- From: Tab Atkins Jr. <jackalmage@gmail.com>
- Date: Mon, 7 May 2012 00:50:59 +0200
- To: Rik Schaaf <coolcat_the_best@hotmail.com>
- Cc: "www-style@w3.org" <www-style@w3.org>
On Mon, May 7, 2012 at 12:36 AM, Rik Schaaf <coolcat_the_best@hotmail.com> wrote: > I understand that old browsers don't support new css properties, but even the big five shows their CSS2.1 based content slightly different. If you make a screenshot, the content should be 100% the same, not 99%. I can say without hesitation that those browsers won't reach that 100%. There are some areas, even of 2.1, that are undefined. Hopefully they're *explicitly* undefined, so we know they need to be filled in later, but I know that not everything is. For all specs, we allow some types of variance as a "quality of implementation" issue. In other words, browsers are allowed to do *better* in some circumstances without violating the spec. Finally, there are still bugs. CSS2.1 is a Recommendation, but that just means that every test we have is passed by two browsers. It doesn't mean that every test is passed by *all* browsers. > Ok that's clear now, but is there anything W3C will do if one of the big five does not implement their properties right? No, the W3C has no power. Standards are purely voluntary. Browsers obey standards because (1) a well-written standard makes their job easier, by defining all the difficult cases and doing all the difficult design work for them, and (2) if everyone else does something one way, your customers will want you to match them rather than doing it a different way. That's it - there's no forcing involved. ~TJ
Received on Sunday, 6 May 2012 22:51:49 UTC