Comments on V1.1 - Future of VOTable (flame bait sigh)

[David Berry]

The difference between a "data format" and a "data model" needs to be
borne in mind. Data formats are what you use to instantiate a data
model.

For example, the "fits-wcs data model" (for 2D celestial coords)
is a shift of origin, followed by a matrix multiplication, followed by a
spherical projection, followed by a 3D spherical rotation. On the other
hand, the "fits-wcs data format" is collection of fits keywords, CRVAL,
CRPIX, etc. But there certainly are other data formats which can be used
to instantiate the fits-wcs data model. The issue of choosing a data
model and a data format should ideally be handled separately, because
otherwise you can easily end up with the data format influencing the nature
of the data model and thus making it difficult to instantiate in other
formats.

VOTable is obviously a useful data format for tablar data, but if it is
also being sold as a general data model, then I'm worried. We have a
distinct effort to decide on data model components, and it may well be
that the outcome of that effort is a model which can be instantiated as a
VOtable (in addition to other data formats).

David


> On Thursday 08 April 2004 1:50 pm, Anita Richards wrote:
> > > >   * re 2: Data Model
> > > >     this section should begin with a note that the VOTable
> > > > representation of a Data Model will be superseded by the work of the
> > > > IVOA Data Model group.
> > >
> > >   I basically disagree -- I do not see VOTable as a temporary patch
> > >   until the (necessarily complex) final data model comes up.
> >
> > I too think that VOTable is here to stay for a decade or so.  The use of
> > VOTable for metadata is being designed-in to major new instruments e.g.
> > ALMA, and for the rest of us more modest data providers, not
> > to mention tool-designers, it has been a major step forwards to get a
> > large proportion of the community to go over to something as standard and
> > useful as VOTable.  It may not be perfect but sometimes you ahve to work
> > with what you ahve got.  The community is no better resourced than we are
> > on average, probably worse; we should minmise the transoformations we
> > require.  It is vital that our prototypes can be turned into real global
> > tools now and incrementally or we will not serve real needs.
> >
> > I realise that there are things which can only be done using VOTable
> > alongside clunky fixes, and Tony's viewpoint is that using VOTable cuts us
> > off from efficient applications of other functionality.  However I think
> > this is an 80:20 situation and VOTable solves the important 80%...
> >
> > Maybe this question should be posted to the dm mailing list?  As an
> > apprentice dm person, I think that we very much need unifying standards
> > based on real models and I don't see anything which is better than
> > VOTable, and real data providers certainly can't wait (or be given
> > excuses to delay).  Using the FITS analogy, for a long time FITS was
> > nearly standard but people used the lack of formal agreement (now reached)
> > to produce all kinds of abominations which ignorred the keywords which
> > already did have general consensus.  If we start dissing VOTable the
> > practical consequence will be more sloppy metadata, not better, in the
> > life-times of most of our projects.  But am I just a paranoid
> > conservative?
> >
> > cheers
> > a
> >
> > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> > Dr. Anita M. S. Richards, AVO Astronomer
> > MERLIN/VLBI National Facility, University of Manchester,
> > Jodrell Bank Observatory, Macclesfield, Cheshire SK11 9DL, U.K.
> > tel +44 (0)1477 572683 (direct); 571321 (switchboard); 571618 (fax).
>
> --
> Martin Hill
> Astrogrid/AVO, ROE
> Tel: 07901 55 24 66
>
>

----------------------------------------------------------------------
Dr David S. Berry    (dsb at ast.man.ac.uk)

STARLINK project		 |	Centre for Astrophysics
(http://www.starlink.ac.uk/)	 |	University of Central Lancashire
Rutherford Appleton Laboratory	 |	PRESTON
DIDCOT				 |	United Kingdom
United Kingdom			 |	PR1 2HE
OX11 0QX                                Tel. 01772 893733
                                             01257 273192