RE: [ubl] Global vs. Local -- Gunther's Recommendation

[Dan Vint <dvint@acord.org>]

I'll jump in here as well and agree with Eve and Mark. I'm new to this 
process, but it seems to me we should be making "proper" XML. Technology 
changes and next week UML might not be the way to model things and FOO is 
the latest methodology, would we change the design/process to then be more 
conformant with that methodology? Why shouldn't we choose ER diagrams as 
the better modeling environment and make our XML look like tables?

I have some trouble with the idea of trying to force XML into tools/methods 
that aren't designed to be used with it or for it. There are things in 
UML/OO that I can do that I can't in XML, and likewise things and methods 
in XML that don't map well to UML/OO.

Until someone comes up with a truly universal modeling language that can 
handle OO, relational databases, XML, etc. without "forcing the process or 
method" I don't see how you can choose something that doesn't work 
completely with the target implementation and say it should conform to the 
other. Maybe OO languages will morph to something that better fits the XML 
world in the near future.

This is another opinion from a non--OO comfortable developer.

..dan


----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>