[sc34wg3] Question on TNC / Montreal minutes

Training & Consulting sc34wg3@isotopicmaps.org
Sun, 8 Sep 2002 22:13:24 -0700


Lars,

>
> * Nikita Ogievetsky
> |
> | ... skipped the national bodies part :-)
>
> Fair enough. :)
>
> * Nikita Ogievetsky
> |
> | I do not see it really that way.
> | In fact, in my example there should have been <baseNameString> in
> | place of <labelString>.
>
> That's your opinion, but knowing *why* that is your opinion might be
> useful.
>
> | RM people should correct me if I am wrong, but according to my
> | understanding inherited from Steve N, these two pairs are similar in
> | relation to the data they point to:
> |
> | resourceRef / subjectIndicatorRef
> | resourceData / baseNameString
>
> I think everyone would agree with that, whether they are RM-heads or
> not.
>
> | I just did not like <baseNameString> as the child element of <label>:
> | its name is too long and unnecessary constrains semantics
> | I like <labelString> or just <string> better.
>
> Actually, I don't see why it couldn't just have been <resourceData>,
> but as I said I feel <baseName> is misnamed in any case.
>
> * Lars Marius Garshol
> |
> | <instanceOf> on base names has now been made a separate issue from the
> | TNC, but it still needs to be settled.
>
> * Nikita Ogievetsky
> |
> | That is true. I just had thrown them together.
>
> :-)
>
> BTW, I added this as a formal issue in the SAM document now, with the
> ID 'names-with-types'. It will appear in the online topic map when I
> publish the next version of the topic map. If anyone knows of
> resources relevant to the issue, please let me know, and I'll add them
> to the topic map.

Damn it :-) I was just going to do it.

> * Nikita Ogievetsky
> |
> | By the way it is worth mentioning that one of the variants
> | considered in Montreal was to specify TNC-ness using baseName
> | typing.
>
> It was. I spoke for that at the meeting, but various people argued
> that the two were orthogonal. I didn't contradict them then, but am
> not 100% convinced that they are right.

As you remember I was also speaking for typed names...
As well as for introducing labels in parallel with names and for merging
along axis.
My feelings are similar to yours. I somewhat like and dislike most of the
proposals.
And still standardizing merging axis still looks the most becoming to me.

--Nikita.


  Nikita Ogievetsky,             Cogitech Inc.
  Topic Maps Tutorials and Consultancy
  nogievet@cogx.com   --   (917) 406-8734
  http://www.cogx.com     Cogito Ergo XML