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

RE: Fallbacks

From: Ernest Cline <ernestcline@mindspring.com>
Date: Tue, 13 Apr 2004 11:13:50 -0400
Message-ID: <410-220044213151350171@mindspring.com>
To: "Anne van Kesteren (fora)" <fora@annevankesteren.nl>
Cc: www-style@w3.org


> [Original Message]
> From: Anne van Kesteren (fora) <fora@annevankesteren.nl>
>
> I think it would be even better if _every_ single property
> could have a comma separated fallback.

<snip>

> Such an extension for every property is also very good for backwards 
> compatibility. Lot's of people wanted a '@useragent' or '@support' rule, 
> having a comma separated list that isn't really necessary anymore (if it 
> was necessary in the first place is a different topic).

I'll have to severely disagree with you here.

The main benefit of an @support rule, is that it would give authors
the flexibility of using a particular value of a property only if the UA
supports another value or property.  Your proposal gives a much
more limited version of such an ability, because it is useful only
with shortcut properties such as 'font' that combine several
related properties together.  The problem with such an approach
is that burdens CSS with the responsibility of deciding which
combinations of properties belong together, and with
implementing the shortcut properties to go with it.  Adding one
or two properties isn't a major burden, but as the quote attributed
to Sen. Dirkson goes, " A billion here, a billion there, and pretty
soon you're talking real money."

@support has the advantage that the CSS spec doesn't have
to decide which properties an author might wish to have depend
on each other. and it doesn't require adding addition properties
to support additional interdependencies.

(Note: I am strongly against @useragent.)
Received on Tuesday, 13 April 2004 11:13:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:29 GMT