Re: On subsetting XML...

On Thursday, January 16, 2003, 4:30:13 PM, Norm wrote:


NW> In any event, I feel very strongly that the subsetting issue and the
NW> other issues (like xml:id) *must* be dealt with separately. It would
NW> be a procedural disaster, IMHO, to combine them.

I agree that the are separate issues but clearly they are related and
influence each other. A "subset XML" that removes all the existing
mechanism to declare IDs has restricted the options down to

a) Use XML Schema if you want IDs, otherwise there are none
b) A new, in-band mechanism will provide IDs.

NW> I am very, very nearly convinced that doing nothing is the wrong
NW> answer


Good. Stay with that thought ;-)

NW> on the subject of xml:id, but I am absolutely certain that it
NW> applies to XML in general and not only some possible subset.

Yes it clearly applies to XML in general - its very important that the
subset XML is a true subset and no a forked, subset-plus.

But its also true that the need for a ID solution becomes more acute
when a subsetted XML (that has no internal or external DTD subset) is
considered.



-- 
 Chris                            mailto:chris@w3.org
 Briefly chiming in on a break at a face-to-face meeting

Received on Thursday, 16 January 2003 23:34:54 UTC