[sc34wg3] what's the most basic issue?

[Steve Pepper]

I don't know if this is the most basic issue, but since SRN has asked
us to state our positions, here goes (with no holds barred):

At 20:44 28.01.2003 +0100, Lars Marius Garshol wrote:

>* Steven R. Newcomb
>|
>| [what *is* the RM?]
>|
>| I'm willing to start there, if everyone else is.  It begs the
>| question: what are the various positions on this issue?  It would be
>| good to understand what they are, exactly.
>
>* Lars Marius Garshol
>
>In the spirit (I hope) of what you started I'll kick off with my own
>view: the RM is whatever SRN thinks it is. You started it, and you are
>very clearly the driving force behind the RM work. So as far as I can
>tell, if you say the RM is X that's what it is. So what I would like
>to hear is what *you* think it is.
>
>Now, I suppose what you meant is just as much what each of us thinks
>it *should* be. "Serious and open-minded introspection," you said.
>Well, personally I am not very interested in the RM. It has been
>proposed that it can be used as a kind of conceptual model or
>reference model for topic maps, and I find that mildly appealing. I
>don't mind if people go off and do that.

This is also where I believe the value of the RM lies in the short
term: As a kind of yard stick against which we can evaluate certain
parts of the SAM and possibly reveal some inconsistencies.

I've been part of the development of this standard long enough that
I remember how we all thought about it back around 1998 and even earlier.
I've seen how everyone's understanding (my own included) has developed
and deepened in the time that has elapsed since:

For example, I saw the emerging realisation that occurrences were
really just specialised associations; followed by first the denial
and then the acceptance (at least by many) that names could be regarded
as specialized occurrences. (It's strange to look back and realise how
little we understood even as 13250 went to press :-)

If we'd realised things like that in time, we might not have ended
up with names that didn't have types (but did have variants), and
occurrences that didn't have variants but did have types.

In just the same way it seems to me that the RM *may* help us
uncover inconsistencies in the SAM - and fix them before we go to
publication.

But that, to me, is the *only* value that the RM has in the short
term.

In the longer term, I think the RM *may* help lead us beyond topic
maps and take us to the next level (which may be the level at which
not a unification but a *synthesis* of topic maps, RDF and other things
takes place). I say that because I do recognise that there are some
valuable insights in the RM as it is today.

However, I think the RM is very far from being "cooked". Very far
indeed. If it really is the "assembly language" of topic maps, then
I would expect a compactness and elegance that transcends even the
TAO model, and that I do not see at the moment. In fact, the RM seems
to become less compact and less elegant at each iteration.

It may be that the formalism is what is primarily at fault. Perhaps
hypergraphs should be taken as the starting point, and maybe that will
render all those clauses and tables unnecessary. I don't know.

What I do know is that the RM as it stands

(1) contains enough interesting insights that it *may* help clarify
certain aspects of the SAM (even though those insights may not yet
have found their ideal expression),

(2) is quite obviously still a long way from prime time, and

(3) has yet to be justified in terms of real user requirements.

Finally, I am very sad that all the energy that has gone into the RM
wasn't channelled into marketing and promoting topic maps as defined
in 13250. If we'd pulled together and done that, I think the market
acceptance of topic maps would be much greater than it is today.

Steve
--
Steve Pepper, Chief Executive Officer <pepper@ontopia.net>
Convenor, ISO/IEC JTC1/SC34/WG3  Editor, XTM (XML Topic Maps)
Ontopia AS, Waldemar Thranes gt. 98, N-0175 Oslo, Norway.
http://www.ontopia.net/ phone: +47-23233080 GSM: +47-90827246