Request to add some parameter types to CEA schema

[Guy Rixon]

On Wed, 10 Oct 2007, Paul Harrison wrote:

>
> On 2007-10 -08, at 14:26, Guy Rixon wrote:
>
> > Hi,
> >
> > recent work with CEA v1 (the AstroGrid one) reveals a need for some
> > extra
> > parameters types. Specifically we need a Table type that is less
> > specific than
> > VOTable. We need this to cover DSA where the output is always a
> > table but can
> > be in any of three formats. Similarly, Image would be a useful
> > type, more
> > useful than "FITS".
> >
> > This need persists into CEA v2 and UWS-PA.
> >
>
> The current CEA v2 proposed schema has the following as parameter
> "types" - (see the ParameterTypes definition in http://www.ivoa.net/
> internal/IVOA/CommonExecutionArchitecture/CEABase.xsd)
>
> integer
> real
> complex
> text
> boolean
> anyURI
> VOTable
> angle
> MJD
> DateTime
> ADQL
> ADQL-S
> STC
> STC-S
> binary
> FITS
> XML
>
>
> So I think that  CEA parameter "type" can probably be thought of a
> sort of  serialization using the vocabulary of the DM group - but
> this has always been a slightly fuzzy concept in CEA  - and the above
> list is probably not entirely self consistent - the main driver for
> this "typing" was to give hints to the User Interface on how to form
> the content of the parameter (e.g. an ADQL widget to help with
> forming adql queries). Each parameter has the possibility of being
> further defined by UCD and UTYPE, which for instance allows angle to
> be subtyped into RA.
>
> I would have no problems with adding Table and Image as types (and
> perhaps Spectrum?), but perhaps it would be a good idea to review
> this whole area to see if there are any more useful classifications.
> - or perhaps the serialization needs to be a separate concept (so
> FITS and VOTable) do not belong in the above list,. Maybe review the
> currently used PLASTIC messages, as they have a similar purpose of
> being suggestive rather than prescriptive about what is actually
> being passed around...

If I were to choose, I would make the types URIs leading to registry
resources, as is done for VOSpace. This would open the vocbulary and save us
some discussion time.


> As a final point - in the future what will DSA have to do with CEA2
> (i.e. UWS-PA)? surely this will be pure TAP...(which will include its
> own UWS)

The stated intent (Bejing, Cambridge) is to do as you say.

>
> Dr. Paul Harrison
> JBCA, Manchester University
> http://www.manchester.ac.uk/jodrellbank
>
>
>

Guy Rixon 				        gtr at ast.cam.ac.uk
Institute of Astronomy   	                Tel: +44-1223-337542
Madingley Road, Cambridge, UK, CB3 0HA		Fax: +44-1223-337523