W3C home > Mailing lists > Public > public-lld@w3.org > December 2010

Re: Modelling library, catalogue, record, and item

From: Owen Stephens <owen@ostephens.com>
Date: Tue, 14 Dec 2010 08:14:06 +0000
Message-Id: <5FDA0F90-07C1-45EB-91BC-4F257FBE53B7@ostephens.com>
Cc: "public-lld@w3.org" <public-lld@w3.org>
To: Jakob Voss <jakob.voss@gbv.de>
One of the things I've hit doing library data -> linked data for the Lucero project at the Open University is how much variety there is in the material we are looking at.

While I can see the advantages of adopting some common sets of vocabulary across these different types  (like DC) if you model (for eg) non-musical recorded sound in the sane way to books, or even musical recorded sound, then you are just creating problems for yourself, and not making any use of the flexibility offered by RDF/LD

In many respects I would have thought that these non-biblographic materials would be low hanging fruit for LD, as they've never fitted well into the MARC representation. 

On 14 Dec 2010, at 07:46, Jakob Voss <jakob.voss@gbv.de> wrote:

> Barbara Tillett wrote:
>> We could also view this with our FRBR glasses on as the things
>> (resources, objects, persons, corporate bodies, families) in our
>> bibliographic universe (whether at a particular collection or library
>> or just available from a creator, publisher, manufacturer, etc.) and
>> the data describing those things (the records and catalogs of
>> attributes and relationships).
> I also looked at FRBR, but it does not help a lot. I can reuse
> frbr:CorporateBody (but also foaf:Organization) and frbr:Concept (but also skos:Concept). Other FRBR classes and properties (frbr:Endeavour, frbr:Work, frbr:Item etc.) require some more analysis of our data.
>> Can't we do better to describe these for the future rather than
>> trying to replicate the catalog and record structures of the past?
>> Those are a big part of our problem just now in trying to break away
>> to a future that enables re-use and sharing/linking of data where
>> there are relationships.
> Sure we should not just put fields of catalog-records into RDF properties, but do some more sophisticated mapping. But I think the "catalog and record structure" is fine, unless you don't understand it as listing of holdings only. In my understanding there is no difference between a catalog and a bibliography - but this distinction between library science and documentation (now information science) has a loooong history.
> Jakob
> -- 
> Jakob Voß <jakob.voss@gbv.de>, skype: nichtich
> Verbundzentrale des GBV (VZG) / Common Library Network
> Platz der Goettinger Sieben 1, 37073 Göttingen, Germany
> +49 (0)551 39-10242, http://www.gbv.de
Received on Tuesday, 14 December 2010 08:19:36 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:27:42 UTC