Re: Help...Current status of formalization of UML



Re: Help...Current status of formalization of UML

From: Pierre-Yves Schobbens ^lt;pys@info.fundp.ac.be>
Date: Sun 30 Apr 2006 - 08:13:43 BST
Message-Id: <75B3C65D-BED6-47AC-B65F-18C17047C8D9@info.fundp.ac.be>
Dear Irum,

The response of Greg is very instructive, and actually I confirm what  
he says

Le 30-avr.-06 à 05:00, Greg O'Keefe a écrit :

> Actually, many people using very
> different approaches are all in this project, and its not clear to me
> how all this work is to be reconciled.

by disagreeing with his comment:

> I have
> recently begun to think that the most exciting work is that using  
> graph
> transformations.  Thus ordinary UML users
> would be able to understand the formal semantics, and this would  
> lead to
> general agreement, between people and tools, about what the diagrams
> mean.

1- Graph transformation is a rather complex subject studied since 30  
years (already by Knuth, Courcelle and other now retired people)
   and we are still far from understanding it.

2- A transformation, by definition, cannot be a semantics.
    Transformations are by definition executable, while semantics  
concepts for UML are not decidable.
    Thus this is impossible to define semantics with transformations  
alone: we need more powerful tools.

In summary: Transformations are very useful tools, but are not  
semantics.
Actually, we need lot of different transformations, each for a  
different purpose: transformations to make the model more analyzable,
to make it executable, to expand abbreviations, etc. These  
transformations must be justified on basis of the semantics.

Best regards,
   PY
Received on Sun Apr 30 08:14:31 2006