[topicmapmail] Topic Naming Constraint

[Miles Thompson]

I'm very happy with the proposed solution.

If I understand it right I even get to (optionally) merge on ticker
symbols without having to move the ticker symbold data from occurances
(where it seemed most natural) into basenames.

Thanks for answering my question.

Miles

> -----Original Message-----
> From: topicmapmail-admin@infoloom.com
[mailto:topicmapmail-admin@infoloom.com] On Behalf
> Of Lars Marius Garshol
> Sent: Monday, January 26, 2004 3:57 AM
> To: Topic Map Mail
> Subject: Re: [topicmapmail] Topic Naming Constraint
> 
> 
> * Miles Thompson
> |
> | Can somebody please explain (or point me to a reference) as to what
> | the working group intends to do with the Topic Naming Constraint in
> | the next version of the specification (XTM1.1).
> 
> You can find this issue in our tm-standards.xtm demo topic map:
> 
> <URL:
http://www.ontopia.net/omnigator/models/topic_complete.jsp?tm=tm-
> standards.xtm&id=topic-naming-constraint >
> 
> It was agreed at the SC34 meeting in Baltimore in December 2002 to
> remove the TNC in the form that it had in ISO 13250:2000 and XTM 1.0,
> but to add a capability that lets you do the same thing (and more!),
> but without all the problems the TNC had.
> 
> The current TMDM draft has this solution in it in section 7.5:
> 
>   <URL: http://www.isotopicmaps.org/sam/sam-model/#sect-name-types >
> 
> | I'm assuming that the working group will continue to recognize the
> | validity of this statement I found (via google) in the working group
> | list (but want to be reassured that it is)...
> |
> | > I see the usefulness of being able to do automatic namespace-based
> | > merges. This is extremely useful, but unfortunately the problem is
> | > that people (myself included) believe that base names are intended
for
> | > labelling purposes, not for identification.
> 
> Yes, this pretty much sums up the motivation for the eventual solution
> to the TNC problem.
> 
> | I am seriously hoping that, by placing certain "scopes" on Company
> | ticker symbols I will be able to merge in data from multiple data
> | sources so all data on the subject of a single company is accessible
> | from one place (er Topic).
> 
> Almost. You need to make the ticker symbols base names of a particular
> type, and then to make that type an instance of unique-characteristic.
> 
> | Personally, what would suit me just fine is if merging on basenames
> | in the unconstrained scope was not automatic, but that in defined
> | scopes, the basenames would merge. mmmm I suppose we strike here the
> | problem with exact definition of 'scope' (whether to 'include' or
> | 'exclude' scopes that share only one topic amongst many).
> |
> | Arghh.. Woah !
> 
> Yep, this was the problem exactly, and that's why we decided to use
> types instead of scope to specify uniqueness. :)
> 
> | But hopefully someone can let me know that name space merging (via
> | basenames in certain scopes) will continue to be available in the
> | next version of the XTM spec as I really need it.
> 
> Look at the draft and let us know what you think.
> 
> --
> Lars Marius Garshol, Ontopian         <URL: http://www.ontopia.net >
> GSM: +47 98 21 55 50                  <URL: http://www.garshol.priv.no
>
> 
> _______________________________________________
> topicmapmail mailing list
> topicmapmail@infoloom.com
> http://www.infoloom.com/mailman/listinfo/topicmapmail