[sc34wg3] Comments to N0277

[Lars Marius Garshol]

I've just read through the XTM 1.0 and HyTM syntax comparison thing,
and jotted down my comments on it. I've posted them here, as it seems
like there will be some back-and-forth, which is likely to be much
faster outside the national body comment ping-pong procedure.

Rough comments follow below.


What is the intended audience for this document? It's probably been
mentioned somewhere, but I've missed or forgotten it. Maybe it would
be an idea to mention this in the document itself? This would make the
document more self-documenting, and make it easier for people to link
to it appropriately.

What is the section "Describing the semantics of both syntaxes"
supposed to do? I've read it twice, but am still not sure. Is it
supposed to say that "the new topic map standard will explain this in
detail, but here is a summary of the syntax differences"? Or is it
meant to say something else? Or should it just be cut?

The "The difference between topic map syntax and topic map
information" section should probably be updated to use the new
terminology adopted in Orlando. The term "topic map parser" sounds a
bit troublesome. Perhaps it is best to remove it? 

The document should perhaps also be updated to define and use the term
"HyTM syntax"?

Comments on the syntax section:

 - the differences in the structure of topic names is not mentioned

 - in HyTM associations consist of (role type, role player) assocrl
   elements, while in XTM 1.0 they consist of (role spec, role
   player+) member elements:
   
   - terminology differences member/role, role type/role spec

   - structural differences: single player vs multiple players

 - the term "theme" is not used in XTM 1.0

 - no added themes in XTM 1.0

 - no "TMBrid" in XTM 1.0

 - the terms "topic link" and "association link" are not used in XTM 1.0

 - no mnemonics in XTM 1.0 (topic / @linktype, occurs / @occrl ...)

 - no scope on topics in XTM 1.0

 - no mergeMap in HyTM

 - there was something about scope that ISO 13250:2000 got wrong,
   according to SRN. I've forgotten what it was, but this may help SRN
   remember. :-)

--Lars M.