Re: Missing formal definition of OclMessageExp in OCL 2.0?

Date view Thread view Subject view Author view Attachment view

From: Heinrich Hussmann (
Date: Wed 29 Jan 2003 - 11:53:12 GMT

Hi Gerrit and Jos

Gerrit Renker wrote:
> Hi,
> I had the same problem -- there is a contradiction in the current OCL
> document:
> --in sec. 1.5.3 it reads:
> ``The equivalance of the two semantic descriptions has not been
> formally established. Wherever they are conflicting,
> the description in section A ( Semantics ) is normative.''
> --on page  1-6 (sec. 1.7) it reads:
>   ``Appendix A ( Semantics ) describes the underlying
>     semantics of OCL using a mathematical formalism.
>    This appendix, however is not normative, ...''
> Gerrit

Thanks for pointing out this inconsistency. In fact, it was the original
intention of the authoring team to make the mathematical semantics the
normative one, but after some discussion in the OMG A&D task force it
was recommended to make the UML-based semantics the normative one. So
se. 1.5.3 needs to be updated, and 1.7 is right.

The formal semantics of OclExpression is currently missing, and the
authors are aware of this. The reason is that iut is not exactly
straightforward what the best semantical abstraction for this concepts
is. Constructive contributions are welcome.

The speaker of the authoring team is Anders Ivner at Borland
(, so any suggestions should be directed to him.

Heinrich Hussmann
Dresden University of Technology

>  --- Jos Fries <> wrote: > Hello,
> >
> > With much interest I read the OCL 2.0 submission, especially the
> > formal definition of the semantics of a subset of UML and of OCL. For
> > application of UML/OCL in the field I'm working in (railway control
> > and safety) such a formal definition will be a prerequisite for
> > acceptance of the method.
> >
> > Paragraph 1.5.3 of the submission states that an equivalent
> > mathematical description is given in appendix A (including semantics
> > for OclMessage). In that appendix however I cannot find a definition
> > for OclMessageExp. Is my observation correct? Is it yet to be added
> > to the definition?
> >
> > Jos Fries
> >
> >
> =====
> --
> Gerrit Renker                     Research Assistant
> Constraints Group                 Computing Technologies Centre
> The Robert Gordon University      Aberdeen AB25 1HG
> __________________________________________________
> Do You Yahoo!?
> Everything you'll ever need on one web page
> from News and Sport to Email and Music Charts
> To remove yourself from this list please mail
> with a message containing the word "unsubscribe".

Date view Thread view Subject view Author view Attachment view