On 2006 Jul 19 , at 12.29, Mark Taylor wrote:
>> 2. Is there any information we can add to the header of the FITS
>> table to enable an application to know that there are pol
>> data in it? (UCD???) [or is the concept of polarimetry so
>> foreign
>> at the moment that no-one has even considered non-mag columns
>> for
>> display in any means other than a cross at the correct place)
>
> To backtrack a bit: rather than using non-standard TUCDnn FITS
> headers,
> a better solution (which is also current practice) would be to
> transport
> the tabular data in a VOTable which has better metadata facilities.
> Whether that is feasible for GAIA is still to be determined depending
> on how we sort out GAIA's VOTable facilities.
I can see the practicality of this, but having thought about it more,
can I add a couple more arguments for the UTYPE experiment.
One is that I'd imagine it's easier for Peter to spot a TUTYPEnn
header than anything else (Peter?), and then act accordingly.
The other is that this would be an experiment both simple and
useful. Simple because it only requires a couple of extra headers in
experimental FITS files, plus Peter doing some string-matching in
GAIA. Useful because it would be the only example I'm aware of of
UTYPEs actually driving processing, and thus actually seeing whether
the whole UTYPE-vs-UCD idea is any good.
Also, this sort of approach would be very easy to use with RDF and
ontologies and all that stuff. I can imagine practical outcomes
fairly promptly. Set things up correctly and you can _deduce_ that
http://www.jach.hawaii.edu/utype/v1.0#stokes-q is a type of
phys.polarization.stokes, as is http://example.com/utype#wibble.
That's not a big deal right away (but the semantic web will save our
souls, blah blah blah), but it opens the door to a variety of
potentially clever things.
> An alternative would
> be to introduce a new message with the semantics "here is a table
> containing polarimetry data - do something with it if you want".
I haven't kept up with PLASTIC, so I'm not really qualified to
comment on its aesthetics, but... ugh! Isn't that legitimising all
sorts of nasty namespace-pollution? It sounds much neater for an
application which gets a broadcast OpenURL message to just look at
the file and see if it can do something with it. This would
genuinely be UTYPEs driving processing, and the Right Thing, and very
webby.
Norman
--
------------------------------------------------------------------------
----
Norman Gray / http://nxg.me.uk
eurovotech.org / University of Leicester, UK
|