[QUAR] Re: Expressing position in RDF

[Rob Seaman]

Hi Matthew,

> The identifying string from the STC in the VOEvent packet is UTC-FK5- 
> TOPO.


Point taken.

> My goal is to represent the RA, Dec information contained in the  
> VOEvent packet in RDFa.

Ok.  Assume a simple case.  Note that this cannot be called a  
"typical" case.  The coordinates for a point-like target fixed on the  
celestial sphere, with negligible proper motion and parallax.  Doppler  
and light travel corrections are to be dealt with downstream ("TOPO" +  
metadata for the observatory + epoch).

How does one convey these coordinates (RA, Dec, coordsys,  
conditionally equinox)?  Well, you either replicate the carefully  
negotiated truce from STC, or you make additional default  
specifications.  In either case, the assumptions are implicit in the  
representation.  One might, for instance, suggest something like:

	X has a SimpleWCS
	SimpleWCS has an RA (and Dec with implicit equinox and coordsys)

My fundamental point is that for proper system engineering you avoid  
saying:

	X has an RA

because this locks you into a specific set of assumptions.  Later,  
perhaps, someone will come up with a SomewhatMoreComplexWCS.  In that  
case, an application can evolve to support this while retaining  
backwards compatibility for SimpleWCS.

However it is done, there will be flavors of VOEvent packets that are  
inappropriate for representation using SimpleWCS.

> I think the answer to my original question is that there is nothing  
> currently existing in the VO to allow me to do this succinctly.

In that case this represents an opportunity for making an original  
contribution to the VO and to broader astronomical practice.

Rob