cc:Mail Link to SMTP Undeliverable Message

Message is undeliverable.
Reason: User "brad_chase@met.bitstream.com" is not found in the cc:Mail Directory.
Original text follows:

Received: from www19.w3.org by met.bitstream.com (ccMail Link to SMTP R8.00.00)
	; Mon, 04 Aug 97 10:29:31 -0500
Return-Path: <www-style-request@w3.org>
Received: by www19.w3.org (8.8.5/8.6.12) id KAA03637; Mon, 4 Aug 1997 10:20:28 -0400 (EDT)
Resent-Date: Mon, 4 Aug 1997 10:20:28 -0400 (EDT)
Resent-Message-Id: <199708041420.KAA03637@www19.w3.org>
X-Authentication-Warning: www10.w3.org: Host SGI.COM [] claimed to be sgi.sgi.com
Message-ID: <33E5E4FF.167E@sgi.com>
Date: Mon, 04 Aug 1997 10:19:43 -0400
From: Douglas Rand <drand@sgi.com>
Organization: Silicon Graphics
X-Mailer: Mozilla 3.01SGoldC-SGI (X11; I; IRIX 6.3 IP22)
MIME-Version: 1.0
To: David Perrell <davidp@earthlink.net>
CC: www-style@w3.org
Subject: Re: CSS vs. transitional markup [was: No Subject]
References: <199708021856.LAA10414@italy.it.earthlink.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-List-URL: http://www.w3.org/pub/WWW/Archives/Public/www-style/
Resent-From: www-style@w3.org
X-Mailing-List: <www-style@w3.org> archive/latest/2697
X-Loop: www-style@w3.org
Sender: www-style-request@w3.org
Resent-Sender: www-style-request@w3.org
Precedence: list

David Perrell wrote:
> How are inline CSS STYLE declarations handled? Seems to me the only
> difference between CSS and HTML inline declarations is that one has
> maximum precedence and the other minimum. Both are valid only within
> the specific element instance in which they're declared. Once effective
> attributes are resolved, trumped ones are discarded. Couldn't this be
> done in one parse of the document, once the effective stylesheet is
> resolved?

Not the way it's been defined.  That's the problem.  They did this
nasty thing in the definition you posted - they placed the properties
*between* the UA and the author's stylesheet.  I'm not convinced that
elements like FONT can even work with that definition.  It looks to me
as if *any* font-family property will turn off FONT FACE, or font-size
turns off FONT SIZE,  even if the properties are for a parent element, 
e.g. BODY or HTML.  Other properties,  like text-align,  are much less
problematic because they won't necessarily appear in the UA's style
> If you're doing dynamic HTML, don't you need an attribute structure for
> each instance of each element?

We don't do dynamic HTML yet,  but I do keep a full document structure
which has all the attributes for every element.  I've thought a bit
about how to handle dynamic property changes,  but not for this release.

Doug Rand				drand@sgi.com
Silicon Graphics/SSO			http://reality.sgi.com/drand
Disclaimer: These are my views,  SGI's views are in 3D