Time Series Cube DM - IVOA Note

[Petr Skoda]

Just after re-reading I have to make a little explanation of context I did 
not mention explicitly :

>
> If we admit  the work of IVOA so far was not based of clear design concept

this is reference to recent mails which give to me feeling that now it 
comes the right time to redevelop everything again using VO-DML and 
models boxing ....
sorry if I understand this not correctly ... as said its my feeling

and the criticism about VOTABLE in SCS ...

My objection was different :

citing from SCS example

http://mycone.org/cgi-bin/search?RA=180.567&DEC=-30.45&SR=0.0125

and SSAP example

http://www.myvo.org/ssa?VERSION=1.0&REQUEST=queryData&POS=22.438,-17.2&SIZE=0.02


- it gives the impression (it is about protocol not DM, sure)

that the IVOA have designed different semantic description for the same 
physical entity (circle on sphere) - not only that radius is SR vs. SIZE 
but the coordinates is ones POS as a tupple and once two entities (in fact 
the SCS is more semantic rich - every astronomer will understand what is 
RA and DEC and the SCS seams to be more convenient for them. The POS is 
strange - and requires explanation still leaving ambiguity - e.g. which 
coordsys is used ...

IMHO the SIAP2 is completely obscured to non-IVOA dedicated people.

The DMs should describe by definition the natural structure of entities 
and so should generate the same semantic description used in protocols as 
well as DB schemas, serializations etc.  - right ?

So it may look from outside that the IVOA has just now realized this and 
will finally define "proper" way of new standards.


Some critics to the statement about "what is the proper spectrum instance"

-----
I am thinking to present a critical review how the spectroscopy in VO 
stands namely using input from community behind stellar spectra libraries 
(recent conference in Brazil ).

In fact the real spectral experts need completely different spectrum model 
than what is currently in IVOA standards.

E.g. the binary table is nonsense - as it does not describe the proper 
wavelength dependence - the WCS description in FITS standard Paper III 
should be followed as it allows to compute clearly wavelenght on subpixel 
level.

another issue - in standard is given vacuum wavelength but most services 
ignore this and do not clearly state what is the wavelenght
as the simple transition is impossible, both should be allowed...
depends on instrumental parameters, environment etc ...)


More - the wavelength is not correct to work with - the radial velocity is 
crucial so the log lambda should be rebinned and WCS used to compute 
partial pixel values..

another - VO does not provide correct solution for echelle spectra, nor 
spectra data cubes !!!!

-----------

When I use my stellar spectropist hat I must say the IVOA have not 
consider most of such cases and so it prevents proper science.

Under hat of VO data provider I am glad that I can at least partially 
publish the data and analyse them in a rough way.... - Say for massive 
visualisation ....

This tiny shift from "almost, roughly" to "precise proper science 
analysis" is crucial for success of VO in scientific communities.


>
> The discussion here has shown the  way how to design future VO standards, but 
> it is terribly complicated in sense of previous way of thinking.

By this previous way of thinking I mean to include all model in 
implementation - which fails even for the SSAP as shown above ...


Best regards,


Petr

*************************************************************************
*  Petr Skoda                         Phone : +420-323-649201, ext. 361 *
*  Stellar Department                         +420-323-620361           *
*  Astronomical Institute CAS         Fax   : +420-323-620250           *
*  251 65 Ondrejov                    e-mail: skoda at sunstel.asu.cas.cz  *
*  Czech Republic                             skoda at asu.cas.cz          *
*************************************************************************