Re: [css3-2d-transforms] support for transforms on inline elements to be removed from spec?

On Friday, 25 February 2011, Sylvain Galineau <sylvaing@microsoft.com> wrote:
>
> [Boris Zbarsky:]
>>
>> On 2/25/11 3:13 AM, Jon Rimmer wrote:
>> > With IE 9 and Firefox 4 close to release, transforms are likely to see
>> increased use, and I am concerned that their implementations are about to
>> be made non-standard by this spec change.
>>
>> Two notes:
>>
>> 1)  Both implementations are prefixed.  So even if the above happens it's
>> not the end of the world (and in particular, there are various other
>> incompatibilities between how transforms work in Webkit and how they work
>> in at least Gecko).
>>
>> 2)  No matter what happens with the spec at this point, the behavior of
>> Firefox 4 for transforms is not going to change for the release, unless we
>> find a security bug in them or something.  I clearly can's speak for the
>> IE team, but they're at RC... which probably means this applies to them
>> too.
>>
>> So I agree we need to decide this; I don't think the IE9 and Fx4 release
>> schedules are relevant to the decision.
>>
> +1 to all the above.

Fair enough, I didn't expect IE or Firefox to change at this point,
but I did want to raise the issue. As an author it's just
disappointing to start using a feature and then discover that Apple's
implementation doesn't match the spec they author, and that, having
not published a new WD of that spec since 2009, they are now planning
a change like this, but too late to affect the implementations in the
two most popular browsers.

As discussed recently, vendor prefixes are useful, but they do not
stop authors using these features in anger. There still needs to be
effort to progress towards interoperability, whether features are
prefixed or not. There is a Webkit bug about transforms on inline
elements[1] from early 2009 in which it was stated that they would
only work on blocks, yet the WD published 9 months later still has
them applying to inlines. It seems like just a little more
communication and co-ordination between implementors could avoid
situations like this, and move features towards stability instead of
having incompatible implementations sitting behind prefixes for years.

[1] https://bugs.webkit.org/show_bug.cgi?id=24919

Jon

Received on Saturday, 26 February 2011 11:06:02 UTC