Vocabulary+SV

[Doug Tody]

On Mon, 10 Sep 2007, Andrea Preite Martinez wrote:

> My point of view is:
> - Yes, if the present format is preventing further evolution and use.
> - Else: No. But we should add that the format (simple-txt or xml) can
> be easily transformed into something more suitable to fulfil the
> needs, should they evolve.

The key thing here is to define a standard vocabulary for astronomical
object types, so that we can treat these consistently within the VO.
For most such purposes, usage will be quite simple (e.g. verifying that
input is valid), and the format is not important so long as it is simple and
easy to deal with in a wide variety of languages and client environments,
without adding any new requirements or software dependencies (this last
is very important).  A simple text and/or XML format, with a single file
containing the entire vocabulary, is a good choice.

To experiment with sophisticated ontology/inference technologies (which is
a secondary priority here) it should be easy to auto-generate RDF, SKOS,
etc. from a simple-format master vocabulary, so long as this contains all
the essential information.  If desired this could probably even be done
on-the-fly by a service, delivering the vocabulary in any desired format.
If/when a new technology comes along, a new translator could easly be added.

For example if someone wants to build a portal which can translate
terminology between a local dialect and a VO standard, this might be
a good application to try out some more advanced semantic technology.
Most likely the format in which the vocabulary data is required will be
specific to the technology being used, hence it makes sense to defer the
translation until the data is to be used.  The only thing I would suggest
is that it might be worthwhile to verify that the model upon which the
master vocabulary is based is general enough to support auto-translation
to various standard ontological formats.

	- Doug