'value'], + ], + }, + + content: { + selectors: [ + ['figure.e-image--hero', '.c-entry-content'], + '.c-entry-content', + ], + + // Is there anything in the content you selected that needs transformed + // before it's consumable content? E.g., unusual lazy loaded images + transforms: { + 'figure .e-image__image noscript': ($node) => { + const imgHtml = $node.html(); + $node.parents('.e-image__image').find('.c-dynamic-image').replaceWith(imgHtml); + }, + + 'figure .e-image__meta': 'figcaption', + }, + + // Is there anything that is in the result that shouldn't be? + // The clean selectors will remove anything that matches from + // the result + clean: [ + + ], + }, +}; diff --git a/src/extractors/custom/www.vox.com/index.test.js b/src/extractors/custom/www.vox.com/index.test.js new file mode 100644 index 000000000..0493ae95e --- /dev/null +++ b/src/extractors/custom/www.vox.com/index.test.js @@ -0,0 +1,97 @@ +import assert from 'assert'; +import fs from 'fs'; +import URL from 'url'; +import cheerio from 'cheerio'; + +import Mercury from 'mercury'; +import getExtractor from 'extractors/get-extractor'; +import { excerptContent } from 'utils/text'; + +describe('WwwVoxComExtractor', () => { + describe('initial test case', () => { + let result; + let url; + beforeAll(() => { + url = + 'http://www.vox.com/culture/2016/12/10/13898352/trump-twitter-harassment-policy-bannable'; + const html = + fs.readFileSync('./fixtures/www.vox.com/1481563623532.html'); + result = + Mercury.parse(url, html, { fallback: false }); + }); + + it('is selected properly', () => { + // This test should be passing by default. + // It sanity checks that the correct parser + // is being selected for URLs from this domain + const extractor = getExtractor(url); + assert.equal(extractor.domain, URL.parse(url).hostname); + }); + + it('returns the title', async () => { + // To pass this test, fill out the title selector + // in ./src/extractors/custom/www.vox.com/index.js. + const { title } = await result; + + // Update these values with the expected values from + // the article. + assert.equal(title, 'Donald Trump’s tweets fit a pattern of harassment Twitter has banned before'); + }); + + it('returns the author', async () => { + // To pass this test, fill out the author selector + // in ./src/extractors/custom/www.vox.com/index.js. + const { author } = await result; + + // Update these values with the expected values from + // the article. + assert.equal(author, 'Aja Romano'); + }); + + it('returns the date_published', async () => { + // To pass this test, fill out the date_published selector + // in ./src/extractors/custom/www.vox.com/index.js. + const { date_published } = await result; + + // Update these values with the expected values from + // the article. + assert.equal(date_published, '2016-12-10T15:20:01.000Z'); + }); + + it('returns the dek', async () => { + // To pass this test, fill out the dek selector + // in ./src/extractors/custom/www.vox.com/index.js. + const { dek } = await result; + + // Update these values with the expected values from + // the article. + assert.equal(dek, 'The President-elect’s targets may endure real-life threats from hordes of his supporters.'); + }); + + it('returns the lead_image_url', async () => { + // To pass this test, fill out the lead_image_url selector + // in ./src/extractors/custom/www.vox.com/index.js. + const { lead_image_url } = await result; + + // Update these values with the expected values from + // the article. + assert.equal(lead_image_url, 'https://cdn0.vox-cdn.com/thumbor/RuJTDlBH9LAp_9uFqYfnPzWXhj0=/0x175:2500x1564/1080x600/cdn0.vox-cdn.com/uploads/chorus_image/image/52223131/628656068.0.jpeg'); + }); + + it('returns the content', async () => { + // To pass this test, fill out the content selector + // in ./src/extractors/custom/www.vox.com/index.js. + // You may also want to make use of the clean and transform + // options. + const { content } = await result; + + const $ = cheerio.load(content || ''); + + const first13 = excerptContent($('*').first().text(), 13); + + // Update these values with the expected values from + // the article. + assert.equal(first13, 'Photo by Steve Pope/Getty Images After the election, Twitter issued a statement to'); + }); + }); +});