I think lot has been said in this thread about this subject. Here's
something more from me ...

UML can depict different kind of concepts: like class diagrams,
sequence diagrams, use case diagrams, activity diagrams and some more.
Some of these concepts *can* be (easily) semantically equated to XSDL
terminology, like for example, class diagrams or perhaps sequence
diagrams. But to my opinion, things like UML use cases have no
equivalence to any of the features of XSDL. So I think, it's not right
to say, "mapping UML to XSDL" without any qualification.

I think, mapping ER diagrams to XSDL is more logical, as both most of
the time reflect data in problem domain. And as I wrote above, we can
map UML classes easily to XSDL features.

I think there is almost no limit to how we use some technology, like
say UML. Mike Kay in his XSLT book has used UML to help readers
understand various kind of nodes and their relationships, in XPath
data model. I in fact liked that example ...

PS: thanks for the links. I'll try to find time to read the literature here ...

On Thu, Sep 4, 2008 at 9:52 PM, Andrew S Halper <ashalper@u...> wrote:
>
> UML is more expressive than most ER diagram notations, thus easier to map
> to XSDL. The down-side is that UML is also more ambiguous and, like most
> ER notations, not rigorously defined mathematically. Every UML to XSDL
> mapping I've seen is different.
>
> Some pages to have a look at:
>
> http://www.agiledata.org/essays/umlDataModelingProfile.html
> http://www.xml.com/pub/a/2002/08/07/wxs_uml.html
> http://www.tdan.com/view-articles/5046/
> http://www.xml.com/pub/a/2002/11/13/normalizing.html
>
> (also, apologies if I've spoken out-of-turn)
>
> Andy Halper
> Computer Specialist
> National Water Information System
> USGS



-- 
Regards,
Mukul Gandhi

From togix@l... Tue Sep 16 04:38:17 2008
Received: from farnsworth.w3.org ([128.30.52.43] helo=wiggum.