W3C home > Mailing lists > Public > public-tt@w3.org > March 2005

RE: [DFXP LC Comment] Some questions (was: Re: [tt] Some questions)

From: Glenn A. Adams <gadams@xfsi.com>
Date: Tue, 22 Mar 2005 09:01:11 -0500
Message-ID: <7249D02C4D2DFD4D80F2E040E8CAF37C0E9093@longxuyen.xfsi.com>
To: "Anne van Kesteren" <fora@annevankesteren.nl>
Cc: <public-tt@w3.org>



> -----Original Message-----
> From: Anne van Kesteren [mailto:fora@annevankesteren.nl]
> Sent: Tuesday, March 22, 2005 8:40 AM
> To: Glenn A. Adams
> Cc: public-tt@w3.org
> Subject: Re: [DFXP LC Comment] Some questions (was: Re: [tt] Some
> questions)
> 
> Glenn A. Adams wrote:
> > [GA] Sorry. Perhaps I wasn't clear, but I was inviting you to make a
> > persuasive argument for use of xml:id that I could take back to the
WG.
> > I am not prepared to make such an argument, particularly since I am
one
> > who would need to be convinced of using xml:id.
> 
> It would remove the need for a validating parser to recognize the 'id'
> attribute is of type ID.

[GA] But it would require a special validating parser that is not yet
generally available, namely, one with intrinsic semantics regarding the
processing of xml:id. I do not believe the argument you cite is
sufficient motivation to adopt xml:id; however, I will communicate your
request and rationale to the WG.

> 
> > A requirement of TT AF was that all information be represented as
XML.
> > We believe we have satisfied that requirement.
> 
> Styling isn't really information. You could have distinquished it as
> such, but that might be a wrong observation. Moving styling inside the
> document takes away the advantages of user specific style sheet or
style
> sheet switching where the author could provide multiple styles for the
> same page and leave the choice to the user.

[GA] As has already been discussed in an earlier thread, DFXP was
explicitly designed to integrate styling information inline with content
(or to optionally aggregate such information then reference it by means
of referential styling features). From the perspective of the TT WG,
style information *is* information of a stylistic nature.

I don't intend to entertain a debate about the relative merits and
demerits of inlined or integrated styling. This discussion has been
repeated far too many times in W3C fora. Let's just say, there are valid
reasons for separating style information from content (semantic)
information and valid uses for integrating the two.
 
> Embedding everything in the document takes away all those advantages
and
> clutters the semantics, just like <font> did.

[GA] The intention of AFXP, which DFXP is not, is to provide an author
the tools they need to implement a policy of separating style
information, potentially even in separate (stylesheet) resources.
However, for DFXP, there was and is an explicit requirement to limit the
complexity of the document representation and to support streamability
scenarios where applicative style rules are problematic, and where
inlined style information is preferred.

In conclusion, I'm afraid it will be necessary to turn down your request
to enforce a separation style information from content information in
DFXP. I don't see any movement in the TT WG position on this point. But
please stay tuned for AFXP.

Regards,
Glenn
Received on Tuesday, 22 March 2005 14:01:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 2 November 2009 22:41:32 GMT