UCDs, metadata and AstroOntology

[Tony Linde]

> If we agree that each column needs a metadata description 
> then where else do we store it?

Does it matter? As long as it accompanies the data (or is easily
accessible), is in a fixed position (eg a section of the VOTable) and
programs know what to do with it (eg provide clickable links from column
name into appropriate place in metadata description), then the column
name need only be reasonable in the context of the table and not in any
wider sense.

I think we need to decide whether UCDs are meant to be column names or
metadata. This will determine how they are developed in the future.
Trying to make them mean both things will make it difficult to decide
their development. 

>From what I've seen, it seems that people want them to be metadata with
relatively simplistic semantic content. In that case, let's forget the
idea of creating rich and meaningful column names and concentrate on
what metadata is required for catalog, table and column data and then
modify VOTable to carry that metadata. CDS can then incorporate the same
ideas into their tools and the column names themselves can be simplified
(or left as they are if that's what is decided).

Cheers,
Tony. 

> -----Original Message-----
> From: owner-semantics at us-vo.org 
> [mailto:owner-semantics at us-vo.org] On Behalf Of Anita Richards
> Sent: 17 October 2002 12:57
> To: semantics at us-vo.org
> Subject: RE: UCDs, metadata and AstroOntology
> 
> 
> 
> > 
> > > ... As I understand it, the UCDs _are_ the meta data
> > > for each column but they are dynamic, ie built up out of 
> > > atoms from the column or the header or possibly higher levels 
> > > depending on the query.
> > 
> > I'm not sure I understand what you mean by them being 
> dynamic. Once a 
> > UCD is assigned to a column, it is fixed, n'est-ce pas?
> 
> I was using 'UCD' to mean metadata built up out of atoms, and 
> I ws thinking that if some of the metadata is in the header 
> and some is identifiying column labels then the relevant UCD 
> for a particular search might contain more or less atoms 
> strung together.  But, yes, the individual atoms have to be 
> fixed I hope!
> 
> ......
> > 
> > As you said above, UCDs are metadata, not column names, so 
> why not use 
> > the column names as the column names and store the metadata 
> separately 
> > in a form that makes it easy to access and use.
> 
> Because the original column names are often very obscure?  Or 
> are we at cross purposes, is Tony assuming that we have 
> already replaced every original column label that says 
> 'brightness' or 'Mags' or ??? with one saying 'PHOT"? etc?  
> If we agree that each column needs a metadata description 
> then where else do we store it?
> 
> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
> Dr. Anita M. S. Richards, AVO 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). 	
> 
>