Relationship between Q and STC

[Arnold Rots]

David,

Why aren't you off to the pub, yet?  :-)
Yes, I think this ties it up, except for Region which is left
dangling.  Just like the Coords element defines a point in space, the
Region (or more precisely, CoordArea) defines a volume in space - it's
a natural extension of single valued quantities.

The CoordSystem is very much a phenomenon (in the singular, please).
As such it is based on reference positions.  Whether those ought to be
considered quantities, I don't particularly care, but that goes back
to more philosophical issues as to what absolute knowledge we have.
"TOPOCENTER" could be a quantity, but it is one without information in
the absence of a related quantity (i.e., coordinate values) that says
where the TOPOS is (I know, "TOPOCENTER" is a bit of an oxymoron, but
at least people know intuitively what it means).

Good night, it's the west coast's turn,

  - Arnold

David Berry wrote:
> Brian,
> 
> > > Because an axis is not a Quantity, it's a phenomenon.
> >
> > 	That is not true. An axis is a phenomena that describes one dimension
> > 	of another phenomena. For example, "FLUX(Time)" [flux measurements
> > 	recorded for a particular time] is a 1-dimensional quantity where the
> > 	"TIME" quantity is both a phenomena (of "time") AND describes a
> > 	dimension of (and is thus an "axis") of another.
> 
> The point I was making that an axis *describes* a phenomenon but does not
> provide any values for that phenomenon. So:
> 
> 1) "topcentric radio velocity in units of km/s" *is* an axis and is
> equivalent to a 1D Frame. This is "the phenomenon".
> 
> 2) "23.345 34.23 100.23" are values and can be stored in a ValuesList or
> generated by a Mapping.
> 
> 3) Put together the Frame and the Mapping (or ValuesList) and you have a
> Quantity, That is, a a list of values for some phenomenon.
> 
> This is why I say an axis by itself (without any associated axis *values*)
> is not a Quantity, it's a Frame.
> 
> 
> > > An STC document does not specify any particular *positions* in space-time
> > > coordinates, it just describes a coodinate system, so what values are
> > > you going to put into an STC Quantity?
> >
> > 	It doesn't have to be providing the actual positions to be a Q. All that
> > 	is needed is that I pair a value with it (Frame information is *optional*,
> > 	and in the STC variant I provide, you cant put coordSystem information
> > 	on any of the STC components as that would be wrong).
> >
> > 	Certainly, for the STC components there are enumerated values (as
> > 	I previously described) which are allowed to be there. Are there no phenomena
> > 	for which a limited, discrete set of values exist? I think so, and hence,
> > 	STC *can* and *is* a Q.
> >
> > >
> > > > > But a Quantity should be "a rule for obtaining values", and STC
> > > > > specifically excludes issues to do with obtaining values - its just
> > > > > defines the coordinate syetem and so is a Frame, not a Q.
> > > >
> > > > Thats not true. I see components of the STC as quantities,
> > > > for example "StandardReferencePosition" is a quantity, albeit one
> > > > which is allowed to  take one of a restricted set of values, e.g.
> > > > "GEOGENTER", "BARYCENTER", etc.
> > >
> > > "StandardReferencePosition" simply corresponds to a property of the Frame
> > > described by the STC document. There is no need to consider it to be a
> > > Quantity.
> > >
> > > > And if I can get all the component parts of the ST to be Q's (and
> > > > I can) then this means at the STC itself can be a Q (as StandardQ
> > > > allows holding "member" Q's).
> > >
> > > But again, the whole notion of an STC document is that it describes a
> > > coordinate system without specifying any positions within that coordinate
> > > system.
> >
> > Quit getting hung up on the values of the actual coordinates, thats a
> > red herring. The STC is a quantity that provides meta-data about coordinates,
> > the actual coordinate values are held by a parent Q.
> 
> This comes back to what a Quantity is. The above says that STC *is a*
> Quantity, but it has no values (the values are held elsewhere, in a
> "parent" Quantity). So we come back to the same point, is the
> purpose of a Quantity to hold *measured values* for some
> phenomenon. Presumably you would say not since you say that STC is an
> example of a Quantity which does not contain any values.
> 
> 
> > > Your star catalogue has columns corresponding to RA and Dec, the
> > > associated STC document describes this (RA,Dec) system but does not itself
> > > include the numerical values stored in the table columns. The STC document
> > > is the Frame, the table columns are the ValuesLists, and both *together*
> > > form a Quantity.
> >
> > 	*sigh* I guess I will have to grab an example and put it in the email since
> > 	nobody is actually looking at the tarball.  Please take a look and comment
> > 	on the example.
> 
> Apologies! Your example made me realise that I have not been very precise
> about what I mean by "STC". I have been equating "STC" with the
> "CoordSystem" element. So in my previous points, make a global edit of
> "STC"->"CoordSystem". Obviously the STC schema also includes the "Coords"
> element which *does* include axis values, and so *is* equivalent to a
> Quantity.
> 
> David
> 
> 
> ----------------------------------------------------------------------
> Dr David S. Berry    (dsb at ast.man.ac.uk)
> 
> STARLINK project		 |	Centre for Astrophysics
> (http://www.starlink.ac.uk/)	 |	University of Central Lancashire
> Rutherford Appleton Laboratory	 |	PRESTON
> DIDCOT				 |	United Kingdom
> United Kingdom			 |	PR1 2HE
> OX11 0QX                                Tel. 01772 893733
>                                              01257 273192
> 
--------------------------------------------------------------------------
Arnold H. Rots                                Chandra X-ray Science Center
Smithsonian Astrophysical Observatory                tel:  +1 617 496 7701
60 Garden Street, MS 67                              fax:  +1 617 495 7356
Cambridge, MA 02138                             arots at head.cfa.harvard.edu
USA                                     http://hea-www.harvard.edu/~arots/
--------------------------------------------------------------------------