semantics: RE: IAU thesaurus in RDF (an update)

From: Frederic V. Hessman <hessman-at-astro.physik.uni-goettingen.de>
Date: Mon, 8 Oct 2007 18:02:09 +0200

> From: Tony Linde <Tony.Linde-at-leicester.ac.uk>
> Date: Fri, 5 Oct 2007 16:19:57 +0100
> << I appreciate your enthusiasm, but going beyond simple broader,
> narrower, and related would demand a full-blown re-analysis of the
> ontological entries. Frankly, updating content, getting the old
> entries into rough correctness and legibility was already enough
> work.>>
>
> I don't think anyone is suggesting that you do all the work
> yourself, Rick. If we agree a format then a subgroup of this
> workgroup should be formed to take the forward.
>
> The question is: whether the extended SKOS is too much work for
> that subgroup or would substantially delay the creation of the
> required SV. If not then it would seem worthwhile to develop a more
> explicit specification than narrower/broader/related, so
> broaderGeneric, broaderPartitive, broaderInstantive (will we be
> recording instances in the SV?).
>
> T.

I think you're right about making this a more communal effort. I really wasn't trying to do an end run-around (for those familiar with American football), but to generate interest in a practicable adoptable solution (well, ... mostly!). Judging by the lack of any real complaints (at least in principle), I think and hope we're converging on this part of this problem.

The present thesaurus is still a bit of a moving target, although I think things are settling, so with a big push and lots of self- control (!), we could get it far enough to make it a real proposal. What I see remaining to do is first:

	Basic Idea:
		- acceptable or should we stop doing/talking about this?   A vote

of hands?

then

	IAU/IVOA Thesaurus:
		- name					I vote for "Updated IAU Thesaurus" (lots of political

brownie points and the blessings of continuity) or, if necessary, "IVOA Standard Vocabulary" (if we have to be purists)

ivorn et al. need to register this and other vocabularies somewhere and determine final resting place at ivoa.net
rules e.g. singular/plural format? restrictions on amount of ontological info (e.g. limit on # of NT's, BT's, RT's in present document)? See the rules we suggested for the SV.
normative format. I vote for text at first unless something simple like N3 is good enough (i.e. just as simple but with tools/ parsers/filters already crafted)
supported formats. Based on previous discussion, sounds like SKOS and OWL. Would like to see concrete translation patterns between text/N3 and SKOS/OWL.
documentation HTML-based help at least at the level of my simple lists/dictionary. One can quickly get bogged down in the ocean of tokens..... This is not about high-level computer post- processing but human-level decisions about what tokens are needed/ missing.
examples How to use both the tokens and the limited ontological info, both as proof-of-concept (read "post-facto justification") and to show potential to external parties
adoption Formulation as a formal proposal, desirable timeline.

and then immediately thereafter (or in parallel):

	IVOA Generic Vocabularies:
		- name					N/A
		- ivorn et al.				vocabularies should be VO resources (but should

be able to live outside as well - not everything happens as a registered VO activity)

rules N/A - every vocabulary can have it's own rules for forming tokens (within the syntactical limits of parsers) and chooses it's own level of ontological info.
normative format same final choice becomes IVOA standard until a new and better one appears
supported formats N/A, WG provides handy conversion tools if necessary
documentation N/A
examples Semantic WG provides UCD, A&A, AOIM, HOU, .....
adoption N/A - let a thousand vocabularies bloom!

Could you and/or Andreas get a (T)Wiki set up? All that I have is on my homepage (download IAU.tar). I'm sure Alaisdair Allan could put up a functional if temporary working site at HTN if the IVOA administrative path is difficult.

Rick

P.S. I will be away for the next 10 days (knee surgery) and may or may not find the time at home to work on this, so the ball is in ya'll's court.

Dr. Frederic V. Hessman     Hessman-at-Astro.physik.Uni-Goettingen.DE
Institut für Astrophysik          Tel.  +49-551-39-5052
Friedrich-Hund-Platz 1         Fax +49-551-39-5043
37077 Goettingen                 Room F04-133

http://www.Astro.physik.Uni-Goettingen.de/~hessman

MONET: a MOnitoring NEtwork of Telescopes

http://monet.Uni-Goettingen.de
------------------------------------------------------------------------ 
-------------------------

Received on 2007-10-08Z18:02:31