Luis J. Martinez
27 Apr 2003 20:18:42 -0400
* Jan Algermissen
> Clearly, there are portions of both, N0393 and N0396 that do make sense
> to constrain applications (e.g. merging behaviour, value equality...),
> but what is the purpose of the conceptual data model? We need it, sure,
> but what for?
I think the purpose for a model for Topic Maps is to describe what is
the abstract structure of Topic Maps, a description to all components
of Topic Maps, and to describe the functional properties of Topic
The main point is to clearly describe what valuable functionalities
Topic Maps provide. The implementation specification is not necessary
under the standard process. Implementation details should be completely
left up to developers. That is way it is so important to describe
clearly the abstract structure of Topic Maps and it semantic
properties in general terms.
For example. The abstract data structure binary search tree can be
implemented as a double link list instead of a tree structure as long
as it follows the prescribed semantics for a binary search tree. My
point being that as long as the operation results are correct, it does
not matter what the implementation looks like. Of course, it is useful
to have a standard interface, but that is just a convenience detail.
I think that the SAM is distracted with specifying implementation
details. If the SAM purpose is to derive interfaces like XTM, HyTM,
TMAPI, then fine, no need to conform to the SAM. Just those
interfaces. But the implementations behind the interfaces have to
conform to the semantics of Topic Maps. I think those Topic Maps
semantics should be specify independently of interfaces. I think the
SAM is try to accomplish two goals simultaneously.
I think the Model should answer "What is Topic Maps and what value it
provides?" not "How to represent Topic Maps with a particular