Re: SciBoard

Hi UCD board,
 Sorry, I've been on travel and then sick. I've tried to catch up on the emails.
 About the solar system UCDs, Seb. said:

> Hi Jonathan,
>
> There might be a slight misunderstanding on this.
> The planetary system community has worked on their needs
> to express the quantities they are measuring in their field
> with proper UCDs. You can see the result of this reflexion
> here:
> http://portail.imcce.fr/fr/expert/ssvo/wgovp/propucd.html

> I think these suggestions are worth examining, and they seem

I have no major problems with this list, I was talking about suggestions that were raised for object types like asteroid, comet, etc.

> quite mature: why not taking them into account for the PR.

I had assumed in Kyoto that we would go for the PR as is, on the basis that the current list is not too bad, and then use the board to make modifications shortly afterwards. I think everyone is happy with the syntax and the methodology, while we can spend ages arguing about the actual list. It sounds like we want to add lots of things to the list before going to PR, which I worry will unnecessarily delay the PR. See Andrea's message of Jun 3 - my view is that we should look for any things in the current list that are BAD, and get rid of them; there are many things that are MISSING, but we can easily add them later. The list will be stable in the sense that valid UCDs should rarely become invalid, but it will never be stable in the sense that it will always grow.

Which doesn't mean the board should delay discussing additions to the list (just that I don't think we have to agree on them before approving the PR). I'll give my comments on the planetary list below.

> I am less confident that we can incorporate in the PR all the
> astronomical object types/phenomenons that were discussed
> for VOEvent (is this what you call "semantics for astronomical
> objects"?).

Yes, particularly the tree of object types: galaxy.Seyfert2 etc. I'm not so worried about the phenomena (outburst, glitch, etc..), and I thought Rick's list was quite good. But I've heard you and others say that you are worried about putting the object types in UCD, whereas it's becoming clear that the object types do need to be somewhere in the VO fairly soon.

OK, now for the particular UCDs:

• meta.id.(x,y,z) Hmm. I thought meta.id was more for names, how about arith.cpt.x, etc.?
• obs.atm instead of obs.air I agree
• obs.atm.refraction, phys.refraction - one the angle, the other the index. I guess I agree, but maybe obs.atm.refractAngle and phys.refractIndex is more explicit?
• obs.elongation, obs.phaseAng - these are both angles. How about pos.angle.elongation, pos.angle.phase ?
• pos.az.azi and pos.eq.ha I agree.
• pos.brc.* In STC and I think in WCS-III the word 'topocentric' is being used for some of this, but body is fine too. I'd prefer pos.body.* which I find more obvious.
• pos.eop.* I like, although as we become more active on other worlds isn't it likely that we'll be needing similar parameters for Mars, etc.? But maybe it's fine for now.
• pos.precess Perhaps pos.eq.precess if we are talking specifically about the effect of terrestrial precession on coordinate systems (as opposed to the precession of the pole of (433)Eros or something)
• phys.magAbs.G Isn't there an H as well? My asteroid friends keep talking about H....
• src.orbital.*, src.veloc.* I agree, with the possible exception of src.veloc.tan. Perhaps 'transverse' rather than 'tangential' is more clear?
  • Jonathan