Minutes MWG 2003-09-25

[Tony Linde]

Hi Ray,

> Slide 2 of Keith's planary presentation at Cambridge
> (http://www.ivoa.net/internal/IVOA/InterOpMay2003ResReg/RWP04p
> lenaryPresentation.pdf)
> entitled "Accomplishments" has the bullet:  "Agreed to adopt VOQL once 
> available".  

Good grief! So we did. It's hard to believe we were so naïve only four
months ago!

> In my recollection, the motivation for this starts with the realization 
> that ADQL is meant for querying VO databases.  A registry is a database.  
> The queries that the database will have to handle have the same 
> difficulties that queries to catalogs will have (e.g. handling coverage 
> constraints).  Thus, it makes sense to use the same query language.  

A registry is not a VO database, in concept at least. 

I agree that the cone search is an issue.

Given schema 0.8.2/3, what would a SQL-like query look like?

Cheers,
Tony. 

> -----Original Message-----
> From: owner-metadata at us-vo.org 
> [mailto:owner-metadata at us-vo.org] On Behalf Of Ray Plante
> Sent: 02 October 2003 21:57
> To: metadata at us-vo.org
> Cc: registry at ivoa.net
> Subject: RE: Minutes MWG 2003-09-25
> 
> 
> On Thu, 2 Oct 2003, Tony Linde wrote:
> > I am absolutely certain that there was no agreement to use ADQL for 
> > registry queries at Cambridge; that makes no sense to me 
> but I'm happy 
> > to be educated
> > :)
> 
> Slide 2 of Keith's planary presentation at Cambridge
> (http://www.ivoa.net/internal/IVOA/InterOpMay2003ResReg/RWP04p
lenaryPresentation.pdf)
entitled "Accomplishments" has the bullet:  "Agreed to adopt VOQL once 
available".  

ADQL was not a term in the general VOQL WG parlance at that time; 
nevertheless, this is what the statement in the minutes refers to.  

In my recollection, the motivation for this starts with the realization 
that ADQL is meant for querying VO databases.  A registry is a database.  
The queries that the database will have to handle have the same 
difficulties that queries to catalogs will have (e.g. handling coverage 
constraints).  Thus, it makes sense to use the same query language.  

I'm not sure if ADQL will be sufficiently "available" for our IVOA demos, so
we may see some short term variation in this arena.  That has benefits for
prototyping purposes.

cheers,
Ray