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

[Michel Biezunski]

> That's a great point - and actually re-inforces Martins point that
> we can't be limited to a single set of rigid tagsets.

The fixed tag set has nothing to do with Martin's request. The XTM spec.
is completely open to any semantic that users want to put, and there are
several
layers of semantics that can be superimposed.

>
> The solution is using Bizcode style markers.  In ebXML parlance
> this is a UID - Universal ID - and then having the UID as an
> ATTRIB of each element.   Therefore machine recognition of
> like parameters is then possible - along with Registry referencing
> and a rich associative behaviour implementation.

XTM 1.0 already works with that. There is no need for adding a new
mechanism.


> I'm sort of contridicting myself a bit - since machine lead interfacing
> is necessary to pull this off.  The union however is in the UID's
> themselves.  Using a prefix/value metaphor always UID's that
> are at least human verifiable - so I htnk we are ultimately talking
> about the same thing.  A blend of machine and human.  We want
> an approach that is sufficiently flexible that humans can interact
> and verify the process, but NOT so cryptic that the human is
> excluded from the underpinning entirely.

This is an application issue and is not -- and should not -- be defined by
the standard.

Michel
==========================================
Michel Biezunski, InfoLoom
Tel +33 1 44 59 84 29 Cell +33 6 03 99 25 29
Email: mb@infoloom.com  Web: www.infoloom.com
==========================================