UCD Status and Perspectives

[jcm at head-cfa.cfa.harvard.edu]

Patricio,
 I realize I never followed up your message of Mar 19. For the record,
what I had in mind with my spectral index example, I had in
mind finding the sources in a subsample and calculating what their
4100A fluxes are, for instance to get a sample for an observing proposal
of the 10 brighest northern objects of a particular class, brightest by
4100A flux. And I want to do it even including objects whose 4100A flux
is not actually known, but can be guessed from other published fluxes assuming
an SED.        
 In general, the existing UCDs are great for getting general information but
not for knowing exactly what you have, that was the broader point. A side point
from this example is that I think the VO will need to allow users to specify
certain assumptions to provide a context for queries (I'm looking for quasars,
and here's what you can assume about quasars - e.g. a typical SED - so that you
can handle my query even in cases where the information is incomplete... a bit
like a Bayesian prior in spirit, I guess).


  Regards, Jonathan