Handling data cubes in VO

[Arnold Rots]

Happy 2006 to you, too!

Yes, actually I was also one of those later contributors, arguing for
additional axes :-)

It may be helpful to provide the context for my original message.
I was asked to provide some use scenarios, not a list of required
tools and by doing so I wanted to dispel the notion which seemed to
prevail that "spatial image slices and individual spectra are all we
ever need".
I'm not trying to argue that we need a full-blown interface to 5-D
data with all possible options, but that we need to ensure that all
implemented protocols are extensible to such a generalized model - by
true extensions, not kludges.
I think that concern is justified since both SIAP and SSAP are so
strongly designed with 2-D images and 1-D spectra in mind that they
are not as easily extensible as they could have been.

Cheers,

  - Arnold

Anita Richards wrote:
> On Mon, 26 Dec 2005, Arnold Rots wrote:
> 
> > Access to 3-D cubes requires a single generalized interface.
> > Otherwise you are going to (a) confuse the users and (b) fail to cover
> > some cases.
> >
> > The access model is simply: everything - full cube, cut-out cube
> > (including those with only one pixel along one or more axes),
> > all possible transpositions, 2-D slices along any plane in three
> > dimensions, 1-D slices along any curve in 3 dimensions.
> >
> > Believe me, they are all needed.
> 
> Happy 2006!
> 
> Arnold is right - but so are later contributors pointing out that >3 axes 
> may be needed.  We very strongly need to keep standards generalisable 
> (e.g. not too bogged down in the details of IFUs*), but we have to get a 
> basic working model which can be tested against tools which astronomers 
> can actually use - not try and do evreything at once.
> 
> I would suggest cocentrating on describing 'science-ready' data, or at 
> least data where there are VO-enabled tools not only to display data, but 
> also if possible to make measurements and/or apply calibration (if not 
> already done) and extract commonly needed products.  Or rather, in some 
> cases, where a standard data model will facilitiate deveoping such a tool. 
> For example, allowing the  the Aladin-based slice tool to convert between 
> frequency, velocity and wavelength or adjust velocities to different 
> origins if there is more than one species in a cube.
> 
> Then we can move to more complex cases.  Arnold's first statement is a 
> very laudable goal which we should move towards, but if we try and cover 
> every eventuality  before we release aything we will never finish (and 
> produce something no-one wants, if it hasn;t been tested on real users).
> So for once I agree with Roy, who said:
> > I guess I would want to keep things as simple as possible, get a solid 
> > implementation and users, and only add sophistication when a real 
> > genuine use case -- or two or three -- scream their heads off and demand 
> > more complexity. That's just how I would do it.
> 
> Having just seen Doug's summary, I am now wondering if we are talking more 
> about data discovery, or manipulation? Of course, they are not totally 
> separate - a tool which allows a qhick non-quantitative look at the data 
> is useful to help decide whether to download a Gb FITS file... but we must 
> not lose sight of the VO goal to use standard models to allow data 
> discovery and manipulation to be integrated into workflows.
> 
> Datacubes (and higher dimensions) also highlight one of the bees in my 
> bonnet - in many cases, even for science-ready data, specialised software 
> will be best for manipulation - both due to the size of the cubes and the 
> range of ways to handle data.  Hence, as well as data discovery, we may 
> need software discovery - i.e., the user does not just need to find a cube 
> to download or view in a standard browser tool, but also to find software 
> to translate 'show linear polarization vectors' into 'if radio take arctan 
> U/Q ...' or whatever the equivalents are for other domains.  This is quite 
> possible, but not much discussed.
> 
> 
> 
> All the best
> 
> Anita
> 
> 
> *As I understand it,  Integral Field Unit data are recorded as single CCD 
> frames, but contain small stripes for each source representing a spectral 
> spread?
> 
--------------------------------------------------------------------------
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/
--------------------------------------------------------------------------