From 00bf93cd87cd624b0db70588b7095943979efc77 Mon Sep 17 00:00:00 2001 From: Barry Date: Tue, 25 Aug 2020 22:40:54 +0100 Subject: [PATCH 1/4] Update percentile wording --- src/content/en/2019/cdn.md | 50 +++++++++++++++++++------------ src/content/en/2019/javascript.md | 16 +++++----- src/content/en/2019/media.md | 22 +++++++------- 3 files changed, 50 insertions(+), 38 deletions(-) diff --git a/src/content/en/2019/cdn.md b/src/content/en/2019/cdn.md index cc0694c08f0..58bb203b1a1 100644 --- a/src/content/en/2019/cdn.md +++ b/src/content/en/2019/cdn.md @@ -486,11 +486,15 @@ In contrast, the median TLS negotiation for the majority of CDN providers is bet - p10 - p25 - p50 - p75 - p90 + Percentile + + + + 10 + 25 + 50 + 75 + 90 @@ -649,7 +653,7 @@ For resource requests (including same-domain and third-party), the TLS negotiati Distribution of TLS negotiation time for site resources broken down by CDN - +
Figure 10. Resource TLS negotiation time.
@@ -683,11 +687,15 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - p10 - p25 - p50 - p75 - p90 + Percentile + + + + 10 + 25 + 50 + 75 + 90 @@ -842,10 +850,10 @@ Most CDNs balance the need for shared certificates and performance. Most cap the
- Figure 13. Resource SAN count (p50). + Figure 13. Resource SAN count (50th percentile). - -
Figure 13. Resource SAN count (p50).
+ +
Figure 13. Resource SAN count (50th percentile).
@@ -853,11 +861,15 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - p10 - p25 - p50 - p75 - p90 + Percentile + + + + 10 + 25 + 50 + 75 + 90 diff --git a/src/content/en/2019/javascript.md b/src/content/en/2019/javascript.md index 332dd3e8102..61bbc37a100 100644 --- a/src/content/en/2019/javascript.md +++ b/src/content/en/2019/javascript.md @@ -31,7 +31,7 @@ Sending smaller JavaScript bundles to the browser is the best way to reduce down Figure 1. Distribution of JavaScript bytes per page. - +
Figure 1. Distribution of JavaScript bytes per page.
@@ -43,7 +43,7 @@ Looking at these numbers, it's only natural to wonder if this is too much JavaSc Figure 2. Distribution of JavaScript per page by device. - +
Figure 2. Distribution of JavaScript per page by device.
@@ -59,7 +59,7 @@ We can get an idea by analyzing main thread processing times for V8 at different Figure 3. V8 Main thread processing times by device. - +
Figure 3. V8 Main thread processing times by device.
@@ -83,7 +83,7 @@ One avenue worth exploring when trying to analyze the amount of JavaScript used Figure 5. Distribution of total JavaScript requests. - +
Figure 5. Distribution of total JavaScript requests.
@@ -99,7 +99,7 @@ Third-party JavaScript can come from any external, third-party source. Ads, anal Figure 6. Distribution of first and third-party scripts on desktop. - +
Figure 6. Distribution of first and third-party scripts on desktop.
@@ -107,7 +107,7 @@ Third-party JavaScript can come from any external, third-party source. Ads, anal Figure 7. Distribution of first and third party scripts on mobile. - +
Figure 7. Distribution of first and third party scripts on mobile.
@@ -117,7 +117,7 @@ For both mobile and desktop clients, more third-party requests are sent than fir Figure 8. Distribution of total JavaScript downloaded on desktop. - +
Figure 8. Distribution of total JavaScript downloaded on desktop.
@@ -125,7 +125,7 @@ For both mobile and desktop clients, more third-party requests are sent than fir Figure 9. Distribution of total JavaScript downloaded on mobile. - +
Figure 9. Distribution of total JavaScript downloaded on mobile.
diff --git a/src/content/en/2019/media.md b/src/content/en/2019/media.md index 6e72e95827c..7522b76fe3f 100644 --- a/src/content/en/2019/media.md +++ b/src/content/en/2019/media.md @@ -22,7 +22,7 @@ From a pure bytes perspective, HTTP Archive has [historically reported](https:// Figure 1. Web page bytes: image and video versus other. - +
Figure 1. Web page bytes: image and video versus other.
@@ -54,7 +54,7 @@ There are three metrics to consider when looking at pixel volume: CSS pixels, na Figure 3. Image pixels per page (mobile): CSS versus actual. - +
Figure 3. Image pixels per page (mobile): CSS versus actual.
