Roger Costello writes:

> For some data requirements there is only one XML validation 
> language that has the needed capability, so the selection of 
> language is clear.

Actually, I'd say it a bit more carefully.  If by "data requirements" you 
mean constraints, I think that's only part of the story.  Sometimes the 
most urgent priority is to express the constraints you really want 
enforced, using whatever technology will do the job.  Sometimes, however, 
it's more important to use a language that is widely deployed, or at least 
one that will be in the places you care about, and that will work with the 
tools you and those you send your XML to need to use.  So, sometimes it's 
a better tradeoff to use a language that doesn't quite capture all the 
constraints you would prefer to have enforced, but to get say 80% of them 
in a form that will be widely understood and work with the tools people 
want to use.  So, I'd be tempted to restate that as "For some data and 
deployment requirements, there is only one XML validation language that 
has the needed capability, so the selection of language is clear."

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------