Spectral quantity in the object's rest frame?

[Markus Demleitner]

Hi,

On Thu, Sep 12, 2024 at 12:38:36PM +0200, Baptiste Cecconi wrote:
> The "pos.obs / pos.rest" extra UCD could be discussed anyway. I
> would be reluctant to put them under "pos", unless it could be
> applicable to other entities than spectral quantities. That is to
> be discussed in a VEP.

Oh, before I write the UCD-VEP, perhaps we can sound out everybody's
taste up front and save a few loops later on?

As I said, a pos.rest atom would make sense in time.period, too, and
it would apply to all the various designations of spectral axes
(wavelength, frequency, ...).  If it were not for that, I would
seriously consider em.wl.rest -- except that that would also conflict
with em.wl.central (say).

So, I *think* I am pretty much convinced that "rest frame" should be
a secondary atom.

I am not insisting on having it in pos, and you sounded like you did
not like that too much.  So:

(1) does anyone here have better ideas for location and form of the
"in the object's rest frame" concept's UCD atom?  Or would it perhaps
be more in the spirit of UCDs to have a more general "relative to the
object" concept and just say that (say) em.wl;obs.object means
"wavelength in the object's restframe"?

The second major question is whether we need something like "as
measured", i.e., the hypothetical pos.obs atom.  So:

(2) Should we declare that UCDs that are not qualified with an
"object" UCD are roughly in the observatory frame?  I think that's
what is largely the existing practice.  Or should we say "the
refposition is open when you don't qualify spectral quanities.
If you need to be explicit on observatory/barycenter/similar, use the
instr atom: wm.wl;instr"?  Or should we really have a new atom like
pos.obs?

Thanks,

             Markus