Call for proposals for SIAP Version 2

[Francois Bonnarel]

hello,
   a couple of days ago I sent this answer to the Call for proposals for SIAP Version 2 on the uS metadata discussion group.
maybe some of the people here do not look at their archive so I
copy my mail here
François
_______________________________________________________________

Hello Doug and other NVO actors,
   I just read the Call for proposal for SIAP version 2 and some
of the comments.
   As I told  clearly discussing with Doug and Bob
Hanish last January, we, at CDS,  want to include SIA to our
Aladin services. We plan to have a new SIA compatible test version of the
Aladin server working for the IAU GA-Hope to have a beta working for Cambridge-.At the same time the Aladin client should admit access to all the SIA services.
   When the first specification of the SIA service were annouced here 
(September 2002), we (F.Ochsenbein and I) made a few comments on this 
specification. These were important for us to be able to join the SIA world 
with full functionnality.

I)   When reading the new document posted by Doug, we find two Image
Attributes parameters which are really helpfull: the data collection
identifier and the dataset identifier. Any service hosting several surveys
or observing programs need that. Please, keep that.

II)   A large part of the discussion is about the regions of interests, and
the POS/SIZE parameters.
    For cutout and mosaic services this parameter is intended to 
be used both for the image query, ie the availability problem, and the
image size definition. 
    Isn'it possible to dissociate these two aspects ?
   
    I think there could be two ways of doing that:
   1) the image query could answer by URL templates ( parametrable)
instead of "hard coded" URL. This can allow a size parameter different
from the initial ROI size. In these conditions the initial ROI can be a cone
instead of a rectangular area.

   2) Have a format = image/METADATA output which will describe the available
images intersecting with the ROI by giving Image attributes on them , but with 
no URL. In a second step (format = fits request) the ROI should be used to 
define the size of the image.

III) More generally I think it would be usefull to have this two step 
interrogation process for other kind of image generation problems
( examples: compress or not, resample or not, mosaic or not ...)

IV) A compression status should be added in the Processing Metadata
list.


By the way as far as DAL is concerned we have recently posted on the DM and
DAL IVOA groups a first description of another acces method designed for 
the AVO prototype demonstrated last january in Jodrell Bank the "IDHA" 
metadata-tree.  I think some of you may have a look to that and give 
interesting comments.


Best regards
Francois
év

=====================================================================
Francois   Bonnarel               Observatoire Astronomique de Strasbourg
CDS (Centre de donnees          11, rue de l'Universite
astronomiques de Strasbourg)    F--67000 Strasbourg (France)

Tel: +33-(0)3 90 24 24 11       WWW: http://cdsweb.u-strasbg.fr/people/fb.html
Fax: +33-(0)3 90 24 24 25       E-mail: bonnarel at astro.u-strasbg.fr
---------------------------------------------------------------------