W3C home > Mailing lists > Public > www-style@w3.org > June 2013

Re: [css-variables] Remaining Issues

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Wed, 12 Jun 2013 12:32:21 +0200
Message-ID: <CAAWBYDAWkMCFuXYMeq5cqW9ov2qUd-qA4LPS=7kQwEMQi1U=sw@mail.gmail.com>
To: Fran├žois REMY <francois.remy.dev@outlook.com>
Cc: Sylvain Galineau <galineau@adobe.com>, www-style list <www-style@w3.org>
On Tue, Jun 11, 2013 at 11:56 AM, Fran├žois REMY
<francois.remy.dev@outlook.com> wrote:
>> Describing the kind of
>> use-cases that would be 'completely broken' adds much more value.
>
> I just had another idea, which is more related to the traditional role of [css-variables] aka replacing constant-like color constants.
>
> I can clearly see how someone would love to animate the 'var-shadow-angle' property of a webpage showing a few objects and shadows generated for them based on a calculation depending on the sun position on the page.
>
> Or, change the color of all the elements on the page using a reference color (for example, bouncing a few times between 'var-player1-color: green' and 'var-player1-color: red' when the player of a Age-of-Empires-like game change of status from alley to enemy and having all the units & buildings of that player switch to red for the parts of the SVG using the var-player1-color as background)).

Heh, this shows off even more strongly why we *can't* do reasonable
custom property animation until we have type annotations.  "Colors"
can be represented as an ident token, a hash token, or a function.
None of these can be interpolated in any reasonable way unless you
already know for a fact that they should be interpreted as colors.

~TJ
Received on Wednesday, 12 June 2013 10:33:08 UTC

This archive was generated by hypermail 2.3.1 : Monday, 2 May 2016 14:39:12 UTC