The tully NBG catalog in RDF

[Ed Shaya]

I was curious how close the IAU vocabulary comes to providing everything 
needed in a typical astronomical catalog to allow sophisticated SPARQL 
queries.  The only way to really know is to actually translate the 
column descriptions into RDF using the SV.  So, I chose Tully's "Nearby 
Galaxies Catalog".  Actually, this is quite a bit heftier than the 
typical table, but by no means the most complicated.

I did not want to muddy it up with owl:Restrictions so I just left them 
out.  This means that in places where it should say things like 
"B_magnitude corrected for [only objects from the] absorption class", it 
is saying "B_magnitue corrected for absorption (ie, the term).  Which 
reads fine, except that it means it uses terms from both an Ontology (to 
create instances of objects with properties) and a Thesaurus (for 
instances of the terms) and this is surely a no-no.
So, a final version document probably will need restrictions and look a 
bit uglier than this.

The terms with either a: or b: are from the SV.  The terms with : are 
locally defined (ie, you just need them when you are looking at this 
catalog).  But, the terms with n: are not in the SV and need to be 
added.  It is not bad at all.  Most of the Measurement classes are 
basically covered.   The SV has none of the properties (eg. has_name, 
has_measurement, has_units, corrected_for, in_band, of, on, at etc). 
And maybe there is no need for these in the SV.  But there are a number 
of useful n: terms as well.  Some may properly be in math and physics 
vocabularies: minor_axis (we have major_axis), ratio, axial_ratio, 
rotational_velocity.

My top choices for inclusion into the SV (aside from the above) from 
this are:
Hubble_Class, isophotal_diameter, reference, altname, name_in_catalog 
(which would include link to catalog).

Clearly we also need units in some format.

Ed
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 7145.rdf.n3
URL: <http://www.ivoa.net/pipermail/semantics/attachments/20071011/b5ba49a1/attachment-0001.ksh>