W3C home > Mailing lists > Public > public-lod@w3.org > August 2009

Re: Top three levels of Dewey Decimal Classification published as linked data

From: Ed Summers <ehs@pobox.com>
Date: Mon, 24 Aug 2009 09:48:46 -0400
Message-ID: <f032cc060908240648p72739100y99835b0b758339b5@mail.gmail.com>
To: public-lod@w3.org
Hi Michael:

On Fri, Aug 21, 2009 at 5:04 PM, Panzer,Michael<panzerm@oclc.org> wrote:
> ... it could be argued that the concept
> http://id.loc.gov/authorities/sh85137241#concept really refers to the
> idea of trees in general, or (probably more in line with common opinion)
> to the idea of trees as framed as an element of the controlled
> vocabulary LCSH.

Yes, that is the intended meaning.

> This thing has (following the definition given in
> webarch) as a non-information resource no representation that can be
> transmitted over the web. So this concept might have broader and
> narrower terms, but can it as a thing have a modification date? Can it
> have a history note? (A car can have a color, but only the description
> of that car can have an dct:issued date or a dct:language.)

Can't the Library of Congress change its mind, or refine its notion of
what Trees are? For example when cataloging a book about a new type of
tree that has been found on the moon, wouldn't it be ok for a new
concept Moon Trees be created, and to assert the relationship between
Moon Trees and Trees? When this happens isn't it ok for dct:modified
to be used to record when LC updated its notion of Trees? And isn't it
ok for LC to use skos:historyNote to record some statement about the
change?

> Tennis and Sutton [1] have suggested (for somewhat different reasons) to
> split skos:Concept into skos:Concept and skos:ConceptInstance to clarify
> what kind of skos:Concept we are talking about; a "thing" or its
> instantiation as an information resource.

Thanks for that reference. It is currently printing out for the train
ride home :-)

Personally I've found the debate around what is, and is not an
information resource to be quite subtle and confusing. Cool URIs for
the Semantic Web [1] was extremely helpful to me because it provided
several patterns to follow when building semantic web applications.
But this document itself doesn't really answer whether or not a SKOS
concept is a "thing" or a "document". The SKOS Primer references Cool
URIs for the Semantic Web in the discussion of skos:Concept resources
[2]. In hindsight, perhaps it should have been more explicit? I see a
lot of value in trying to follow patterns in use elsewhere to build
community and consensus. But this is also a new area of development,
so experimentation, and testing the boundaries is also good. Perhaps
this could be a good thread to kick off on the skos discussion list?

I remain curious why you chose to go through the effort of minting
URIs for things like <http://dewey.info/class/37/> which 303, but you
haven't used these URIs in your RDF assertions. It seems like you've
followed the patterns in the Cool URIs for the Semantic Web, but then
not used those URIs in your RDF descriptions. If your position is that
your skos concept resources are documents then why bother minting the
thing URIs at all?

//Ed

[1] http://www.w3.org/TR/cooluris/
[2] http://www.w3.org/TR/skos-primer/#secconcept
Received on Monday, 24 August 2009 13:49:35 UTC

This archive was generated by hypermail 2.3.1 : Sunday, 31 March 2013 14:24:23 UTC