W3C home > Mailing lists > Public > www-style@w3.org > September 2002

Re: Specified values: what are they?

From: Etan Wexler <ewexler@stickdog.com>
Date: 26 Sep 2002 18:00 -0700
To: www-style@w3.org
Message-ID: <CSS-values@d20020926.etan.wexler>

Ian Hickson wrote to <www-style@w3.org> on 26 September 2002 in "Re:
Specified values: what are they?"
(<mid:Pine.LNX.4.21.0209261350240.15733-100000@dhalsim.dreamhost.com>):

> So in CSS3, you want "specified" and "cascaded" to mean the same
> thing?

No, that is not quite the case.  I want CSS3 to avoid the term
"specified value" altogether.  "Specified value" will retain its
unambiguous meaning from CSS2, which, yes, has the same meaning as is
proposed for "cascaded value".

What I seek to achieve is disambiguation.  When somebody writes
"specified value", I should not have to ask, "Do you mean 'specified
value' like in CSS2 or like in CSS3?"

> > Does replacement of 'attr()' forms really take place during
> > assignment of cascaded values?  Why replace 'attr()' forms
> > then instead of during assignment of computed values?
>
> The value extracted out of attr() could require further computation,
> e.g. if it is a percentage.

Certainly this is true. However, I had envisioned that the "attr()" form
be the cascaded value.  In this case, the CSS processor would, after
assigning the cascaded value but before assigning the computed value, both
retrieve the referenced attribute value and process that attribute value.

-- 
Etan Wexler <mailto:ewexler@stickdog.com>
This message has been brought to you by Pontiac.
"Pontiac: a passion for performance."
Received on Thursday, 26 September 2002 20:48:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:16 GMT