W3C home > Mailing lists > Public > www-style@w3.org > December 2004

Re: Multiple Background Images

From: Ben Ward <benmward@gmail.com>
Date: Thu, 2 Dec 2004 09:44:09 +0000
Message-ID: <ef5d0f2f0412020144549391f4@mail.gmail.com>
To: Staffan Mhln <staffan.mahlen@comhem.se>, fantasai <fantasai.lists@inkedblade.net>, www-style@w3.org

> Finally, should we expect other properties that can require extra elements to be
> arrayed in addition to the background-X ones? (Multiple borders and outlines seems
> like potential use cases for instance? But then again, would a list of 1 to 4 values make
> sense?)

This is where my concern arises. I personally see nothing /wrong/ with
comma separated syntax. I just think that doing it through an indexed
syntax would be better (as well leaving us open for a fallback,
'alternate background' syntax with commas which would be far
more consistent with the font-family comma separated list behaviour).

Assuming, for a moment that commas was adopted and that CSS isn't
interested in degrading, alternate backgrounds, does this not cause
another problem for CSS generally? It locks out other multivalue
behaviour for the background property forever, without consideration
for a consistent array model.

I'm increasingly drawn toward trying to draw up something stable and
usable that would permit any multivalued property, and not restricted
to whether the existing property already has a multivalued function.
But if (as looks likely) that cannot be worked out acceptably, then
does it suggest that we're reaching the limits of what CSS can achieve
in its current syntax? I'm worried that it places a brick wall on
CSS's future usefulness and hints that an XPATH/XML replacement for
CSS is on the fringe of need to accommodate these new ideas. As
technically 'cool' (to XML type people, maybe) as that idea could be,
calling time on CSS over multivalued properties could really burn the
www. If it [universal, multivalued properties] can be made to work,
should they be? Or should we be striving only for 'perfection' and
reject all else?

Ultimately, is this just something that has to be accepted?
Received on Thursday, 2 December 2004 09:44:10 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:27:16 UTC