Re: Some site performance analysis from Google
Le 31 mars 2015 à 07:43, James Graham ja...@hoppipolla.co.uk a écrit : There is also a view with some extra comments at [1] which are worth reading. [1] https://docs.google.com/document/d/1K-mKOqiUiSjgZTEscBLjtjd6E67oiK8H2ztOiq5tigk/view# And to not forget the title, if people missed the topic of that email from Google Perf Audits for Blink DevTools, CNet, Wikipedia, Time — *Blink* -- Karl Dubost, Mozilla http://www.la-grange.net/karl/moz ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Some site performance analysis from Google
On Tue, Mar 31, 2015 at 10:49 AM, Robert O'Callahan rob...@ocallahan.org wrote: This is a quick read and an interesting analysis of how various sites fail to achieve acceptable performance on mobile: https://docs.google.com/document/d/1K-mKOqiUiSjgZTEscBLjtjd6E67oiK8H2ztOiq5tigk/pub One clear message: blocking tracking - performance wins. Rob -- oIo otoeololo oyooouo otohoaoto oaonoyooonoeo owohooo oioso oaonogoroyo owoiotoho oao oboroootohoeoro oooro osoiosotoeoro owoiololo oboeo osouobojoeocoto otooo ojouodogomoeonoto.o oAogoaoiono,o oaonoyooonoeo owohooo osoaoyoso otooo oao oboroootohoeoro oooro osoiosotoeoro,o o‘oRoaocoao,o’o oioso oaonosowoeoroaoboloeo otooo otohoeo ocooouoroto.o oAonodo oaonoyooonoeo owohooo osoaoyoso,o o‘oYooouo ofolo!o’o owoiololo oboeo oiono odoaonogoeoro ooofo otohoeo ofoioroeo ooofo ohoeololo. ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Some site performance analysis from Google
On 30/03/15 22:49, Robert O'Callahan wrote: This is a quick read and an interesting analysis of how various sites fail to achieve acceptable performance on mobile: https://docs.google.com/document/d/1K-mKOqiUiSjgZTEscBLjtjd6E67oiK8H2ztOiq5tigk/pub There is also a view with some extra comments at [1] which are worth reading. [1] https://docs.google.com/document/d/1K-mKOqiUiSjgZTEscBLjtjd6E67oiK8H2ztOiq5tigk/view# ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Some site performance analysis from Google
On 30/03/15 23:55, Karl Dubost wrote: Le 31 mars 2015 à 07:43, James Graham ja...@hoppipolla.co.uk a écrit : There is also a view with some extra comments at [1] which are worth reading. [1] https://docs.google.com/document/d/1K-mKOqiUiSjgZTEscBLjtjd6E67oiK8H2ztOiq5tigk/view# And to not forget the title, if people missed the topic of that email from Google Perf Audits for Blink DevTools, CNet, Wikipedia, Time — *Blink* If you just mean that the specifics of this analysis have all been done in Blink then yes, that's true and the document certainly mentions Blink bugs that aren't directly relevant to Gecko. But there are a lot of very general lessons in there about the perf-destroying things that authors do, which provides an indication of where we should be improving the platform to provide more performant APIs, the kind of data that we need to surface in devtools to ensure that authors can investigate their performance issues, and the kind of outreach we ought to be doing to get past the DOM is slow meme into a more reasonable discussion about web perf and how to optimise sites. It would certainly be interesting to do a similar analysis in Gecko to try and surface specific perf bugs that we could fix. ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
Re: Some site performance analysis from Google
Le 31 mars 2015 à 08:14, James Graham ja...@hoppipolla.co.uk a écrit : If you just mean that the specifics of this analysis have all been done in Blink then yes, That's all. That was context. very general lessons in there about the perf-destroying things that authors do, which provides an indication of where we should be improving the platform to provide more performant APIs, Definitely. the kind of data that we need to surface in devtools to ensure that authors can investigate their performance issues ok here why I gave context. Some of these sites (example CNet) have a tendency to send very different content based on UA sniffing, JS script path, etc. etc. So when testing in Gecko devtools, the results might differ for some of the findings because anyway Gecko does not receive the same thing. :/ Also devtools are sometimes not neutral themselves with regard to the analysis. Another layer of variability. But yes 1+ very interesting study. To reinforce the point that rob made: Slow Web == Ads Quantified Web. Things like µblock helps to understand how fast the Web can be ;) https://addons.mozilla.org/en-US/firefox/addon/ublock/ -- Karl Dubost, Mozilla http://www.la-grange.net/karl/moz ___ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
