[CubeDM] - Target problem

[CresitelloDittmar, Mark]

All,

I have noticed a problem with my diagrams w.r.t. the Target object(s)
Current diagrams show:
  Observation Experiment contains Target class
  Spectral model extends Target -> AstroTarget
      adding astronomical object attributes (spectralClass, redshift,
varAmpl)

Obviously, datasets other than Spectral might have an Astronomical target..
so
this should be re-arranged as:

  Observation contains <BasicTarget>[1]

  <BasicTarget>
     + name:string[1]
     + description:string[0:1]
     + position:Position[0:1]

  Target - extends BasicTarget
     + objectClass:string[0:1]

  AstroTarget - extends BasicTarget
     + objectClass:string[0:1]
     + spectralClass:string[0:1]
     + redshift:real[0:1]
     + varAmpl:real[0:1]

I've added the abstract layer so that Target and AstroTarget are siblings
and one can specify one without allowing the other.  (ie. not all cases
would want to allow AstroTarget as Target).

I've kept the multiplicities as they currently are, but looking at this set
it seems that if I make an extension, at least 1 of the new elements should
be required [1].
  ** objectClass:string[0:1] -> string[1].  (default="") This is something
like "star", "galaxy", "calib"?
  ** spectralClass:string[0:1] -> string[1]. (default="") spectral class..
~ "Ia"


Side Note:
  In my researching the object, I found this IVOA Note regarding Spectral
Class.  The current models do not have any specs for Spectral class, but
suggest one would be good.  This note was addressing that request.  I will
want to discuss whether or not the new doc(s) should pull this in as a
suggested format/syntax.


http://www.ivoa.net/documents/Notes/SpectClasses/20111215/SpectClasses-20111215.pdf

It is used/referenced by MAST:
  http://archive.stsci.edu/spec_class/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.ivoa.net/pipermail/dm/attachments/20140403/3d50882b/attachment.html>