Re: Object Role Modelling (ORM) or UML or ?? for designing Schemas

[Dave Carlson <dcarlson@ontogenics.com>]

Not to worry Robin, my ears were not deaf :-)

I find this subject of conceptual modeling for XML to be better received
outside of the traditional XML circles.  However, as XML infiltrates all
aspects of system design, this is not very limiting.  I did a presentation
at
the OOPSLA conference in October (which is the main object technology
conference) on the subject of modeling XML schemas with UML.  It was very
well received.  Although, in that environment, it was like selling candy on
Valentine's Day...

I began prototyping the generation of DTDs and (at that time) XML-DATA from
UML about 2.5 years ago, which was over a year before Matt Fuchs
(Rational's) paper on the subject.  This was also the subject of the XMI
specs from the OMG, which began as draft in the summer of 1998.

Dave

----- Original Message -----
From: "Robin Cover" <robin@isogen.com>

> A couple weeks ago (see archives for late January) there was an XML-DEV
> thread on conceptual modeling and mapping UML <--> XML schemas.
> I wrote in that connection,
>
> "...  My guess is that the syntax-centric
> stranglehold will not be broken until there's a conscious focus
> on conceptual modeling, accompanied by an unspoken agreement
> that XML schemas and other markup syntaxes can readily be
> generated from conceptual model notations..."
>
> but the comment appeared to have fallen on deaf ears.  Most
> people are apparently content to design in an environment that
> does not test/enforce constraint-based relational semantics
> at the object level; using visual-graphical modeling software
> and syntax-centric modeling formalisms is "good enough, even
> if not ideal..."
>