Call for proposals for SIAP Version 2

[Steve Lowe]

John Good wrote:

>While this might belong in the same "family" of services, I think it
>muddies the waters to try and fold it and the SIA together operationally.
>A positional constraint on the sky, whether for SIA or a Catalog Search
>replacement for the Cone Search, is pretty fundemental and (in various
>forms) already used extensively.  I think it makes more sense to keep
>it separate from frequency/wavelength/etc. issues.
>  
>
Do positional and wavelength constraints just seem different because of 
the way data is currently organized? Since an archive is often 
associated with a single instrument or collection of similar 
instruments, wavelength is already somewhat narrowed down just by the 
choice of archive, so sky position becomes the dominant search 
constraint *within* an archive. But if you're thinking of posing a query 
to the union of all archives, mightn't a wavelength constraint make as 
much sense as a positional constraint? Suppose you want to assemble a 
whole sky survey in a particular frequency band?

>Sky coordinates, because they are two dimensional on a sphere, have to be
>queried differently from typical relational constraints.  In our services,
>everything else is handled by simply allowing fragments of SQL syntax
>(e.g. "date between 10 and 20 or J_magnitude < 8.").  I can *imagine* needing
>more complex constructs in other subspaces but we've never encountered
>that need in practice.  Do you have instances of specific constraints that
>don't work with fairly the simple SQL WHERE-clause construct?
>
I'll have to think about this a little more, but I'm not sure we're 
proposing anything more complicated than WHERE. We're just including 
units and coordinate systems in the WHERE.

Steve

-- 
Steve Lowe
Smithsonian Astrophysical Observatory
Harvard-Smithsonian Center for Astrophysics
slowe at cfa.harvard.edu
1-617-496-1661