W3C home > Mailing lists > Public > www-dom@w3.org > October to December 1999

Re: TransformNode method

From: <keshlam@us.ibm.com>
Date: Tue, 2 Nov 1999 19:31:46 -0500
To: "Michael Champion" <michael_champion@ameritech.net>
cc: www-dom@w3.org
Message-ID: <8525681E.0003840C.00@D51MTA03.pok.ibm.com>
The XSL Working Group has certainly expressed an interest in this. But the
question of whether XSL processing is something applied _by_ a DOM, or _to_
a DOM, is open to debate. Note that the DOM's CSS support currently
consists only of associating style rules with the document, and doesn't
(yet) address actually running the CSS processor.

The strongest argument in favor of building styling into the DOM is that
once you start talking about actively working with a transformed page, you
may need to associate from the source document to the styled version and
back -- and that may require active support from the two documents
themselves.

The strongest argument against is that you may want to use the same basic
API to apply XSL to other document models... and that the DOM may not be
the optimum representation for an XSL processor to use internally. The DOM
is _a_ W3C API; it doesn't have to be the only W3C API.

This is definitely an area where the two working groups will have to work
with each other to determine what will best address the needs of the users.

______________________________________
Joe Kesselman  / IBM Research
Received on Tuesday, 2 November 1999 19:38:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 22 June 2012 06:13:46 GMT