David,
On 2006 Aug 14 , at 18.25, David Berry wrote:
>> The link to units (getting back to that, David) is that when
>> assembling and using this information, I really couldn't see a place
>> for the units information which is in the VODataService element
>> above. The statement "USNO-B's 'ra' column is a type of pos.eq.ra"
>> is true independently of units. Once I've established just what this
>> USNO-B column is supposed to be (aha, an RA!), then I'm going to have
>> to discover what units the data there has, in order to actually read
>> it.
>
> What about all the other metadata needed to interpret the RA value?
> Things like, reference system (FK4, FK5, ICRS, etc), equinox, epoch...
> Without these things, it's still not possible to convert between two
> general RA values (indeed, even *with* these values, it still would
> not be
> possible unless you also had the associated Declination value).
Yes, you certainly need more metadata in order to convert
coordinates. But before you can convert the values in the RA and Dec
columns in a table (using the other metadata), you must first _find_
the RA and Dec columns (or spectral columns, or...).
The situation I'm aiming to address is where TOPCAT or GAIA, say, is
given a file in FITS or VOTable format, from who-knows-where, and
asked to display it. If it's a FITS file, then the headers will
contains all the standard FITS WCS information -- all the metadata
required to do all the transformations that are required, including
units information.
I haven't forgotten Mark's description of the problems of column-
spotting. However, in the example I gave, the Utype <ivo://roe.ac.uk/
DSA_USNOB/TDB#ra> was essentially a resource-specific nonce Utype
formed from the table ID and column name, so that even if there were
no Utypes explicitly present in the file, you could formulate such a
utype and see if your local reasoner had been fed any information
about it (how it's fed that information is of course another story).
[Utype, UTYPE, uType, bah!]
And I forgot to mention the other thing: the query
prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#>
select ?u
where {
?u rdfs:subClassOf <http://cdsweb.u-strasbg.fr/UCD/
words#pos.eq.ra;meta.main>
}
allows you to go the other way, and get all the specialisations of
pos.eq.ra;meta.main:
http://cdsweb.u-strasbg.fr/UCD/old#POS_EQ_RA_MAIN
http://cdsweb.u-strasbg.fr/UCD/words#pos.eq.ra;meta.main
ivo://roe.ac.uk/DSA_USNOB/TDB#ra
Thanks also to those who pointed at column-UCD maps. I've tried
getting these out of the Vizier SOAP services. That partly works,
but what you get back from their describe-a-catalogue service is a
sliver of SOAP wrapping an undocumented text file, so parsing it
involves a certain amount of guesswork.
See you,
Norman
--
------------------------------------------------------------------------
----
Norman Gray / http://nxg.me.uk
eurovotech.org / University of Leicester, UK
