[sc34wg3] DM conformance

[Patrick Durusau]

Lars,

Lars Marius Garshol wrote:
> * Patrick Durusau
> | 
> | Taking the idea that conformance to the XML Infoset is a
> | "conformance is a property of specifications that use those
> | definitions" one finds, for example:
> | 
> | The XQuery 1.0 and XPath 2.0 spec claims conformance to the XML
> | Infoset Appendix A: XML Information Set Conformance,
> | http://www.w3.org/TR/2003/WD-xpath-datamodel-20031112/, which lists
> | the properties used from the Infoset.
> | 
> | If we follow this path, does seem like conformance to the data model
> | is something that would be claimed in the other parts of ISO
> | 13250.
> 
> I used to think along these lines initially, and I think it was SRN
> who steered me in this direction at the Berline meeting, but I've
> since turned around on this. Don't you think it would be exceedingly
> strange if ISO 13250-3 *didn't* conform to ISO 13250-2? 
> 
Yes.

> And does it really conform to the data model, anyway? I'd say that it
> *uses* the data model, and as far as I can see the same goes for every
> other specifications I can think of, except possibly an API spec.
> 
Not sure what conformance is other than "using" the data model. If 
13250-3 defined the same terms as 13250-2 differently and proceeded to 
use those terms, I would say it was not conforming to the data model.

> I'm not sure we really *care* about DM conformance, either. If
> standards are about interoperability, then surely conformance is what
> enables interoperability, and it should be tailored to that end. Now,
> how does the DM actually affect interoperability? I don't think it
> does; it's just a tool used to define what XTM, TMQL, and TMCL
> actually mean. People must conform to XTM, TMQL, or TMCL, but I don't
> see how they can conform to the DM.
>  
Quite right but I was suggesting that XTM, TMQL, or TMCL conform to the 
data model, but not people in general.

> | I rather like the idea of specifications claiming conformance to the
> | data model as well as documenting why they are conformant.
> 
> Well, how could they not be? What could they do that wouldn't be
> conformant? 
> 
Consider the following example from the TMDM (still parsing the prose):

*******1 Scope*******

1 Scope

...

The purpose of the data model is to define the interpretation of the 
topic map interchange syntaxes, and to serve
as a foundation for the definition of supporting standards for 
canonicalization, querying, constraints, and so on. All
of these standards fall outside the scope of this part of ISO/IEC 13250, 
however.
*******/1 Scope******

Do we really want to say: "define the interpretation of the topic map 
interchange syntaxes..."?

Seems to me that 13250-2 defines the interpretation of topic map 
interchange syntaxes and standards for canonicalization, querying, 
constraints, etc., that rely upon the data model.

Fully agree with you that the data model is providing a foundation and 
that anything that relies upon that foundation must of necessity be in 
conformance with it, but how do I (as a reader) know what foundation a 
particular standard claims?

You could say it is obvious since it is all part of 13250 overall, but 
then one could make the same claim about the XML InfoSet. It is the only 
one at the W3C so what else would one be using for an XML standard.

The other problem is that I think we are construing the existence of 
topic map interchange syntaxes, querying, constraints, etc., with only 
those that appear from WG3. What if some third person wants to write a 
topic map interchange syntax? Not part of 13250 so there is no 
presumption, implication, whatever, that it uses the data model. If we 
adopt the practice of having other parts of 13250 say they are 
conforming (which I interpret for data models as using) to the data 
model, then we have set a practice in place for others to follow.

If topic maps become as successful as we all want, I don't think the 
interchange, querying and constraint languages of 13250 will be the only 
ones in existence. There is an SQL standard but probably more variant 
SQL syntaxes than people who have served on the various committees for 
the "official" SQL standard.

Hope you are having a great day!

Patrick

-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model

Topic Maps: Human, not artificial, intelligence at work!