New UCD1+ for SIA protocol

[Alberto Micol]

>1. In the list of ucd1+ that I derived, the atoms "image" and "scale" are
>present in instr.scale and obs.image . To me the concept of scale in this
>context is more important than the concept of image, then I supported
>"instr.scale".
>A much better alternative could be the definition of the new ucd1+:
>obs.image.scale
>

Sorry Andrea,

But if we go for obs.image.scale
then one would also think that the following should exist:

obs.spectrum.scale

And furthermore, personally if I see a scale at that level
then I also expect:

obs.image.wcs.ctype
obs.image.wcs.crpix
obs.image.wcs.crval
obs.image.wcs.cdmatrix
etc.
and similarly for obs.spectrum ...


Hence, I like much better the pos than the obs.image
because it is more generic.

So in the end I like much better:

pos.wcs.ctype
pos.wcs.crval
pos.wcs.scale  !!!
etc etc

because we can reuse it in many different contexts (spectra, images etc).
Let's keep together all the "how is the dataset sampled?" elements under
the same UCD (pos) and not let them spread under different ones.

Alberto
PS: Yes Anita they did a good job indeed!