Re: UML 2.0 Issues



Date view Thread view Subject view Author view Attachment view

From: Andy Evans (andye@cs.york.ac.uk)
Date: Fri 14 Feb 2003 - 11:18:52 GMT


Gerrit,

There are currently two main UML proposals on the table: U2P and 2U. The 2U
proposal has incorporated the best of the MMF and MML work and so supports a
meta-modelled semantics for UML and OCL + Catalysis like package extension
and templates. This submission therefore supersedes the MMF/MML work.

The OCL submission for UML 2.0 has wide agreement now and the semantics of
the OCL component in 2U are pretty compliant to that of the OCL submission,
so you can ignore any differences there. The OCL you see in the OCL 2.0
submission is thus the only contender.

I believe that your questions regarding delay are ones being generally asked
by the user community. As a result there seems to be growing political
pressure to accept the larger consortium's (U2P) proposal now and address
any problems during finalization. Whilst some of the things we want are in
there (e.g. package extension and templates to a certain degree), I
personally believe that many compromises have been made and many problems
remain un-addressed - the question will be whether these compromises are a
price worth paying in the long run.

Andy

On 14/2/03 9:02 am, "Gerrit Renker" <gerrit_renker@yahoo.co.uk> wrote:

> Hi,
> 
> after reading a lot of proposals and perusing several UML 2.0
> related websites I find myself in a state of more questions than
> initially. I hope that someone on the list can help clarify some UML
> 2.0 issues -- much in the interest of others who are likely to also be
> lost in a flood of proposals.
> 
> 1/ STANDARDIZATION ISSUES
> I think that the ideas to MMF and the MML as proposed may be
> good starting points -- but which of the proposals is likely
> to be adopted by the OMG. What will we be talking -- MMF,
> U2P,  2U. The point is there is a flood of proposals out
> there but I think I am not wrong in assuming that the last
> say will be the OMG. So at the end of the day, are there
> any indications yet which of the many proposals will make
> it to the OMG.
> 
> 2/ EXTENSION MECHANISM
> Further to an email I sent earlier on this list, I am
> probably not the only one having problems with the UML 1.4
> extension mechanism (there was a paper at UML-99 about
> that). Again, are there any indications -- will we finally
> have the long-awaited Catalysis style package templates?
> 
> 3/ WHEN?
> In an earlier email on this list I heard someone predict
> 2004 for the release of UML 2.0. It is late anyway (the
> CACM article by Kobryn predicted 2001 as release date).
> Does anyone have nearer informations?
> 
> 4/ DOMAIN TASK FORCE
> We are working on AI extensions for UML. We would be
> interested in developing profiles and model libraries. Are
> there any interest groups, birds of a feather or task force
> groups for this area yet?
> 
> 5/ FINALLY
> On a personal note, I find the current state truly Babelian,
> lots of upcoming great ideas but the (OMG) future entirely
> opaque. An example is the semantics in OCL -- people have
> come up at least with 3 different OCL semantics. Richters'
> is adopted in the appendix of the current OCL spec, but the
> normative one is still the UML 1.4 semantics, based on
> meta-models and natural language.
> 
> It would definetely help if someone could say a few clarifying words
> here.
> 
> Gerrit Renker
> 
> 
> __________________________________________________
> Do You Yahoo!?
> Everything you'll ever need on one web page
> from News and Sport to Email and Music Charts
> http://uk.my.yahoo.com
> 
> 
> 
> To remove yourself from this list please mail puml-list-request@cs.york.ac.uk
> with a message containing the word "unsubscribe".
> 
> 

Date view Thread view Subject view Author view Attachment view