Re: Ontology metadata

Martin,

Thank you for your comments. You raise a good point in wanting to be
able to attach an author or modification date to individual elements in
the ontology. In RDF Schema, this can be done by simply defining a new
property whose domain is a Class or Property. Whether or not we want to
have predefined vocabulary for this is open to debate. We will certainly
consider it before releasing the next version of the requirement
documents.

As for your comment on "View mechansim," I agree that maintaining
updates between parallel ontologies is a difficult problem, but the
problem of maintaining consistent views in the face of changing
ontologies is almost as difficult. As for the note about the ontology
mapping mechanism, the only point is that we might be able find a
suitable mapping method that is identical to creating a view.

Jeff Heflin

Martin Bryan wrote:
> 
> Congratulations on a great first draft for the Requirements for a Web
> Ontology Language. Your statement of requirements is extremely
> comprehensive.
> 
> One of the few areas in which an improvement is to be desired relates to:
> 
> Ontology metadata
> It must be possible to provide meta-data for each ontology, such as author,
> publish-date, etc. The language should provide a standard set of common
> metadata properties. These properties may or may not be borrowed from the
> Dublin Core element set.
> 
> It should be possible to identify the author and update date of an
> individual term within an ontology. Ontologies are rarely the work of a
> single person, and rarely created at a single point in time. You need to be
> able to identify which changes have been applied in each version. This can
> only be done if individual terms carry enough information as to their
> currency.
> 
> I would also suggest you review the suggestion, under View Mechanism, that:
> "This is particularly useful in developing multicultural versions of an
> ontology. Note that this requirement may be satisfied by having multiple
> ontologies and using an ontology mapping mechanism."
> This is certainly not the most efficient way to identify relationships
> between different sets of terms in multilingual ontologies. There are severe
> problems with ensuring the update of parallel ontologies when they are
> maintained separately.
> 
> Martin Bryan
> The SGML Centre, 29 Oldbury Orchard, Churchdown, Glos GL3 2PU, UK
> Phone/Fax: +44 1452 714029  E-mail: mtbryan@sgml.u-net.com
> 
> For further details about The SGML Centre visit http://www.sgml.u-net.com

Received on Wednesday, 27 March 2002 10:57:08 UTC