SV: do we need it?

[Doug Tody]

On Fri, 14 Sep 2007, Tony Linde wrote:
> It has been suggested that the need for the SV is obvious and pressing. Is
> it? What for? Maybe people could reply to this with their own views.

One of the original motivations for the SV came from DAL/DM: we
have a need to describe the type of astronomical object observed
(Target.Class), or search for data for a particular class of object
(TargetClass input parameter) and found that, while some partial
compilations of this type had been created within astronomy, there
was nothing comprehensive.

As a starting point it would be good to have such a list so that
data publishers can assign a standard classification to an object
when an observation is published, rather than just make one up as
we do currently (this does not require infererence, just a good
well thought out and comprehensive classification system).  Later,
once we have this, is will also be desirable to be able to map less
well-formed user inputs for astronomical object types to the SV in
a portal, so that we can use this for data discovery (this could
benefit from inference capabilities).

I imagine VOEvent (at least) has a very similar problem.

	- Doug