[Data Characterisation]: Spatial domain models comparison

[Doug Tody]

Alberto -

> >The revised strategy is to replace the "bandpass" data model with a
> >more general "sampling" data model which includes the bandpass information
> >as the sampling range
> >
> Indeed I like much better the "sampling" than the "bandpass".

Good!  The "bandpass" model was ok so far as it went, it just wasn't
including all the essential information to characterize the data.

> Aside comment, all you are saying here is new to me, since it was not
> circulated on this mailing list. I understand that the NVO has other
> communication channels, but I would think that NVO decisions which are going
> to affect the IVOA DM should also be posted here. Can that be arranged ?

There has been no discussion other than what you have seen here, except
for some mail I exchanged with Pat clarifying details of the CVO model.
Our first iteration identified some significant issues, so I thought it
best to incorporate these before proceeding further.

I did bring this up briefly in the NVO telecon this morning however,
mainly to discuss integration issues.  We need to look at RSM and at the
registry resource schema for compatibility at the top end, and should
also look at Arnold's STC stuff to see where there is overlap.  I see
you have already done some of that in the analysis you posted yesterday.

> Pat gave me his comments on the spatial characterisation comparison page,
> which I updated accordingly:
> 
> http://archive.eso.org/~amicol/VO/DM/coverage/spatial_comparison.html

Ok, I will use that version instead of the earlier email.

	- Doug