[sc34wg3] Individual contribution on the U.S. N.B. position o n the progress ion of Topic Map standards

[Mason, James David (MXM)]

Thanks, Kal.

I think the RM is something that has evolved -- especially in the eyes of
its creators. And I think its evolution has to do, in part, with their
having come to understand that they left too many questions unanswered in
the original 13250 -- and the second edition did nothing to fix that.

I suspect that one reason things are lacking in the original 13250 is that
the authors were fresh from revising HyTime and took it as a given. As we
get further from HyTime, we need to backfill so that 13250 can stand on its
own.

I can't speak for the authors of the RM, but I wouldn't expect a general
theory of knowledge representation in 13250. That's too big for the scope of
SC34. What I'm looking for is just enough basis to go on to the application
oriented portions, starting with XTM, which got in by way of a shotgun
marriage in which I willingly complied. And at the same time, I want to
relieve the other parts of the family from having to mess with a formal
basis.

I watched the ODA people turn themselves inside out with a "formal
definition" of ODA -- indeed I egged them on because it kept them busy and
out of our way while we quietly kept doing SGML things. We don't need that
in SC34. But what I see in the RM, if people would really look at it and
pare it down, is worlds away from that thing which turned into an existence
proof of ODA -- a useless project if there ever was one. (So in that sense,
don't take my axioms/theorems stuff too literally.)

What I see in WG3 is a bunch of really sharp application developers who are,
furthermore, quite articulate. That's good. It means that things like TMQL
will be grounded in stuff that works. I've enjoyed the recent little thread
about whitespace in CXTM because it shows people are really working on code.

But at the same time, I want to see "just enough" in the RM to give support
to the developers who are going hot and heavy on the stuff that's close to
application. There are a handfull of people who want to get that part right,
too. They need some help, and it's not really right to dismiss their
contributions with "oh, we know how to do that in a query language" (not
that that's your position).

> -----Original Message-----
> From: Kal Ahmed [mailto:kal@techquila.com] 
> Sent: Tuesday, March 30, 2004 2:44 PM
> To: sc34wg3@isotopicmaps.org
> Cc: 'Lynne A. Price'; 'USA HOD: Patrick Durusau'; 'USA/W3C 
> XML: Jon Bosak'; 'Zarella Rendon'; 'SC34 Secretariat 
> (jtc1sc34@scc.ca)'; Mason, James David (MXM)
> Subject: Re: [sc34wg3] Individual contribution on the U.S. 
> N.B. position on the progress ion of Topic Map standards
> 
> 
> Jim,
> 
> Thank you for this - I'm sure that it will give us all a lot 
> to think about in the Amsterdam meetings (assuming that we 
> can hear ourselves think over the sound and the fury of the 
> standardisation process ;-)
> 
> I feel that I have to make a response now though to put my 
> personal position on this issue - I'm not sure that I will be 
> present when this is discussed at the meetings, although I 
> will be attending from Friday afternoon.
> 
> I believe that the four parts of ISO 13250 in progress at the 
> moment address all four of your points, but as you noted, 
> there is currently no way for an application to specify 
> merging rules declartively.
> 
> However, IMHO, that this is a showstopper for ISO 13250. The 
> original version of the standard, and its second edition 
> lacked declarative rules for merging and yet the take-up of 
> the standard has progressed and people are managing to create 
> all kinds of systems and to interchange information successfully.
> 
> Topic maps are one form of information organisation. I 
> believe that there *is* a more generalised theory of 
> information organisation which can be developed to encompass 
> topic maps, RDF, thesaurii, classification systems and all 
> kinds of information organisation systems. I don't think that 
> Topic Maps (as stated in ISO 13250-2000 and again in the second
> edition) are that generalised theory.
> 
> The work on the RM is valuable work which will inform not 
> only topic maps but also many other apporaches to organising 
> information and communicating assertions between machines 
> (and ultimately between human beings). It would be a valuable 
> piece of work to be done either inside ISO or in some other 
> forum. I don't believe that it is necessary for the RM to be 
> completed or for the work on ISO 13250 to continue, nor do I 
> believe that it necessarily need hold up the initial stages 
> of work on TMCL or TMQL (although if the RM progresses 
> quickly, it may be that it informs the work on those standards).
> 
> Cheers,
> 
> Kal 
> -- 
> Kal Ahmed, Techquila
> Standards-based Information Management
> e: kal@techquila.com
> w: www.techquila.com
> p: +44 7968 529531
>