A note on a possible extension to SIAP

[Pedro Osuna]

Hi Roy,

> Please Pedro and Francois, can you supply an alternate logic that > I
would use if your schemes were to be accepted?

We would still keep an image CDATA pointer in our "top level" structure;
this means that if you would be using our service, you would not have to
modify your code at all: you just stay at the highest (let's say, the
"flattest") level in our hierarchy, and that's it, you stop there.

On the other hand, I believe that in the process of building a whole VO,
having to change our code will be a common task for everyone. 

With respect to a use case that would require some structure, I have
one, coming from the AVO demo, that clearly illustrates the point: 

- guess whether a Young Stellar Object is  of type I or II.

In this use case, images in different X-Ray bands should be used, as
YSOs of different types emit in different X-Ray regions. You can check
this use case at:

http://www.euro-vo.org/twiki/bin/view/Avo/XMMUseCases


With respect to how much "beautiful" a flat structure is as opposed to a
structure where each data is identified and structured in the display, I
have my doubts as -fortunately- beauty is quite a subjective thing.

Cheers,
P.



On Tue, 2004-04-20 at 18:37, Roy Williams wrote:
> >      Human Interaction or Computer Interoperability?
> 
> Thank you Alberto, good question. In fact, maybe we could broaden the
> question to the one we should start with -- What are the Use Cases for SIAP?
> 
> Here is one at least.
> 
> I have invested some time in building software that gets its data through
> SIAP (Atlasmaker). I am worried that these new schemes will force me to do a
> lot of recoding.
> 
> Currently, my code gets the VOTable response from the SIAP and looks for the
> first table in the first resource. (Else fail).
> 
> In the table, it looks for the UCDs for RA, Dec, URL, and Format, and then
> dereferences all the URLs that have the right values of RA, Dec, and Format.
> 
> Please Pedro and Francois, can you supply an alternate logic that I would
> use if your schemes were to be accepted?
> 
> Roy
-- 
Pedro Osuna Alcalaya

 
Software Engineer
VILSPA Archive Development Team
e-mail: Pedro.Osuna at esa.int
Tel + 34 91 8131314
                                                                                
European Space Agency
VILLAFRANCA Satellites Tracking Station
P.O. Box 50727
E-28080 Villafranca del Castillo
MADRID - SPAIN