[iaufwg] FITS Registry Proposal

[Mark Calabretta]

On Tue 2006/03/21 17:26:56 CDT, William Pence wrote
in a message to: IAU-FWG <iaufwg at nrao.edu>

Dear Bill,

>1.  Bob and Doug both suggested opening up the registry so that proposed 
>conventions, or working drafts of conventions could be registered.  I think 
>this is fine as long as there is a clear distinction between the 2 (e.g., by 
>having separate web pages for the "proposed" and the "approved" 
>conventions).  So when someone submits a convention for approval it would 
>first be put on the "proposed" web page during the public comment period. 
>Then if the IAU-FWG approves it, it would be moved to the "approved" web 
>page.  Otherwise it would remain on the "Proposed" web page indefinitely, 
>(unless the proposers want to withdraw it).

There seems to be a conflict here between "proposed" and pre-existing
conventions.  I am thinking specifically of the SDFITS convention
(http://www.cv.nrao.edu/fits/documents/drafts/dishfits/sdfits.ps,
http://www.adass.org/adass/proceedings/adass99/P2-32/) which the ATNF
currently uses (in slightly modified form) for Parkes and Mopra data
and which consequently is used by other radio observatories that have
adopted our software.

The SDFITS convention exists already and has been used to transport and
archive a large mass of data from single dish radio telescopes including
several major surveys (e.g. Parkes/Multibeam and Arecibo/ALFA).  In this
sense it has already been "approved" by widespread usage; the main goal
in registering such a convention should simply be in documenting it (and
its variants).

In fact, SDFITS as originally described, would probably not pass muster
if offered now as a new convention because its WCS keywords do not
conform to the syntax prescribed for bintables in WCS Paper I - but
SDFITS predates that and in any case its keyword usage is not illegal.

>2.  Doug felt that the proposed approval process is too strict and the 
>requirements should be cut back to a minimum.  I agree that the requirements 
>should be keep to a minimum and am open to specific suggestions for how to 
>change the wording (at http://fits.gsfc.nasa.gov/fits_registry.html)  I 
>still think however that a convention must meet some minimum requirements 
>before it can be given the IAU-FWG Seal of Approval.  Doug specifically 
>mentioned the requirement of not duplicating other conventions, but the 
>current wording  does not forbid this, and only says that conventions should 
>not duplicate other conventions "without good justification".

This is fine for proposed conventions but for pre-existing conventions
(some of which may already duplicate others) I suggest opening it right
up and creating a wiki.

Cheers, Mark