W3C home > Mailing lists > Public > www-xml-infoset-comments@w3.org > July to September 2001

Re: Response to your comment on the XML Infoset CR draft

From: Richard Tobin <richard@cogsci.ed.ac.uk>
Date: Thu, 2 Aug 2001 13:58:48 +0100 (BST)
Message-Id: <200108021258.NAA05970@mcpherson.cogsci.ed.ac.uk>
To: Noah_Mendelsohn@lotus.com
Cc: www-xml-infoset-comments@w3.org
I hope I can lay your worries to rest...

>Let's say I create an empty DOM on my program, and then use the DOM API's 
>to add element and attribute nodes.   I pass the resulting DOM to some 
>other program, perhaps as input to an XSLT transform.  I then free up the 
>DOM.  This is a case in which the document has never existed in 
>well-formed "<...>" syntax.  Are you really intending to say, and indeed 
>to emphasize, that your infoset specification does not cover this case?

It all depends on what you mean by "cover".  Certainly this is exactly
the sort of thing that we expect to happen.  But what would the
Infoset spec say about it?  In particular, what would it *specify*
about it?  It wouldn't specify how you create the infoset - the DOM
would specify that.  It wouldn't specify how you pass it to some other
program - perhaps some future processing model spec might.  It
wouldn't specify that the DOM-created infoset is acceptable input to
XSLT - XSLT will implicitly specify that if it is changed to describe
its input in Infoset terms.  I don't say how we can say more than we
do at present: that infosets may be created by means other than
parsing a document.

-- Richard
Received on Thursday, 2 August 2001 08:58:51 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 16 March 2009 11:12:24 GMT