SSAP and images

[Guy Rixon]

Anita,

I don't know whether SSAP is insufficiently defined. I do know that  
it is quite hard to work out what is allowed, what is recommended and  
what is required, so very easy to implement wrongly. In this respect  
it was easy to get details wrong on SIAP 1 (I did it up several  
times) but, on rereading the spec it was always clear what should  
have been present. SSAP is much more complicated and is defined less  
directly, with more use of external data models.

Cheers,
Guy

On 2 May 2007, at 09:09, Anita M. S. Richards wrote:

> On Tue, 1 May 2007, Guy Rixon wrote:
>
>> Anita,
>>
>> by my reading of the SSAP spec, the position encoding is part of  
>> the Spectrum data model:
>>
>> Coverage.Location.Sky.Value pos.eq RA and Dec deg, R
>> (or sexagesimal string)
>>
>> If I understand this correctly, each column of the VOtable  
>> returned by SSA is a node in the Spectrum data model, with an  
>> associated Utype. This position node is defined as a coordinate  
>> pair, so the column isn't a scalar, numeric type. It would have to  
>> be either a string or an array of float or double. Would the array  
>> form work for you? I guess the string form covers the case where  
>> the coordinates are sexagesimal.
> -----
> On Tue, 1 May 2007, Doug Tody wrote:
>
>> Hi Anita -
>>
>> SSAP and SIAP both return positions as type double.  I think you  
>> need to address this at the level of applications integration.   
>> (Also, this just goes to show, that although we may know the type  
>> of an object, that doesn't say much about what to do with it).
>
> Thanks to all for the comments.
>
> I have just done a quick comparison, and in fact some SSAP votables  
> do have RA and Dec (in decimal degrees) as type Double, so that the  
> positions can be easily plotted in TopCat, as can most SIAP  
> VOTables.  So I think that there are two issues:
>
> Firstly, that SSAP and SIAP tables can have position coordinates in  
> a form which can be plotted by TopCat, but don't always.  I don't  
> know whether this is because the standard is insufficiently  
> defined, or whether it is OK but open to misinterpretation (those  
> pesky users again) or because some archives get it plain wrong (as  
> a data provider, I know I commit all three sorts of error...).  I  
> _think_ that is covered by what Guy says - and, following from  
> that, if coordinates are in decimal then they are an array of two  
> doubles or floats; if in sexagesimal, then strings?.
>
> Secondly, it is up to the tool authors to provide a means of  
> overplotting and I will take that up separately.
>
> Thanks very much
>
> Anita
>
>
>
>
>
>
>> On 1 May 2007, at 17:07, Anita M. S. Richards wrote:
>>
>>> Hi people,
>>> Today we were doing a demo at the E-Science North West open day  
>>> in Manchester and someone asked 'can you display the positions of  
>>> spectra on top of images'.  So I used AstroScope to get some  
>>> spectra and images and sent an image to Aladin, and sent the SSAP  
>>> VOTable to Aladin and TopCat. Two problems emerged:
>>> The position is given in decimal degrees but type is string.  If  
>>> I change to float, TopCat will plot the positions.
>>> However, even then, Aladin will only offer to display it in a  
>>> spectral tool - it won't plot the positions.
>>> Probably we ahve discussed this and there are good reasons for it  
>>> not working? Or is it something to do with retrieving SSAP stuff  
>>> via AstroScope? But otherwise, it seems easy to fix and would  
>>> please users.
>>> Thanks
>>> Anita
>>> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -  
>>> - - - -
>>> Dr. Anita M. S. Richards, AstroGrid Astronomer
>>> MERLIN/VLBI National Facility, University of Manchester, Jodrell  
>>> Bank Observatory, Macclesfield, Cheshire SK11 9DL, U.K. tel +44  
>>> (0)1477 572683 (direct); 571321 (switchboard); 571618 (fax).
>>
>>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Guy Rixon.vcf
Type: text/directory
Size: 16315 bytes
Desc: not available
URL: <http://www.ivoa.net/pipermail/dal/attachments/20070502/cacc720e/attachment-0001.bin>
-------------- next part --------------