Karen:
I think I like the first option, because, of course, it also applies to
the RDA vocabularies ... :-)
Diane
Karen Coyle wrote:
> Diane, it sounds like the offending paragraph is this one, right?:
>
> *subject*
> We want to record the *subject* using the Library of Congress
> Subject Headings. Typically, we would indicate the subject with a
> string (e.g., "Islam and Science") together with the vocabulary
> encoding scheme dcterms:LCSH, which identifies the heading "Islam
> and Science" as a member of the Library of Congress Subject
> Headings. Alternately, if the individual terms of the controlled
> vocabularies have already been given URIs as part of the work to
> express existing vocabularies using the RDF vocabulary Simple
> Knowledge Organization System [SKOS
> <http://dublincore.org/documents/2008/11/03/profile-guidelines/#SKOS>],
> then that URI may be used. In this case, the Library of Congress
> subject heading "Islam and Science" has been assigned the URI
> http://lcsh.info/sh85068424#concept. The DCMI property
> dcterms:subject has a "non-literal" range, which means that it can
> support the use of value strings, value URIs, and vocabulary
> encoding scheme URIs as needed."
>
> Would you want that to read: "... has provisionally been assigned the
> URI..."? Or to re-word the sentence saying something like: LCSH has
> been expressed in SKOS using Cool URIs on the lcsh.info
> <http://lcsh.info> site, and therefore we can use the URI http.....
>
> kc
>
> On Tue, Nov 11, 2008 at 6:13 AM, Diane I. Hillmann <[log in to unmask]
> <mailto:[log in to unmask]>> wrote:
>
> My point remains: if we're using lcsh.info <http://lcsh.info> URIs
> as examples, we need to be careful not to imply that these are
> more than what Ed says they are--an experiment. A very nice and
> useful experiment, which we all hope will prod LC into doing LCSH
> officially, under whatever domain. Far as I'm concerned,
> lcsh.info <http://lcsh.info> is a fine domain, but it's Ed's not LCs.
>
> But please, re-read that section without *knowing* that it's an
> example, think of yourself as someone coming to this for the first
> time, and tell me that it doesn't need some kind of "flag."
>
> Diane
>
>
> Pete Johnston wrote:
>
> Ed said:
>
>
>
> If lcsh.loc.gov <http://lcsh.loc.gov> came online tomorrow
> and all the concept URIs at lcsh.info <http://lcsh.info>
> redirected appropriately to lcsh.loc.gov
> <http://lcsh.loc.gov> would that count as persistence?
> What does "maintained properly" mean in this context?
> Isn't lcsh.info <http://lcsh.info> just used as an
> example, not as an instruction to use concepts from that
> concept scheme.
>
>
>
> I strongly agree with Ed here: the LCSH example is being used to
> illustrate the notion of treating a concept-as-subject as a
> resource in
> its own right, and identifying it with a URI, following the
> "Cool URIs"
> conventions.
> I think it might also be useful to include a short section
> about URI
> persistence and the potential role of organisational
> commitment to URIs
> in supporting persistence, but I think that is a distinct issue.
>
> I don't think there is any need to limit the URIs used in the
> document
> to URIs for which there is some sort of currently published
> organisational policy. On the contrary, if anything, I think it is
> important to make the point that any URIs owned by any agency
> can be
> used in RDF/DC metadata - while also recognising that URI
> ownership/persistence may be an issue to be considered.
> Having said that, I'd also add that, depending on the context, the
> weight given to that factor may vary: in some circumstances,
> "This group
> of three people guarantees to maintain ownership of these
> URIs, manage
> them sensibly, and serve representations/descriptions of the
> resources
> for five years" _may_ be sufficient for the purposes at hand.
> Obviously
> that may have implications for the longer term stability of a
> DCAP using
> term URIs based on such a policy, but that doesn't stop it
> being a DCAP.
>
>
>
>
> Oh there's lots you could use: dbpedia [1], geonames [2]
> opencyc [3] and more [4]. But perhaps these don't meet
> your criteria for being real either? I'm not really
> suggesting that GEM vocabs not be used--they seem
> excellent as well.
>
>
>
> Again, agreed. I really don't think there is any shortage of
> terms or
> other resources that can be used in examples, and - within
> reason, given
> that we want to keep the document reasonably concise - it is
> helpful to
> illustrate the breadth of possibilities, and the value of a
> URI-based
> approach, rather than, albeit inadvertently, giving an
> impression that
> only a small subset of "endorsed" URIs, owned by a small number of
> parties, is appropriate.
>
> Cheers
>
> Pete
> ---
> Pete Johnston
> Technical Researcher, Eduserv Foundation
> [log in to unmask]
> <mailto:[log in to unmask]> +44 (0)1225 474323
> http://www.eduserv.org.uk/foundation/
> http://efoundations.typepad.com/efoundations/
>
>
>
>
>
>
> --
> -- ---
> Karen Coyle / Digital Library Consultant
> [log in to unmask] <mailto:[log in to unmask]> http://www.kcoyle.net
> ph.: 510-540-7596 skype: kcoylenet
> mo.: 510-435-8234
> ------------------------------------
|