W3C home > Mailing lists > Public > w3c-wai-hc@w3.org > October to December 1997

RE: CSS 2: priorities in cascading order

From: Charles (Chuck) Oppermann <chuckop@MICROSOFT.com>
Date: Thu, 18 Dec 1997 16:34:51 -0800
Message-ID: <E3A3FFB80F5CD1119CED00805FBECA2F013BBABF@red-msg-55.dns.microsoft.com>
To: "'Jason White'" <jasonw@ariel.ucs.unimelb.EDU.AU>, WAI HC Working Group <w3c-wai-hc@w3.org>
Jason,
It was my impression that the "author can always win" scenario was put in
the achieve acceptance of CSS1 by authoring groups.  To change that, I see
three possible paths:

1)	Additional syntax changes to allow for more attributes (your option
#2)
2)	Change the definition of CSS to say that a reader attribute "wins"
when faced with the same level author attribute (your option #1)
3)	Recommend that browsers do what IE4 does - provide absolute
overrides in the user interface.

If we add more attributes, what's to prevent someone from always using the
"!SUPER-DUPER_INCREDIBLY_IMPORTANT_ULTIMATE" attribute on their site simply
because it's a guarantee of appearance.

Option #3 I feel is best because it's the easiest to implement:  Millions of
web authors, verses hundreds of thousands of folks needing their user styles
verses a handful of browsers (minus the one that already does it).

Charles Oppermann
Active Accessibility, Microsoft Corporation
mailto:chuckop@microsoft.com http://microsoft.com/enable/
"A computer on every desk and in every home, usable by everyone!"


	-----Original Message-----
	From:	Jason White [SMTP:jasonw@ariel.ucs.unimelb.EDU.AU]
	Sent:	Thursday, December 18, 1997 2:36 PM
	To:	WAI HC Working Group
	Subject:	RE: CSS 2: priorities in cascading order

	On Thu, 18 Dec 1997, Charles (Chuck) Oppermann wrote:

	> Up on http://microsoft.com/enable/products/ie4.htm
	> <http://microsoft.com/enable/products/ie4.htm>  we have a couple
of High
	> Contrast style sheets that can be downloaded.
	> 
	I have visited the Microsoft site and found the style sheets, which
	demonstrate the problem and show that it is not merely hypothetical.
Any
	of these style sheets would be overridden by contrary provisions in
an
	author's style sheet.

	The fact that a user agent may include controls in its user
interface
	permitting the user to disable the application of certain style
sheets is
	helpful, but not sufficient. Rather, the user should be able to
determine
	which of the provisions in his or her default style sheet will apply
	regardless of the author's style sheet, and which can be overridden.

	For this reason I would reiterate the importance of either (1)
changing
	the definition of !important; so that an author's styles always
carry
	normal weight; or (2) adding an !absolute; directive to achieve the
same
	effect. The "specificity" problem as outlined in one of my earlier
	contributions to this discussion also requires attention.

	The fact that Microsoft have implemented "readers' style sheets" has
	changed my thinking somewhat with regard to the importance of this
issue.
	The fact that it is not a major practical problem today is merely a
	consequence of there being so few "authors' style sheets" available
on the
	web. This will change in the future, perhaps quite rapidly,
particularly
	if and when CSS 2 becomes a recommendation and is widely supported
by
	authoring tools and web browsers.
	
Received on Thursday, 18 December 1997 19:35:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:56:12 UTC