Vocabularies issues

[Rob Seaman]

Norman Gray wrote:

> Matthew Graham wrote:
>
>> # Working Group prepares Working Draft (version ≥1.0) and submits  
>> to Document Coordinator for posting in the IVOA document collection.
>
> That's the important first step.  I have the impression (partly  
> encouraged by Rob) that the document is in a state where Andrea  
> might legitimately send it to Bruno as a WD.

Yes, although I think we're confusing a couple of issues here.  The  
altercation a couple of months ago was about whether work products  
(excellent such) from a small group corresponded to an official WD of  
the WG.  This has since been resolved and the current draft already  
has the nature of a group effort.  But the second question then  
becomes whether we've jumped through the succession of hoops required  
by the IVOA.  That is where we stand at the moment.

>> # Working Group reviews the Working Draft. Two reference  
>> implementations of any associated software are recommended.
>
> I think that the VOEvent group is primed to start using vocabularies  
> as soon as the rest of us can give them a fixed target.  Alasdair  
> Gray, in the Explicator project, is working on a web service which  
> can help a (human or application) client explore a vocabulary.  Do  
> these sound like `two implementations'?; if not, I share your  
> curiosity about what would.

Note the "any associated software".  We only need reference  
implementations if implementations are needed.  I don't think this  
will be a problem - if there aren't two or more groups interested in  
using vocabularies, what the heck are we doing here?  However,  
defining a format for vocabularies is not like designing a transport  
protocol for VOEventNet.  While there is no VOEventNet if two or more  
brokers don't talk, there certainly can be a vocabulary even if all  
speakers of a language are extinct.

>> # The Chair of the Working Group, with consent of the WG, promotes  
>> the document to a Proposed Recommendation and submits it to the  
>> Document Coordinator for posting in the IVOA document collection.
>
> Given that these implementations (if that is what they are deemed to  
> be) appear in the next couple of months, it might be that we can  
> manage a couple of iterations of the document in that time.  If so,  
> I'm optimistic that we could get to the RFC stage four weeks before  
> Trieste (that is, before 21 April).
>
>> # The Chair of the Working Group issues a formal Request for  
>> Comments (RFC) to the e-mail distribution listinterop at ivoa.net. The  
>> RFC and all comments must be logged on a TWiki page whose URL is  
>> given in the RFC. A minimum comment period of 4 weeks must be  
>> allowed.
>> # The Working Group Chair responds to comments on the TWiki page.  
>> If comments lead to significant changes to the document, the status  
>> reverts to Working Draft (back to Step 1).
>
> I'm not in a mad rush here, but we seem to have got some momentum  
> here, and I don't believe we have a massively hard document to  
> develop.
>
> How does that sound?

Sounds good.  There is nothing special about Trieste other than that  
we can use it as a milestone to push the effort forward.  On the other  
hand, we need to avoid a situation in which the overhead of the IVOA  
process itself hampers the success of the working group efforts.  As  
Norman says, there appear to be no blockers.

Rob