[topicmapmail] Announcement of XML Schema for ISO 13250 Topic Maps

[Murray Altheim]

Martin Bryan wrote:
> 
> Murray
> 
> > I'm not sure where you got the idea that there was no way to express
> facets
> > in XTM syntax.
> 
> From one of the earlier messages on this list stating that XTM did not
> support Facets.

Well, you are correct that we didn't implement explicit facet syntax,
but my understanding is that the current syntax suffices.

> > The XTM 1.0 syntax does not have explicit facet syntax, but
> > this doesn't stop anyone from expressing facet relationships.
> 
> The question is how they generalize them to make them interworkable, which I
> happen to believe they need to be.

Certainly.
 
> > We added baseName back into occurrence for this reason, at Steve Pepper's
> > request, if I remember correctly.
> 
> I wondered what that was for - the lack of adequate XTM documentation is a
> distinct problem with understanding what these new extensions to the ISO
> spec are there for.

What you call a "lack of adequate XTM documentation" is a bit unfair. The
specifications are not even finished, much less has anyone had a chance to
write the definitive guide. We're working as fast as we can -- all in good
time. If you wanted to volunteer to fill in any holes in the documentation,
we're all ears. BTW, there still isn't really any decent documentation on RDF.
 
[...]
> My name must not appear in the Acknowledgements of XTM - if it does then the
> editors will lay themselves open for legal action.

I'll remove your name if you truly want it off of the Acknowledgements. I
see no reason to threaten legal action. Your name was included because you've
been part of the activity and we had no prior foreknowledge or notice from you 
indicating you wanted your name removed. If you don't want your clients to
know you were involved, don't tell them. But if XTM takes off (and I fully
expect it will, regardless of your advocacy), then removing your name is
probably not a good business decision. I hope you understand that inclusion
of your name in the Acknowledgements was done as a "thank you" to you for 
your valuable input, not as an endorsement. There's also somebody else we'd 
very much like to Acknowledge for their input but cannot due to political 
considerations, so you're not alone. But the final decision is certainly up
to you. If I receive either a public or private confirmation, I'll be happy
to oblige whatever decision you make. And rest assured, if your name is 
removed, you'll still receive my thanks.
 
> > As Nikita has demonstrably shown in his release of various XSLT
> stylesheets
> > that transform from the various ISO 13250-based DTDs into XTM syntax, the
> > specific syntax used is somewhat irrelevant. Given your background, I can
> > only guess that you're ignoring this in order to make some point. Calling
> > the work on XTM 'insane' insults the efforts of all of us who have put an
> > enormous amount of energy into its development, and gives little credit
> > or acknowledgement of the work and the specific requirements we operated
> > under. XTM 1.0 syntax is distinctly and explicitly an *interchange*
> syntax.
> 
> When I see a real need to interchange maps I'll change my mind. What is
> important is that maps interwork, not that they are interchanged. In my
> opinion they can interwork better by adopting a common abstract schema than
> they can be adopting a fixed tag set. I know you guys have no faith in XML
> Schemas, but when you come back to this listing in two years time you will
> understand why I have chosen to take an opposing stance at this stage,
> before Topic Maps get cast in unbreakable concrete in the XML world.

I'm not sure of the distinction you're making between "interchange" and
"interwork." These mean rougly the same to me (absent any differentiating
information).

Having provided a full and public review of XML Schemas to the W3C working
group, I'm quite aware of what their strengths and weaknesses. I see almost
no need for XML Schemas in the world of topic maps, in fact I only see the
danger of them being abused by companies with proprietary agendas. Everything
we need is easily supplied by the current syntax, a template language, and a
query language. None of the XML "equivalents" for the latter two will do, as
the processing model for topic maps (either ISO or XTM) disallows common XML
views of topic map documents (merging, scoping, etc. must be accounted for
properly, as described in our XTM Processing Model spec).
 
[...]
> What I am also trying to point out to people is that XTM is only a stop gap,
> and a relatively inefficient one at that. You should be aware that its not
> just XTM I am complaining about, but RDF, which is also not efficient enough
> for the long-term management of semantic

Well then, I guess we'll have to agree to disagree. I've been spending the 
last week or so finishing up a parser that builds an XTM topic map from a
Cyc ontology file. I've almost got the Linear Topic Map (LTM) to XTM parser
finished, and I'm looking at other mapping syntaxes. Perhaps next I'll try
RDF. Given that and the existence of Nikita's functional ISO-to-XTM XSLT
stylesheets, I can't see why anyone would continue to argue about syntax.
Syntax is irrelevant. Anything that can fit into the ISO 13250 processing
model (as elucidated in the XTM Processing Model spec) can "interwork." If
I can mix Cycorp ontologies, LTM, RDF, and four or five ISO 13250-based 
DTDs, what more can you ask for? The proof is in the pudding, and we're 
already eating it.

Murray

...........................................................................
Murray Altheim, SGML/XML Grease Monkey     <mailto:altheim&#64;eng.sun.com>
XML Technology Center
Sun Microsystems, 1601 Willow Rd., MS UMPK17-102, Menlo Park, CA 94025

      In the evening
      The rice leaves in the garden
      Rustle in the autumn wind
      That blows through my reed hut.  -- Minamoto no Tsunenobu