WD for the Obscoredata model and its implementation on TAP

[Anita M. S. Richards]

>
> Like Tom McGlynn, I find the prescribed unit for the spectral axis
> (wavelength) very awkward. Only 1/3 of the community thinks in this
> inverse unit. I would strongly suggest allowing wavelength, frequency,
> or energy; if we really need to stick with only one, I would suggest
> frequency as the most physically meaningful one.
>

Hear hear, allowing all 3 is relatively simple and with the new broad-band 
radio instruments, which observe in up to a span of half the observing 
frequency, evenly sampled in frequency, even a 'representative' wavelength 
could be misleading by a factor of 2.

> TimeScale is missing from the list.
>
> Reference positions are missing, too. They really need to be there.
> One might require them to be the same for space and time, but the
> spectral reference needs to be separate.
>
> I commented earlier on the Redshift axis. Its specification requires a
> Redshift-Doppler toggle, Doppler definition (optical/radio/relativistic),
> a reference position, and possibly the Doppler units (though km/s is
> pretty much globally accepted).
>
> I won't harp on my long-standing objections ;-)

Thanks Arnold (apart from the last comment ;-( ) - you put what I was 
going to 
say much better!

Seasons greetings

Anita