real_ozcoder's profile picture.

real_ozcoder ๐Ÿ’‰๐Ÿ’‰๐Ÿ’‰๐Ÿ’‰

@real_ozcoder

Is it better for #WebPerf to under or overestimate contain-intrinsic-height in pixels (up to 24px wrong maybe) for different sections in a page which itself is greater than 4000px tall?


Anybody got any #GitTricks to extend #git usage? I want to be able to diff a file between 2 branches, but with a regex to ignore certain (debugging messages) lines.


I forgot that Javascriptแต€แดน is trademarked. Any #webperf professionals (and up and comers) want to add to the @rough__sea letter to Oracle at javascript.tm , as heard on the @changelog podcast.


I didn't realise that clicking on the purple ๐˜€๐—ฐ๐—ฟ๐—ผ๐—น๐—น button in #FireFox Developer Tools would highlight which elements are causing the overflow. Nice feature @FirefoxDevTools . I also like using the grey ๐—ฒ๐˜ƒ๐—ฒ๐—ป๐˜ button.

real_ozcoder's tweet image. I didn't realise that clicking on the purple ๐˜€๐—ฐ๐—ฟ๐—ผ๐—น๐—น button in #FireFox Developer Tools would highlight which elements are causing the overflow. Nice feature @FirefoxDevTools . I also like using the grey ๐—ฒ๐˜ƒ๐—ฒ๐—ป๐˜ button.

Finally last Friday my แดข๊œฑแด›แด… test finished. 49,376,250 variations of options ๐•ค๐•ฅ๐•ฃ๐•’๐•ฅ, ๐•จ๐•๐• ๐•˜, ๐•ค๐•๐• ๐•˜, ๐•™๐•๐• ๐•˜, ๐•”๐•๐• ๐•˜ and ๐•ฅ๐•๐•–๐•Ÿ produced smallest size of 80,581. Input ๐˜€๐—ถ๐˜๐—ฒ.๐—ฐ๐˜€๐˜€ is 514,725. Brotli compresses better at 77,849. Even Bzip2 gives 72,265 #webperf


Iโ€™m attending JSNation โ€“ the main JavaScript conference of the year. You may join me there for free with 10k other JS engineers and 40+ great speakers. Just follow this badge. #javascript portal.gitnation.org/badges/jsnatioโ€ฆ #jsnation


I was thinking, for #WebPerf , you probably shouldn't preconnect to your #font provider for logged in customers, since they have already downloaded the fonts when they logged into your site.


I wonder if anybody uses github.com/Genymobile/scrโ€ฆ or something similar for #webperf measurements and automated testing?


I would be nice if @newrelic Genesys chat assets had versioned (or MD5 suffixes) filenames so that they could have a breakable 30 day cache settings. Don't really the same 15+ files downloaded on each page of a site. #WebPerf


After some simple reordering in HEAD using #capo for #WebPerf analysis. Might be able to do a little bit more.

real_ozcoder's tweet image. After some simple reordering in HEAD using #capo for #WebPerf analysis. Might be able to do a little bit more.

Someones old #techno page on #tripod monstermo.tripod.com/techno.htm from last century make me feel a little bit nostalgic.


It appears that @salesforce is using an old Java server to serve their JavaScript (deployment.js 42.5Kb) for Chat. It might be better for #webperf if they used a CDN Eg. @cloudflare etc. that served Brotli (10Kb) or Gzip (12.5-11.5Kb) from the Edge.


Do people get Apache to gzip on-the-fly returned JSON from an API calls for #WebPerf reasons? If so do you only do it if the JSON is bigger than a set amount, say 1Kb?


I wonder who will be the first #rapper in #space to film a #MusicVideo and claim some of the #spaceflight on tax?


Loading...

Something went wrong.


Something went wrong.