[sc34wg3] 4.3 Constraints

[Lars Marius Garshol]

* Patrick Durusau
| 
| If we don't have conformance or as Lars as said, what does it mean
| to conform to a data model, then the data model can't very well
| constrain "allowed instances of the model." 

Actually, it can. The structure of the model as a set of information
item types with properties gives much of the rules for the model
instances, but not all rules can be covered in that way (same as EBNF
isn't enough to describe XML and you need the WFCs and VCs in
addition).

Now, while I don't think an implementation can reasonably claim
conformance to the model, an XTM processor can claim conformance to
XTM. XTM requires such processors to detect cases where the
deserialization builds DM instances that violate the constraints of
the model, and to report these as errors.

That's one reason to have the constraints there. Another is to make it
clear what is and isn't allowed in topic maps, so that TMQL, TMCL,
CXTM, and other specifications can take that into account. Even LTM
needs to be aware of that so that the rules can be followed.

| (BTW, what are instances of the model?)

If you have a topic map information item you have an instance of the
model.
 
| Hence my redundant proposal (following a much earlier one by SRN),
| to follow the practice of the W3C with regard to the InfoSet. At
| least in that case I don't have to guess whether a particular XML
| standard follows the InfoSet or not, and, more importantly for me, I
| know where it departs from the InfoSet.

Well, the thinking is that *every* topic map standard will follow the
DM, and that if it doesn't it won't be a topic map standard, but
something else. Also, I don't think such a declaration will be of much
use if you don't trust the authors to be able to follow the DM, since
then you can't trust the declaration either.
 
| BTW, I missed the "constraints on the allowed instances of the
| model" part. Pointers anyone?

They are listed throughout the document. Just search for "constraint"
and you'll find them.

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