[sc34wg3] SAM issue: sam-conformance

[Lars Marius Garshol]

* Martin Bryan
|
| I think that there are two levels of conformance:
| 
| 1) conformance of applications to a particular syntax

So far, this matches my proposal that SAM have no conformance clause,
and that only the syntaxes and TM?L define conformance.

| 2) conformance of a particular syntax to a particular model.

I'm not sure this one is meaningful. Do we want to draw a line between
SAM-conformant syntaxes and syntaxes that are not SAM-conformant? If
so, why?

We will define XTM and HyTM as topic map interchange syntaxes.
Obviously, those will very much have the ISO stamp of approval, and so
long as we do our job properly I don't think anything more is needed.

There will also be things like LTM and AsTMa=, alternative syntaxes
for topic maps. These will not be ISO-approved and may or may not
support all of topic maps. Are we interested in some sort of QA
program for them? I don't think so. What would the purpose be? (They
should have deserialization specifications that map them to SAM, but
that's *not* the same as meeting some syntax conformance criterion.)

Also, there will be things like XFML, NewsML, and the RSS Easy News
Topics which, although they are not full topic map syntaxes, have
definite connections to topic maps. These also can be mapped to the
SAM, but they would never meet a SAM conformance criterion in any
meaningful way.

Finally, there will be things like RDF, XML, and RDBMSs, all of which
can have application-specific mappings to SAM.

This is what I've been thinking, anyway.

-- 
Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no >