Hi Jonathan,
In SeaDataNet we've pretty much accepted that more than one vocabulary will be involved and that we will have to build maps between them. For the types of data I foresee going into CF then the existing semantics (Standard Name plus Cell method) should cover us, but we are also using an ASCII format to handle other types of data, particularly biological data. Sometimes the biological data will have parameters such as salinity in there and what I want to do is hook up the salinities in a little ASCII file labelled using say the BODC PUV with salinities in a CF file. The way I'm thinking of doing this is following the Semantic Web methodology by setting up an RDF triple of the form:
URI in ASCII SameAs URI in CF
in a resource such as an OWL file mounted on an internet server.
The exact form of the SeaDataNet URIs is yet to be finalised, but they woruld be a string along the lines of 'URN:NDG:Term:P011:25:PSALST01' or '
http://vocab.ndg.nerc.ac.uk/Term/P011/25/PSALST01'. Means nothing to a human, but means a lot to a Semantic Web aware software agent. Whist these strings could be built from the Standard Name using a bit of database lookup it's much easier to have them stored 'ready made' as strings in an attribute called something like 'SeaDataNet_URI' for each parameter.
Is that any clearer?
Cheers, Roy.
>>> Jonathan Gregory <j.m.gregory at reading.ac.uk> 11/29/2006 5:23:43 pm >>>
Dear Roy
> Next week I am about to recommend to the SeaDataNet project that the data
> transport protocol for gridded, and possibly other data, be CF-compliant
> NetCDF.
That's good.
> The parameter semantics will be incorporated as a Standard Name plus Cell Method. These are currently stored as terms, which can be converted to the URIs we will need for the semantic maps, but they are not permanent and should they change the maps will start to break. What I would like to be able to do is to store URIs as attributes within the CF file. Note that this is in addition to the standard CF parameter labelling, not as an alternative.
> Is this permissible and is there a recommended way in which it should be done?
I don't fully understand what you mean about terms and semantic maps. Can you
give an example? It is certainly permissible to store any other attributes in
the file.
Cheers
Jonathan
_______________________________________________
CF-metadata mailing list
CF-metadata at cgd.ucar.edu
http://www.cgd.ucar.edu/mailman/listinfo/cf-metadata
--
This message (and any attachments) is for the recipient only. NERC
is subject to the Freedom of Information Act 2000 and the contents
of this email and any reply you make may be disclosed by NERC unless
it is exempt from release under the Act. Any material supplied to
NERC may be stored in an electronic records management system.
Received on Wed Nov 29 2006 - 10:43:50 GMT
