SIAV2 prototype / reference implementation

[Douglas Tody]

I guess the suggestion here is that an event list dataset could be
considered an alternative "image format" - the main queryData response
could describe data in this format as well as the associated sampled
image if available.  We are stretching the concept a bit, but yes this
is an interesting suggestion that we should consider.  - Doug

On Fri, 27 Sep 2013, Arnold Rots wrote:

> I would like to add something to Doug's message, regarding
> event lists.
> Just like filled data cubes are a special case of sparse cubes
> (defined as a collection of sub-cubes within the frame of a
> "master" cube), events lists are the other extreme: so
> sparse that each individual voxel is listed separately.
> They truly are multi-dimensional (usually at least 4-D)
> images with a limited number of voxels called out individually.
> 
> Mark Cresitello-Dittmar developed a proof-of-concept prototype extension
> of Doug's prototype that handles event data and provides cut-out
> services, creating either event list subsets or n-D images - thus proving
> the following concepts:
> - The inclusion of event data is quite feasible and does not require
> a great effort
> - There is no reason to relegate event datasets to "pass-through" only
> - It is feasible to integrate server-specific custom code as plug-in
> 
> The HEA community has been asking for the inclusion of its data
> in the VO for several years. Event lists are well- standardized, but
> the reactions has generally been that this must be too complicated.
> Mark's proof of concept proves that this is not the case and that
> there is no reason to exclude event list data from data cube
> protocol development - or even to put off its inclusion.
> Needless to say that adding HEA event data to the standard types
> of data products in the VO greatly enhances interoperability.
> 
> Cheers,
>
>   - Arnold
> 
> 
> 
> 
> 
> ---------------------------------------------------------------------------------------------
> ----------------
> 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 cfa.harvard.edu
> USA 
> http://hea-www.harvard.edu/~arots/
> ---------------------------------------------------------------------------------------------
> -----------------
> 
> 
> 
> On Fri, Sep 27, 2013 at 1:25 PM, Douglas Tody <dtody at nrao.edu> wrote:
>       Hi DALers -
>
>       If you want to try out the developmental/reference implementation of our
>       SIAV2 prototype you can exercise the service yourself at the following
>       link:
>
>           http://vaosa-vm1.aoc.nrao.edu/ivoa-dal/siav2query.html
>
>       This is just a crude Web interface used to test the prototype, but it
>       provides an easy way to see how it all comes together.  The query
>       response should render as a VOTable in your Web browser.
>
>       In the current version the MODE parameter, used to enable cutout mode
>       (and what I will be mostly focusing on in my talk in the DAL/DM session
>       later today), is disabled.  While most of the functionality for this is
>       implemented in the prototype service, it was not finished in time to be
>       deployed to the demo service in time for this meeting.
>
>       However the way it will work is simple enough: if for example one issues
>       a query with "MODE=archival,cutout", the query response will be what you
>       see now except that for each archival image for which a cutout is
>       possible, there will be an additional row in the query response
>       describing the cutout virtual image.  The metadata for the cutout image
>       will describe the image that will be created if retrieved via the access
>       reference.  No datalink is required in this case since the cutout image
>       is described directly in the query response, and already has an access
>       reference URL for retrieval.  All image retrieval, including retrieval
>       of archival images as well as virtual image generation, is performed by
>       the accessData request (the access reference URL returned in a query
>       response is actually a call to accessData).  AccessData can also be
>       called explicitly by a client to directly access a particular image
>       identified by its publisher dataset identifier.
>
>               - Doug
> 
> 
> 
>