[tmql-wg] subjectIndicatorRef and resourceRef in TMQL

[Lars Marius Garshol]

* Lars Marius Garshol
|
| The TMQL we are working with will operate on the TMDM, hence it will
| support XTM 1.0, XTM 1.1, HyTM, LTM, AsTMa=, and so on. Or did I
| misunderstand your question?

* Rani Pinchuk
|
| I guess I fail to understand the limits of the planned TMQL. 

Ah, I see.

| In other email you answered to Steve Pepper:
| 	| and what about internal object IDs?
| 	I don't think TMQL should have any notion of such IDs.

Yep.
 
| As I understand it, you have a very clear vision of what is
| important in the TMDM and what isn't.

Well, object IDs aren't in TMDM. It doesn't have any such thing, and
that's why I think TMQL shouldn't use object IDs. Most topic map
implementations have some form of object ID, which is what Steve was
referring to. For example, in tolog you can say @93222, which will
refer to the topic map object with that object ID. There's no
guarantee that any implementation will have this, though, nor is there
any common syntactical restriction on this IDs. (I think some
implementations have used handles which are just opaque objects, for
example, which may not even have a textual serialization.)

Maybe you mixed up object IDs of this type (which Steve was referring
to) and XML IDs? The latter turn into source locators, and are what
we've used in the use case solutions.

| However, I am not sure if that vision is documented somewhere. The
| current design of Toma is in light of XTM 1.0. I now want to fix it
| to support TMDM, but I am not sure what to include and what not to
| include from TMDM.

Basically, TMQL should be able to query all of TMDM. 
 
| Another example is the base names types. Although it is defined in
| TMDM, it is not mentioned in the use cases. Is this also something
| TMQL should not take care of?

TMQL should definitely support it. I think the reason Robert did not
include it in the use cases is that it's quite new in XTM 1.1, and so
nobody supports it yet. I think you are right (in what you imply) that
it ought to be added to the use cases.

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