A partial archive of discourse.wicg.io as of Saturday February 24, 2024.

Update on Vendor Prefixing

stuartpb
2015-04-21

Wasn’t there some hubbub a couple years ago about how the web was heading toward some kind of Prefix-ageddon and there needed to be some kind of overhaul of the vendor-prefix convention, lest we all be hurtled into an oblivion where every UA would be handling -webkit- rules?

Whatever happened with that? It seems to me that most of these concerns were ameliorated somewhere around the advent of CSS pre/postprocessors, with Autoprefixer more or less bringing vendor prefixes back into the realm of “good ideas”.

tabatkins
2015-04-22

What happened is, Firefox and Blink stopped using prefixes, opting instead to hide/show things based on flags. WebKit decided to keep using prefixes (when only one person is doing it, I guess it’s not quite as terrible?) and I think IE is on the no-prefix bandwagon, but I’m not sure.

simevidas
2015-04-24

A few details about Microsoft’s browsers: IE11 is frozen and the version of IE in Windows 10 will be pretty much the same old IE11 from before. Meanwhile, Project Spartan supports a set of -webkit- properties in order to improve interoperability with the Web.

myakura
2015-04-24

pre/postpcoressors only helps newly-developed-or-maintained sites; browsers still need to handle sites coming only with -webkit-prefixed features. Mozilla recently enabled “CSS Unprefixing Service” for some hi-traffic sites, doing pretty much the same thing as IE does (converting -webkit-prefixed CSS properties to the standard equivallents).

stuartpb
2015-06-05

Did they, though? It feels like there are still a ton of new CSS features (and JS APIs) coming from Blink that are behind -webkit- vendor prefixes, and I’m pretty sure they don’t all predate the flags approach.

tabatkins
2015-06-05

Can you give some examples?