@@ -62,7 +62,7 @@ There are three metrics to consider when looking at pixel volume: CSS pixels, na Figure 4. Image pixels per page (desktop): CSS versus actual. - +
Figure 4. Image pixels per page (desktop): CSS versus actual.
@@ -84,7 +84,7 @@ If we had one image that filled the entire screen perfectly, this would be a 1x Figure 5. Image pixel volume versus screen size (CSS pixels). - +
Figure 5. Image pixel volume versus screen size (CSS pixels).
@@ -168,7 +168,7 @@ Of course, web pages are not uniform in their use of image content. Some depend Figure 8. Image format usage per page. - +
Figure 8. Image format usage per page.
@@ -192,7 +192,7 @@ There are two ways to look at image file sizes: absolute bytes per resource and A comparison of image formats by file size - +
Figure 10. File size (KB) by image format.
@@ -202,7 +202,7 @@ From this we can start to get a sense of how large or small a typical resource i Figure 11. Bytes per pixel. - +
Figure 11. Bytes per pixel.
@@ -226,17 +226,17 @@ One [Lighthouse](./methodology#lighthouse) test is an A/B comparing baseline wit Figure 12. Percent 'optimized' images. - +
Figure 12. Percent "optimized" images.
-The savings in this AB Lighthouse test is not just about potential byte savings, which can accrue to several MBs at the p95, it also demonstrates the page performance improvement. +The savings in this AB Lighthouse test is not just about potential byte savings, which can accrue to several MBs at the 95th percentile, it also demonstrates the page performance improvement.
Figure 13. Projected page performance improvements from image optimization from Lighthouse. - +
Figure 13. Projected page performance improvements from image optimization from Lighthouse.
@@ -331,7 +331,7 @@ Earlier, in Figure 5, we showed that the volume of image co Figure 19. Lighthouse audit: Offscreen. - +
Figure 19. Lighthouse audit: Offscreen.
From 18fbd339a38d5ffc58aef93561cfe5f7a21f6490 Mon Sep 17 00:00:00 2001 From: Barry Date: Tue, 25 Aug 2020 23:49:01 +0100 Subject: [PATCH 2/4] Add colgroup scope --- src/content/en/2019/cdn.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/content/en/2019/cdn.md b/src/content/en/2019/cdn.md index 58bb203b1a1..eac7aa39e9c 100644 --- a/src/content/en/2019/cdn.md +++ b/src/content/en/2019/cdn.md @@ -486,7 +486,7 @@ In contrast, the median TLS negotiation for the majority of CDN providers is bet - Percentile + Percentile @@ -687,7 +687,7 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - Percentile + Percentile @@ -861,7 +861,7 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - Percentile + Percentile From 0c39f9c6c5d8d143779127f97a3e39a5616d6396 Mon Sep 17 00:00:00 2001 From: Barry Date: Fri, 28 Aug 2020 14:52:54 +0100 Subject: [PATCH 3/4] Better wway of handling percentile headers --- src/content/en/2019/cdn.md | 28 ++++++++++------------------ 1 file changed, 10 insertions(+), 18 deletions(-) diff --git a/src/content/en/2019/cdn.md b/src/content/en/2019/cdn.md index eac7aa39e9c..91310f4c0f4 100644 --- a/src/content/en/2019/cdn.md +++ b/src/content/en/2019/cdn.md @@ -486,15 +486,11 @@ In contrast, the median TLS negotiation for the majority of CDN providers is bet - Percentile - - - - 10 - 25 - 50 - 75 - 90 + + + + + @@ -687,15 +683,11 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - Percentile - - - - 10 - 25 - 50 - 75 - 90 + + + + + From e7fb6b38ae10bbee0403be6fe4a5367708e39fb4 Mon Sep 17 00:00:00 2001 From: Barry Date: Fri, 28 Aug 2020 14:59:34 +0100 Subject: [PATCH 4/4] Missed one --- src/content/en/2019/cdn.md | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/src/content/en/2019/cdn.md b/src/content/en/2019/cdn.md index 91310f4c0f4..3d824bd6bda 100644 --- a/src/content/en/2019/cdn.md +++ b/src/content/en/2019/cdn.md @@ -853,15 +853,11 @@ Most CDNs balance the need for shared certificates and performance. Most cap the - Percentile - - - - 10 - 25 - 50 - 75 - 90 + + + + +