DOM and Xapi-J

[Peter Murray-Rust]

In message <33EA6992.B6615F6D at datachannel.com> john at datachannel.com (John Tigue) writes:
[...]
> 
> I think we can declare Xapi-J 1.0 complete at any time now. When the DOM

I think this is a great achievement, and I'd like to thank John both for the
API and for continuing the momentum. Also thanks to everyone who has 
contributed ideas.

This group is not, of course, part of the formal process of XML under W3C, but 
I believe that anyone involved in XML development will take Xapi-J as a central
reference. I would suggest that those who have pages publicising XML resources 
should include this.  John - is there now a definitive URL that should be used?


> is done I think Xapi-J should be reved to be a direct subset of the
> DOM's object model using the DOM's object model and method signatures
> exactly. That is the only part I see where there is overlap and it would
> be a shame to have two very similar but different object models of an
> XML document. The original goal of Xapi-J was to come up with a unified
> model/api for Java developers who are using/writing XML processors. To
> not reflect the work of the DOM WG would defeat the whole idea.

It seems clear to me that there will continue to be revisions to many parts
of XML (we do not yet have a definitive version). So revision of Xapi-J to 
be consistent with DOM will be one of several such adjustments or extensions.

I hope that other reference documents will come out of public debate on XML-DEV.
I think there are going to be a large number of problems which are not defined
by the spec and which are not felt appropriate for discussion by the WG (the 
formal W3C body) or the SIG (now not public).   From time to time it is 
suggested that 'this is an implementation problem, - perhaps XML-DEV would be 
appropriate?'  

Some of us are concerned that uncoordinated (though well-meant) implementation
of XML applications and tools will create a range of inconsistent approaches.
At present XML-DEV is the only forum for discussing these and I think we have
a critical role here. Obviously any contributions are voluntary, not part of the
W3C process, but if we continue to come up with well-thought out documents or 
proposals they should have an important role. 

Some areas where I think guidance for implementers is critically needed NOW 
(in rough priority) are:
	- whitespace processing
	- error processing
	- treatment of defaults and inheritance
	- interpretation of XML-LINK constructs
 
Volunteers? Should we adopt any sort of informal process?  Thoughts? :-) Perhaps
those who are able to be present at XML-DEV day might wish to discuss this?

	P.


-- 
Peter Murray-Rust, domestic net connection
Virtual School of Molecular Sciences
http://www.vsms.nottingham.ac.uk/

xml-dev: A list for W3C XML Developers
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To unsubscribe, send to majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
List coordinator, Henry Rzepa (rzepa at ic.ac.uk)