Re: several messages

On Fri, 2 Apr 2004, Boris Zbarsky wrote:
>> Those people were not the focus of the discussion. The discussion
>> started because you said that changes in the specs were an important
>> reason for delays in UAs reaching good compliance levels.
> Yes. And they are.

In the cases of features that _changed_ (inherit, cascade of style and
presentational hints), as opposed to the introduction of new features or
extensions to the syntax, I think the changes are helping speed up the
convergence towards interoperability. Do you disagree?

In the cases of features that were added, they are easy relatively easy
(addition of _ and - to certain identifiers), necessary for better
interopability with already widespread UAs (background-position), or
simply common sense extensions (multiple classes, display on generated
content). I don't see that they are going to have a particularly big
impact on how long it takes for UAs to reach good compliance levels.

> Since it seems that the authoring community (and with CSS2.1 the WG
> itself, even if for excellent reasons) constantly pushes to implement
> whatever the newest level is, effort is diverted to implementing bits of
> this instead of being focused on fixing issues in the existing
> implementation of the oldest spec level.  Hence the lack of bug-free
> CSS1 support in any browser currently on the market (I'm ignoring the
> places where CSS2 contradicts CSS1 for now; I believe even the rest of
> CSS1 is not properly implemented in anything in existence).

Most of the holes in CSS1 support (list items, margin collapsing, floats)
are actually bugs in CSS2.1 support, because CSS1 was very vague about
those features, and CSS2.1 finally specified them well.

'display: list-item' is about the only thing Mozilla gets wrong from CSS1,
if you don't count bugs that are within CSS1's vagueness but wrong per
CSS2.1 (changes that were needed to ensure interoperability). (Well, there
are also a few float bugs that might be CSS1 bugs, and there are minor
things with padding on inline elements that you can spot if you look very
closely in the CSS1 test suite, and some minor issues with font-weight,
and some paining errors, but that's about it. I'll _always_ be able to
find a bug if I look hard enough, or I'm not doing my job right...)

Ian Hickson                                      )\._.,--....,'``.    fL
U+1047E                                         /,   _.. \   _\  ;`._ ,.                         `._.-(,_..'--(,_..'`-.;.'

Received on Friday, 2 April 2004 10:27:29 UTC