[sc34wg3] N0429: Topic Maps Reference Model: Requirements

Steve Pepper sc34wg3@isotopicmaps.org
Fri, 1 Aug 2003 09:26:13 +0200


Thanks for your quick response to my comments. I don't
have time to reply in as much depth as I would like before
the WG3 meetings get under way, but maybe a few more short
comments would help the discussion.

| > The thing I (at any rate) was looking for was something more
| > along the lines of what is called "Purpose and justification" in
| > New Work Item proposals.
| Will have to look at the requirements (sorry, no pun intended) for
| "Purpose and Justification" statements. Perhaps I can find ones in other
| New Work Item proposals to use as a guide.

Rather than you go looking for the meta-requirements let me simply
say that my view of what "Purpose and Justification" is about is:
Explaining to the world outside SC34 (in particular the National
Bodies that vote for or against a NWI), *why* we are proposing a
certain work item (in this case, the RM) *in terms they understand*.

For me, the latter means that the explanation has to be couched
in terms of one or more real-world business problems. At present
N429 is still too abstract in this regard. (Elaborating the IRS
example would probably help at lot, especially if you could provide
a couple of other examples as well.)

| Not meaning to be difficult but the SAM/TAO are not presently part of
| ISO 13250. XTM syntax has been added but not the XTM specification.

I was not suggesting that either the SAM, the TAO, or the prose of
the XTM specification are parts of the approved 13250 as of today.
Obviously they are not. I was simply using SAM/TAO/XTM as a label for
one interpretation or view of 13250 - the one which I believe is both
the dominant and the more historically accurate one (which is not to
say that it is the view we necessarily want to end up with).

| In brief I was trying to say, probably poorly, that the RM should make
| explicit the underpinning ideas of topic maps, which can only be
| inferred from the current version of ISO 13250. Since it defined an
| interchange syntax, it did not ever make the reasoning for various
| aspects of the interchange syntax clear.

Making explicit the underpinning ideas is what the whole restatement
of 13250 is about - not just the RM. There is no doubt that we need
such a restatement; what is not (yet) clear to everyone is that we
need the RM in order to do that. Hence my insistence that we need
to justify the need for a RM before discussing how it should work
or what it should look like.

| I thought the document made it clear that there was no intent to go
| beyond ISO 13250, which is as was originally approved, with the addition
| of the XTM syntax as an appendix.
| I really don't think, and here is where a conference call might (or
| might not) help, that either Newcomb or myself are proposing anything
| that could be considered "beyond 13250." It would be helpful (realizing
| this is an imposition on your time when other matters are probably
| pressing) if you could say how any particular part of the requirements
| document shows an effort to go "beyond 13250."
| Don't limit yourself to the explicit language in the document as I don't
| want to play word games but honestly would like to know in what way you
| think the RM (or the requirements document) proposes to go "beyond 13250."

I don't think I can answer that question to either your satisfaction
or mine in the short time I have available right now before the start
of the meeting. I readily admit that N429 has far less new terminology
and concepts than previous iterations of the PM4TM/RM/TMMM, which in
my opinion is a BIG step in the right direction. However, I still
think there is a very large residue of thinking which says that there
are some "things" out there, which legitimately might be called "topic
maps", which cannot be expressed in either HyTM or XTM syntax, and that
do not conform to the model described in the draft of 13250-2. If that
is the case, and the RM is designed to cater for those "things", then
the RM, in my opinion, does indeed go beyond the current 13250.

If such "things" do exist, I need to know what they are, *why* they
cannot be expressed in terms of 13250-2 and -3, and why they still might
legitimately be called "topic maps". Once I am convinced of their
existence and understand their nature, the first thing I will want to
do is look at 13250-2 and -3 to see whether they can be accomodated,
without having to introduce a whole new "reference model".

I hope this clarifies the position a little.

Have a good meeting everyone!


Steve Pepper <pepper@ontopia.net>
Chief Executive Officer, Ontopia
Convenor, ISO/IEC JTC 1/SC 34/WG 3
Editor, XTM (XML Topic Maps 1.0)