[QUANTITY] Use-cases, role in larger scheme (Was: Re: [QUANTITY] Quantity "arguments")

[Steve Lowe]

I advocate much more detail. Let me tackle case 1 to show the level of 
detail that I think is needed for a proper use case. I'll be putting in 
a lot of questions instead of answers, just because of my ignorance. But 
the proposer of the case would presumably know the answers and could 
easily fill them in. Other questions may be ones for which even the 
proposer would be uncertain, but then they lead to further discussion.


Brian Thomas wrote:
 > The following [use case was] taken from the NVO SWG list.
> -----------------------------------------
> 1. GRB follow-up service
> 
> - Upon a receipt of a GRB detection message (from GCN, SAX, HETE,
> Swift...):

We get some information in the message. What is it? Position, time 
estimated errors in each? Specify the allowed reference/coordinate 
systems and the requirements on the error model (e.g., does you need 2-d 
gaussian standard error?)

> - Grab the image and catalog data from all linked archives covering the GRB
> error box
> 
> - Cross-match any sources in this set
> 
So what do you need these source catalogs to be? What does the 
cross-matching algorithm need? Maybe a catalog is a list of objects, 
each with a position, observation time, error estimates. Same questions 
as before regarding allowed systems and error model.

Do you download the whole catalog, or filter it first? I would think 
you'd want to do the latter. So for each catalog, you ask for all the 
entries within some distance of the GRB position, as determined by the 
GRB error. Specify the query you will run on the catalog(s). How big are 
the tables you expect to get back---huge or small? Are there problems if 
the queries don't really do the same thing on different catalogs (e.g., 
maybe it approximates the boundaries).

How does your cross-match algorithm chew on the data? I'd guess that you 
start by getting everything into the same coordinate system. Could you 
treat it as a unitless problem from this point on? Then you have to load 
the points into some data structure to do the cross match. What is this 
structure? There are probably some parameters that control the 
false-match/miss tradeoff; what are they, what are their units, are they 
determined by astronomical or mathematical considerations.

> - Identify possible permanent-but-variable sources (i.e., not 1-time
> transients) in this area from multiple observations at the same (+-)
> wavelengths (e.g., optical, others if available)
> 
Very much the same questions as for the cross-match.

> - Prepare the finding charts with all the sources labeled, links to the
> pertinent info
> 
What list is used as source for stars that will appear in the finding 
charts? ...

> - Have it all on a website
> 
Describe web pages and functions.
> - Email to the subscriber list that this is available
> 
echo "See all our GRBs at grb-master.ivoa.net" | \
mail -s "Got another one!" grb-lovers at ivoa.net
> - Update as more data come in
loop

Gee, after typing all that I'm tired. Anyway, that's the kind of 
information that I'd like to get from the use cases.

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