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

Re: CSS Suggestion - based-on:

From: David Woolley <david@djwhome.demon.co.uk>
Date: Sat, 11 Dec 2004 11:18:02 +0000 (GMT)
Message-Id: <200412111118.iBBBI2n02805@djwhome.demon.co.uk>
To: www-style@w3.org

> I assume the feature you meantion continually gets rejected too? I'm

I'm only on the public list, so all I know is that it is always the
consensus of the list to reject it and it never appears in the official
proposals.

>  sure there's a reason for it though. Is there an explanation somewhere
>  in the archives. If there's a fundamental problem with that, that also

Every time it is proposed the reasons against are given.  I tried to give
a quick summary.  The formal decisions are made in private and the
reasoning is not published - presumably because it can reveal product
plans, etc.

> How does it do that?

Because you may end up basing on a style that you didn't provide, and,
possibly because of various circularity/forward referencing issues.

> I don't see how this in anyway would prevent the user from changing
>  your style if they wished, since the user stylesheet would be applied
>  last, and has the highest priority.

Actually, that's an oversimplification of the precedence rules, but, in
any case, your "based on" rule might end up based on their override.
You, at least, need to work out all these sorts of combinations and 
specify how they are resolved.

> That's fine when it's a simple example there, but when you've got a
>  large site with many different rules the CSS tends to get
>  exceptionally messy. You end up with things like this, which are
>  unavoidable.

If you have got many different rules, you almost certainly have a 
design that is styled arbitrarily.  The house style for a good design ought
to be specifiable in a quite small amount of plain language and, hopefull,
in a similarly small amount of CSS.

If we assume that styling is supposed to aid usability, users can cope with
remembering very few exceptions to rules, especially in the public web
context, where they may have had to learn the rules by example for a site
they may only visit once.

> #navigation #products.opened a:link, #navigation #products.opened
>  a:hover (except with about 5 rules added on there)

But you should want :hover to produce the same change everywhere on the
page.

> > [And, if you find this not powerful enough, to use an authoring time
> > pre-processor to generate the expanded CSS from your proprietary
> > format.
> 
> That seems like a pretty flippant remark. I'm a little shocked by that.
>  The W3C is all about maintaining standards. If there's a limitation to

There are many limitations in all the standards imposed as deliberate
design decisions, e.g. many result from the policy of allowing the browser
user to override styles, and many arise from trying to keep the authoring
langauge simple, and many from the requirment that documents display 
reasonably on devices with wide ranges in age and capability.

>  them, then surely something should figured out to resolve that, rather

There isn't a limitation in terms of the terms of reference of CSS,
as a styling language for the public web, as you can pre-process in 
the authoring tool and send the resultant CSS.  CSS is not an authoring
tool file format it is the format for transmission to end users.

>  that telling someone to use a proprietary pre-processor. (does such a
>  thing even exist?)

There will be many non-productised batch pre-processors.  Getting an authoring
tool implementor to do this sort of pre-processing will be a lot quicker
process than getting all the browser developers to do it and then waiting
for users to upgrade.
Received on Saturday, 11 December 2004 12:46:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:35 GMT