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

Re: [open-bibliography] Draft development vocabulary for bibliographic metadata

From: William Waites <william.waites@okfn.org>
Date: Wed, 16 Jun 2010 15:10:00 +0100
Message-ID: <4C18DB38.30309@okfn.org>
To: Ross Singer <ross.singer@talis.com>
CC: List for Working Group on Open Bibliographic Data <open-bibliography@lists.okfn.org>, "rufus.pollock@okfn.org" <rufus.pollock@okfn.org>, Ben O'Steen <bosteen@gmail.com>, public-xg-lld@w3.org
Hi Ross,

I really need to polish the presentation of some of this... That email
was written very quickly before I had to run out of the office. But

On 10-06-16 10:50, Ross Singer wrote:
> This seems like a pragmatic approach to this.  Is there any reason why
> you coined new classes/properties instead of reusing and extending
> existing ones?
> For example, it seems like you could coopt the Work and Manifestation
> classes from http://vocab.org/frbr/core and add the (useful!)
> hasManifestation property.

In the case of FRBR this was to avoid confusion since coopting its
classes means using them in ways that have different meaning.
frbr:Manifestation implies a frbr:Expression whereas obp:Manifestation
doesn't. Making a new obp:Work also happens in places where FRBR would
expect simply a new frbr:Expression.

> Seems like the ISBN/ISSN/LCCN properties could be taken from BIBO, as
> well.  Granted, BIBO doesn't set a domain for these properties (which
> may have been your motivation), but I'm not sure of the accuracy of
> inferring that every LCCN is a Manifestation, (subject authorities
> have LCCNs, as well as archival finding aids:
> http://lccn.loc.gov/92162930), etc.

I think you're right here. There don't seem to be much in the way of
domain/range restrictions on these properties so I don't see why they
couldn't be used.

For the W3 LLD list in the crosspost -- please let me know if this is
inappropriate for the list, might be getting ahead of ourselves a bit.
This is meant specifically in the context of the use case that is taking
a MARC record and turning it into RDF.

As it is, it looks like a MARC record at a minimum consists in:

    * one xyz:Manifestation
          o one dc:publisher
    * one (implied) xyz:Work
          o one or more dc:contributor (or sub-properties like author,
            translator, etc)
          o one or more identifiers, bibo:isbn bibo:issn etc
          o one or more dc:subject from a controlled vocabulary


William Waites           <william.waites@okfn.org>
Mob: +44 789 798 9965    Open Knowledge Foundation
Fax: +44 131 464 4948                Edinburgh, UK
Received on Wednesday, 16 June 2010 14:11:25 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:38:34 UTC