Libraries for working with STC?

Noel Winstanley noel.winstanley at manchester.ac.uk
Tue Aug 12 06:51:59 PDT 2008


Hi.
Three weeks ago I sent message quoted below.

>
> On 21 Jul 2008, at 17:28, Noel Winstanley wrote:
>> Hi,
>> In voexplorer, I'd like to present resource coverage information  
>> in an astronomer-friendly manner.  The coverage of a service is  
>> described in a registry resource by a block of STC/x.
>>
>> Does anyone know of a library / xslt / webservice that can convert  
>> STC/x into something (such as STC/s) that is more readable?
>>
>> I'd hoped to find an xslt stylesheet that converted from STC/x to  
>> STC/c (to my mind, surely is required by the standard to define  
>> the equivalence of the two representations), but I've had no luck  
>> so far.
>>

I think I should feed back to the lists what I've found:

Well, I've not managed to find a solution for the problem described  
above (I need a library or service I can call from a Java desktop  
application).

In fact, no-one replied to my request on-list.

Off-list, I was informed of two pieces of software that process STC:

    * Starlink AST  (http://www.starlink.ac.uk/ast ) is a C library  
for coordinate conversion which implements an early draft of STC.  
AFAIK, there's no plans to support the STC standard.

    * The Footprint Service at http://voservices.net/footprint/  
allows the user to upload a STC description and then download it in  
different formats. However, this functionality doesn't seem to be  
programmatically callable as a webservice. I've asked whether the  
implementation could be open-sourced or shared in some way, but  
haven't yet heard back.

Off-list, I also received a number of very negative responses.  
Typical quotes include:

> Welcome to VO world, where just spec exists and
> nobody cares about implementing them.


and

> Hah!  You'll be lucky.  STC/anything wasn't distinguished by  
> multiple implementations.  It was distinguished by having no  
> complete implementations at all.  That's what standards are for, eh?

Other replies were even more emphatically pessimistic.

So, I'm hardly encouraged! Now, I could dive down the rabbit hole and  
try to implement my solution from scratch  - but I'm surprised that  
there's been so little implementation of STC so far, considering the  
central place it takes in the IVOA standards (as re-emphasized in  
sect 3.4 of the 2008 IVOA roadmap).

Considering the negative replies,  I'm wondering whether there's some  
technical / software engineering barriers to implementing STC parsers  
and transformers. Can anyone else share their experience of even  
_attempting_ to work with STC?

Finally, I spotted in the 2008 IVOA roadmap (section 2.4) that one of  
the activities for the DM workgroup this year is to

> 1) Improve and facilitate the uptake of the existing IVOA  
> specifications: STC,
> Characterisation and Spectrum data models.
> 	*  Develop a library for STC classes
>

Which sounds like an excellent idea. Could I ask who it is that's  
working on this, and what the planned implementation language and  
functionality is to be? If it looks like it'll solve the problem  
described in my original email, I"d be glad to assist.

cheers
Noel Winstanley

---
http://wiki.astrogrid.org/bin/view/Main/NoelWinstanley
Senior Developer, AstroGrid Project.
University of Manchester





More information about the dm mailing list