W3C home > Mailing lists > Public > www-tag@w3.org > October 2003

RE: Action item on syntax-based interoperability

From: Elliotte Rusty Harold <elharo@metalab.unc.edu>
Date: Wed, 22 Oct 2003 20:51:42 -0400
Message-Id: <p0600200abbbcd5e01d70@[192.168.254.4]>
To: "Joshua Allen" <joshuaa@microsoft.com>, "Tyler Close" <tyler@waterken.com>, "Tim Bray" <tbray@textuality.com>
Cc: <www-tag@w3.org>

At 5:05 PM -0700 10/22/03, Joshua Allen wrote:
>>  XInclude is another effort that's been excessively hindered by
>>  defining itself in terms of infoset rather than syntax. It enables
>
>Are you saying that people found it difficult to understand or implement
>XInclude because of basis in infoset?  If so, I would say that
>observation does not apply to us, nor to the guys who did XInclude.NET.
>The spec was not all that bad, and if anything the potential confusing
>or ambiguous issues are all about "processing model".

Yes I'm saying exactly that. The issues of handling notations and 
unparsed entities were the worst that came from the infoset nature. 
Namespace declarations and base URIs were another problematic point.

Perhaps you're smarter than the rest of us, or perhaps you just got 
lucky, or had a better local model to build on top of, so you didn't 
encounter the problems, I, the Apache team, and the Gnome Project 
have had to struggle with. Or perhaps you have bugs in your 
implementation you aren't aware of yet. I haven't looked at 
XInclude.NET personally so I don't know.

-- 

   Elliotte Rusty Harold
   elharo@metalab.unc.edu
   Processing XML with Java (Addison-Wesley, 2002)
   http://www.cafeconleche.org/books/xmljava
   http://www.amazon.com/exec/obidos/ISBN%3D0201771861/cafeaulaitA
Received on Wednesday, 22 October 2003 22:23:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:22 GMT