OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

ebxml-core message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Subject: Re: autogeneration of UML to XML

Thank you Brian for the information, but the problem with the document is
that it is designed for the very specific case of mapping between UML and
XML (SOX) and it order to be able to do something UML has been heavily
adapted through the use of stereotypes and tagged values. Also one sees that
the stereotypes have been created to cater for XML. In other words virtually
you model your XML schema instead of modelling in a neutral form.
Consequently you are no longer really using UML per se but a tailored
language to model XML schemas. Let me just take an example. A stereotype
called "enumeration" has been defined for the purpose of defining codesets
(enumerations being generally used within UML to give a list of values for
an attribute type!!). In the example there is a class with the stereotype
"enumeration" called "CountryCode". Whats curious is that this class has
attributes which are none other than the codes!! (in the example : USA, ENG,
GER, FR). I find a very curious mix here between classes and objects !!! But
then maybe its the stereotype which clears all that up :-)))

With the way UML has been specialised and doctored, I think I could do a
pretty good job of defining a similar UML to map to EDIFACT or even X12. But
I don't really think thats the point.

So for the moment from my point of view all that the Rational document has
managed to demonstrate is that our core components in UML cannot be used to
autogenerate an XML schema. Also I'd submit that we're a long way off from a
syntax neutral solution.



----- Original Message -----
From: "Hayes, Brian" <Brian.Hayes@Commerceone.com>
To: "Mike Conroy" <michael.conroy@wanadoo.fr>; "ebXML Core"
Sent: Saturday, April 07, 2001 12:43 AM
Subject: RE: autogeneration of UML to XML

> See UML for XML Schema Mapping Specification
> http://www.rational.com/uml/resources/documentation/index.jsp
> <http://www.rational.com/uml/resources/documentation/index.jsp>
> -----Original Message-----
> From: Mike Conroy [mailto:michael.conroy@wanadoo.fr]
> Sent: Friday, April 06, 2001 2:19 PM
> To: ebXML Core
> Subject: autogeneration of UML to XML
> Ever since William Kammerer said : "Heck, I haven't even yet seen anyone
> take a UML data model and generate schemas from it -  and that should be a
> lot simpler" theres been a flood of e-mails from people such as Duane
> Nickull of XML GLobal Technologies, Arofan Gregory of Commerceone, James
> Whittle of  e-centre, Philip goatly of Bolero, and Kris KETELS of Swift,
> have all jumped into the thread to tell us that they currently have the
> capability to "autogenerating from UML" XML schemas.
> Well I for one am most impressed. However, I do have a grain of doubting
> Thomas, in my vains. I made a UML class model of the core components,
> I am led to believe is a faithful rendering of the Excel spreadsheet.
> Just to give me, and perhaps many others who are listening in, an idea of
> how the autogeneration from UML works, would someone like to demonstrate
> their capabilities on the enclosed core components model ? It might help
> settle a lot of the dust that has been raised on the subject.
> Regards
> Mike

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]

Search: Match: Sort by:
Words: | Help

Powered by eList eXpress LLC