[sc34wg3] Analysis of TMRM Use Cases
Mon, 12 Apr 2004 16:38:18 +0100
On Mon, 2004-04-12 at 14:33, Mason, James David (MXM) wrote:
> See below, snipped from Steve P's comments on Robert's message.
> Jim Mason
> > Again, there is a certain intellectual satisfaction in this,
> > but it begs further important questions:
> > * What other "instantiations" already exist or might exist?
> I think I've mentioned TMs in databases in another message, just for a
> starter. From the end-user's prespective, TMs should be a black box: if it
> acts like a TM, it is a TM. We may think a real TM engine offers better
> performance, but if you can fake it (even a static set of generated HTML
> pages, for a collection that't static), why not call it one?
I wouldn't want to call a set of HTML pages a TM as it lacks merging and
update features - I wouldn't mind it being called a rendition of a topic
Databases are another matter, I imagine that with some clever SQL stored
procedures you could probably make a database behave like a topic map,
but again I would draw the line at saying that all databases are topic
> > * Do we want to call those "instantiations" Topic Maps?
> Yes. That was part of the intent of the original 13250, which is what is
> still in force.
> > * If so, to what extent does it serve or damage the interests
> > of Topic Maps users for there to be multiple models, all of
> > which can legitimately be called Topic Maps?
> I believe it helps us immensely in that we can point to other things (maybe
> even in RDF, horror of horrors), and say, this, too, is a TM. This helps us
> make our case for TMs being a widespread phenomenon, not just a SC34 niche.
On the other hand, doing this leads to a dilution of the term "Topic
Maps" and potential confusion in the user community. Most people that I
deal with in topic maps equate topic maps with topics, associations and
occurrences and with a mental model that looks a lot like TMDM. I think
that saying that RDF is topic maps, that databases are topic maps and
that anything can be a topic map is a different proposition - it is a
different proposition entirely from what I read in ISO 13250.
> > * To what extent should multiple "instantiations" of the
> > generic model be interoperable?
> Most likely we want import to things we think of now as TMs.. The RM,
> combined with a DM, might show us how to transform the odd representations
> to something we're more comfortable with (i.e., XTM). It might be that the
> transformations will not be fully reversable, it might be that some must be
> assembled on a case-by-case basis, but it would help to be able to say we
> have an understanding of how to create them, based on the RM.
I think that adding import from other data models to the topic maps
standard is just that, an addition - and that while it is no doubt a
worthwhile thing to do, I just don't think it is a fundamental part of
the ISO 13250 standard.
Kal Ahmed, Techquila
Standards-based Information Management
p: +44 7968 529531