SSA-1.1

[Rob Seaman]

There seems to be some notion here of setting a broad precedent for IVOA interfaces.  As Myron says, it surely will be significant for numerous use cases to preserve both case and whitespace.  Can't there be a toggle to select between strict and relaxed string matching algorithms?  There could even be separate toggles/queries for fixing case and/or excising whitespace (and maybe other heuristics).

Also, this is the sort of issue that must have some paper trail of historical discussion in both the astronomical and library science communities.  Perhaps there is already a precedent (IAU, ADS, SIMBAD, NED?) that IVOA can borrow?

Rob
--
On Apr 26, 2011, at 6:04 PM, Myron Smith wrote:

> But Petr ...
> 
> 2) ignore case
> 
> I don't see how on earth you can ignore case in a target name. This introduces many kinds
> of ambiguities, of which variable star designations is only one....
> 
> 
> Cheers - Myron
> 
> 
> ________________________________________
> From: dal-bounces at ivoa.net [dal-bounces at ivoa.net] on behalf of Petr Skoda [skoda at sunstel.asu.cas.cz]
> Sent: Tuesday, April 26, 2011 8:54 PM
> To: Randy Thompson
> Cc: dal at ivoa.net
> Subject: Re: SSA-1.1
> 
>> Just to let you know, all the MAST spectral services support TARGETNAME
>> queries. It
>> assumes no name resolution, so you can submit a query such as:
>> http://archive.stsci.edu/ssap/search2.php?targetname=mars
> 
> Its very nice - but to guess the precise name of the target may be
> difficult. So you have to provide some rules like - ignore the speces,
> ignore the case of letter etc ...
> Moreover some wildcard will help
> 
> As I see (and want to propose as a standard) you have already most of this
> working on your web form of MAST - I tried M81%, M81*, m81* and even
>  *m8*
> 
> lets take
> 
> m8*
> (returns e.g. M87, M81FIELD03 etc)
> 
> and in the SQL it is really
> 
> WHERE  (fes_targname LIKE 'M8%')
> 
> 
> So in fact you do some translation of wildcards to SQL LIKE %...
> 
> If I use SSA directly it works as well
> 
> http://archive.stsci.edu/ssap/search2.php?targetname=*m8*
> 
> It is exactly what I wanted to put in SSA -
> define UNIQE way how to encode the wildcards patterns in TARGETNAME=
> with the respect to permitted characters in URL GET
> 
> Now I know (after reading RFC 2616 and RFC 2396) that probably the only
> one reasonable wildcard characte which does not require escaping in URI is
> the star *
> 
> So what I would suggest in TARGETNAME is to
> 1) strip spaces befre and after string
> 2) ignore case
> 3) the * in TARGETNAME replace by % and call TARGETNAME LIKE '%text%'
> 
> Even if it did not enter the SSA1.1 I suggest to discuss this issue and
> prepare the rules anyway - it may then be added to other discovery
> interfaces using TARGTENAME (including SSA1.X) x>1 or retroactively as the
> recommended way of using OPTIONAL TARGETNAME
> 
> But I still think that we are expected to provide such a standard and
> recommend the TARGETNAME capability of (not only) spectra archives.
> 
> And the MAST case simply shows the practical need to implement this
> (driven by the practical request of scientific community instead of of our
> ad-hoc decision that all celestial objects are accessible by coordinates
> as a unique key ...................)
> 
> 
>> http://archive.stsci.edu/ssap/search2.php?target=M81&resolver=Don%27t%20Res
> 
> Petr Skoda
> 
> *************************************************************************
> *  Petr Skoda                         Phone : +420-323-649201, ext. 361 *
> *  Stellar Department                         +420-323-620361           *
> *  Astronomical Institute AS CR       Fax   : +420-323-620250           *
> *  251 65 Ondrejov                    e-mail: skoda at sunstel.asu.cas.cz  *
> *  Czech Republic                                                       *
> *************************************************************************