Re: [topicmaps-comment] uml and topic maps

[Lars Marius Garshol <larsga@garshol.priv.no>]

* W. Eliot Kimber
| 
| There are two different issues here:
| 
| 1. Modeling the abstract data model for topic maps using UML. 
| 2. Modeling topic map instances using UML.
 
For the record, I have tried both 1 and 2 using EXPRESS, which is a
very powerful data modeling language (part of the STEP standard). It
worked very well for 1. (In fact, the infoset-based model proposal is
derived from that original EXPRESS model.)

For 2, however, it didn't really work very well. Both Steve Pepper and
tried this, and none of us managed to come up with anything that we
could claim was very useful.

I think a graphic modelling notation for topic maps could really make
a noticeable difference for the adoption of the standard. It would
also be something that would be highly useful in its own right. I'm
not the right person to work on it, though. However, I think all we
need is for someone to step forward with a reasonable proposal. Once
that is done we can try and develop it into something that can be
standardized.

--Lars M.