Re: UML

[Marcus G. Daniels]

>>>>> "PT" == Pietro Terna <address@hidden> writes:

PT>     In my opinion UML is a good tool to describe models as
PT> software structures (in an enterprise environment, for example);
PT> I'm not sure that this kind of formalism could be completely
PT> usefule where we have to develop, with our simulation models) not
PT> a 'world' but 'a metaphor of a world'.

I think that could work. I once developed a UML project in Rational Rose.

   ftp://ftp.swarm.org/pub/swarm/src/users-contrib/anarchy/anl-0.0.4.tar.gz

See xml/swarm.mdl for a description of heatbugs.  MDL is Rational's 
ad-hoc file-format for UML. 

On the plus side, UML parameterized classes work fine for the
+createBegin:-createEnd kind of instantiation.

But one problem I have with UML is that it has a bunch of notation that I
think would never be used by Swarm users, in practice.  Another
problem is all that notation maps to a really complicated XML format
called XMI that (last time I checked a few months ago), isn't very
well-documented and for which there isn't a good freely-available
support library.


                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.