Dear Henson, (01)
thanks for these further details about MBSE and SysML. (02)
As you wrote this to Peter, Leo and me, and after Peter had included you
into the discussion that I had started on the OntoIOp side, may I assume
that it is OK to share it with the whole OntoIOp community? (I am Ccing
them in this reply, quoting the relevant parts from your email below.) (03)
This is because I am neither a UML nor a SysML expert, but just happened
to be the connection point between the MCS special issue and OntoIOp. (04)
Cheers, (05)
Christoph (06)
2013-10-06 14:01 henson:
> As
> you now know I have been involved with various ontolog efforts. Also in
> theory I lead the INCOSE MBSE ontology action team which you can find at
> http://www.omgwiki.org/MBSE/doku.php?id=mbse:ontology. Have a look, I
> could use a lot of help.
>
> My work on formal semantics for SysML is very much ongoing, in progress.
> I can fill you in on that. The quick version is that I realized several
> years ago that I could map a large class of SysML models as axiom sets
> in a logic formalism based on the map and type (object) constructions of
> topos theory (ABD in the paper). The topos version simply axiomatizes
> set theory constructions in an algebraic way. It also has advantages for
> dynamics. This formalism can in turn be loaded within the right kind of
> inference engine and used to do some of the kinds of examples that I
> think engineers would understand and like.
>
> I am familiar with the work on giving UML class diagrams a formal
> semantics, e.g., work by Berardi, D., Calvanese, D., and De Giacomoa,
> G., 2005. While this is very good work, it suffers from the same
> Description Logic insufficiencies that have led Ian Horrocks and friends
> to first do the description graph extension and now the Logic
> Programming Description Graph extension. In my opinion none of these
> extensions are sufficient. That statement requires discussion. Brief
> form is that none of the DL formalisms really represent the diagrams in
> the languages well. However, they can be represented easily within the
> topos framework, which uses a two sorted signature for maps (arrows) and
> types (nodes). By using two sorted FOL approach one can express
> statements that are higher order in other approaches.
>
> You may find all of this a bit confusing, but ...
>
> Best Regards
> Henson (07)
--
Christoph Lange, School of Computer Science, University of Birmingham
http://cs.bham.ac.uk/~langec/, Skype duke4701 (08)
→ Mathematics in Computer Science Special Issue on “Enabling Domain
Experts to use Formalised Reasoning”; submission until 31 October.
http://cs.bham.ac.uk/research/projects/formare/pubs/mcs-doform/ (09)
_________________________________________________________________
To Post: mailto:ontoiop-forum@xxxxxxxxxxxxxxxx
Message Archives: http://ontolog.cim3.net/forum/ontoiop-forum/
Config/Unsubscribe: http://ontolog.cim3.net/mailman/listinfo/ontoiop-forum/
Community Files (open): http://interop.cim3.net/file/pub/OntoIOp/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntoIOp (010)
|