Vocabularies "steward"

[Markus Demleitner]

Dear Semantics, dear TCG,

The somewhat odd distribution is because this is mainly a, say,
governance question, but tightly related to Semantics.

Carlo has made a very good point in the RFC in Vocabularies 2
(https://wiki.ivoa.net/twiki/bin/view/IVOA/VocabulariesV20RFC).
Basically, VocInVO2 assigns to the Chair/Vice Chair of the Semantics
WG something like stewardship functions, in particular:

* Maintaining the list of VEPs
* Maintaining the vocabularies (i.e., adding terms as VEPs are passed,
  build the artefacts, etc)
* Run/maintain adapters to externally managed IVOA vocabularies

When writing the spec, I've made all ofthese responsibilities of the
Semantics chairs -- not because I think that's where they belong but
because there is nobody else we could plausibly assign them to; in
particular, the document coordinator has quite a high work load
already.

But Carlo is right: This *is* a bit of a precendent in terms of what
IVOA WG chairs do or don't do, and it is also right that
administrative (if you will) functions will probably profit from
somewhat stronger continuity than we have with the chairs and their
limited terms.

I believe we already have (at least) two models how this kind of
thing is handled in the VO:

(a) the document coordinator.  I *think* s/he is appointed as a
person by the exec.

(b) the IVOA web page, the mailing lists, and the Registry of
Registries.  I *think* these are entrusted to institutions that then
internally distribute the work load.

Both of these models could also be applied to either a
"vocabularies coordinator" or "the vocabularies repo".


At least for while I'm the semantics chair, I'm happy to shoulder the
administrative load, and so as far as I am concerned, no changes to
the PR are really required.

On the other hand, it's clear that a WG chairs' work load, in
particular with document reviews, but also as regards managing the
WGs' own standard processes, already may be discouraging to future
candidates.   Adding further chores almost certainly will not be
helpful then.

Hence, I at least wanted to raise the question from the depths of the
RFC page -- and in case you have opinions, alternative ideas or even
motivation to take on the chores, by all means chime in.

Thanks,

        Markus