Re: Open Library and RDF

Would it be possible for me to give a presentation to the joint session in
I am involved in the development of an RDF representation of the International
Standard Bibliographic Description (ISBD). ISBD is a mix of bibliographic
attributes (RDF properties) with a generic domain of (bibliographic) Resource
(unlike the WEMI model in FRBR and RDA), content guidelines (like RDA), and
display format based on punctuation (unlike FRBR or RDA). Preliminary
registration of RDF properties can be found in the Open Metadata Registry. ISBD
specifies repeatability and mandatoriness constraints (unlike RDA, which just
recommends "core" properties for each of WEMI) and sequence of elements when
displayed (along with punctuation). We have decided that the best way to model
all this is with an ISBD application profile (and XSLT or equivalent for
generating the punctuation). The ISBD Review Group is assisting me with expenses
to attend the Pittsburgh meeting ... and I'll be submitting a use case to the
LLD XG real soon now.
I'd like to think of me and my colleagues as gamma-testers of the application
profile approach.
At some point in the future, probably under the aegis of the newly-approved (but
not-yet formally constituted) IFLA Namespaces Technical Group (more news from
IFLA!), we'll attempt to relate ISBD, FRBR/AD/SAD, and (as a result of stronger
links with JSC) RDA, 'cos there's huge overlap between properties.
(And a reminder that the third goal of the DCMI RDA Task Group is the
development of an RDA application profile - as far as I'm concerned, that is
still our intention, and now that JSC has brought RDA to the market-place, I'm
hoping we can resume progress ...)
I'd also be willing to work with Marcia (of course :-) on your proposal for a
presentation on the subject domains, which I consider to be the most significant
area where libraryland can contribute to linked-data and the core of the
Semantic Web. And libraryland really does need advice and support from
ontologyland; I suspect "further work" is indeed needed.
ps I've already updated Karen's wiki page on library data and added more
non-U.S. stuff - and mentioned the ISBD application profile work.

On 16 August 2010 at 20:23 Thomas Baker <> wrote:

> Marcia,
> On Mon, Aug 16, 2010 at 11:51:13AM -0400, Marcia Zeng wrote:
> > On the other hand, due to its high level super-class/property
> > nature, there are still practitioners in the library
> > community who were/are/will be reluctant to accept this model
> > (specifically because frsad did not model for pre-framed
> > frbr Group 3 entities (i.e., concept, object, event,
> > and place)).  Our approach is to allow who would like to
> > keep such differentiating to develop application profiles,
> > for example, under frsad:Thema they could differentiate thema
> > 'type' according to Group 3 or other ways.  We believe other
> > communities and subject domains (e.g., art, medical, business,
> > etc.) would have very different 'type' of themas from Group 3.
> Marcia, would you be willing to present the idea of "application
> profiles" for subject domains in the Friday afternoon joint session
> in Pittsburgh?  It would be very interesting to hear how formally
> (or not) the notion of application profile has been defined for
> this use case.  Perhaps we could talk about whether the requirements
> for this type of application profile are significantly different from
> requirements for descriptive-metadata application profiles?   Is it
> clear to the frsad community how an application profile should be
> constructed, or is this an area where further work is needed?
> > The FRSAD model and approach have received strong support
> > from IFLA (a long story...).  My point is that, in the FR*
> > family, this commitment to the sharing and reuse within and
> > beyond library sectors is very determined.  I hope that Gordon
> > will lead a good reconciled solution very soon.
> That's wonderful to hear! :-)
> Tom
> --
> Thomas Baker <>

Received on Monday, 16 August 2010 20:39:27 UTC