Le 22 oct. 2012 à 18:10, Philip TAYLOR <[email protected]> a écrit :

> Thank you for your comments, Philippe, for which I am very grateful.
> I am, however, puzzled by your view that it can be considered a feature
> (albeit an experimental feature) rather than a bug.
> 
> If an implementation chooses to ignore the wording of the current
> specification, which according to :
> 
>       http://www.w3.org/Style/CSS/Overview.en.html
> 
> is CSS 2.1 as amended by CSS Color Level 3, CSS Namespaces and
> Selectors Level 3, and implements a behaviour that directly
> contradicts that specification, how can that be classified as
> anything other than a bug ?
> 
> In my view (which I do not think is heretical), an author should be
> able to /rely/ on a W3C specification, not have to test his/her work
> against every extant browser -- would you not agree ?

1. CSS 2.1 is a REC (a recommendation), it is not a law punishable with the 
death penalty (we wouldn't have much browser in that case)
2. For ages, work has been going on to extend CSS 2.1; the noted CSS-content 
module is one such example. Browser implentators can start implementing that 
work on an experimental basis and test it. Ideally, that would happen behind a 
vendor-prefix flag; obviously, that didn't happen that much at the time that 
CSS content module was first produced (10 years ago); there are (quite a few) 
other properties and values that are implemented in one or more rendering 
engines that way. Some would argue that such experimental feature shouldn't 
ship with release version of browsers. I think that would be sad; many of the 
features that are commonly used in designing pages nowadays wouldn't be as 
solid as they are today without those experimental features being included in 
release version of browsers.

So no, not a bug, but an experimental feature; an unfortunate one maybe, at 
least for you apparently, and I don’t see much use for that feature, 
personally, but nothing more. It is extending the boundaries of the possible. 
And some of those experiments stick, some fail. Some get improved, some turn 
out to be useless.

A bug is something more irrational, like, for example, Opera running on OS X 
making crapshoot of an element with { font-weight:  bold; font-style: italic; } 
- they are slanting the font, & substituting glyphs even though the specific 
font-family has all required faces. Creepily ugly doesn’t start describing it. 
Or Gecko being unable to draw a nice dotted border in combination with the 
border-radius property (an other real old bug). Real ugly results again.

Sometimes I wish I could ‘rely’ on the REC status of CSS 2.1 or other modules 
and save on cross browsers testing time. That always turn out to be wishful 
thinking. Without fail.  IOW, a pipe dream.

PS - If one makes an error in a stylesheet (did you wrote E { content: 'foo'; } 
instead of E::after {} ?) then sometimes strange things happen…

PPS - E { content: 'foo'; } actually validates.

Philippe
--
Philippe Wittenbergh
http://l-c-n.com




______________________________________________________________________
css-discuss [[email protected]]
http://www.css-discuss.org/mailman/listinfo/css-d
List wiki/FAQ -- http://css-discuss.incutio.com/
List policies -- http://css-discuss.org/policies.html
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/

Reply via email to