Firstly thanks to Ross for picking up on the thread and pointing people
at Ian Davis' 2005 presentation on our research site.
> > Talis still doesn't make that abundantly clear. An answer
> would need
> > to be such that it at once grips you by its crispness and
> clarity so
> > you can have no doubt it can sweep MARC away.
> > And even from there to a universally recognized and applied
> standard
> > would still be a long way to go. Which is, however, what we
> need, for
> > interoperability will be key.
We are not necessarily looking to sweep Marc away, in fact the Talis
Platform ingests and preserves Marc, and come to that any other format.
During the ingesting process the data would be mapped in to RDF triple
statements under the control of the owner of the Platform Store. So RDF
statements about a record that is stored, could not only indicate that a
record has a property of type title with a value of Hamlet, but also a
property of type marc21-tag-245 also with a value of Hamlet. These
mappings can then be used in indexing and querying processes. When a
record is thus identified, the builder of the application that is
accessing the store should then have the option to either return the
original record or to construct a record type of their choice from the
RDF data stored about it.
(In a real example 'title' and 'marc21-tag-245' would actually be
defined as URIs in a schema)
This means that systems using the data could infer relationships between
say records that were available in Dublin Core and others in Marc; or
between bibliographic records and author biography records which share
the same creator property.
It would be would be folly to degrade the vast amount of meta and
semantic data preserved in Marc records in a move to another format.
Nevertheless Marc is becoming increasingly imperfect for the uses that
are being demanded of it, and we could gain much value by extracting
some of that value in a way that can be used to relate it to, and enrich
it with, data that is available in other formats.
There are already emerging schemas for FRBR
[http://vocab.org/frbr/core], Dublin Core
[http://dublincore.org/schemas/rdfs/] and others and my colleague Ian
Davis has been looking at the issues around transliterating Marc in to
RDF for some while
[http://iandavis.com/blog/2005/12/marc-transliteration].
As to a universally recognised and applied standard, that is not as
crucial in an RDF world as you might think. Obviously the fewer
standards the better, but there is no real problem relating two data
sets together if for instance one uses 'author' and another 'creator'
providing the user of the data has the ability to assert happiness that
the two properties essentially describe the same attribute.
And finally - you don't really want to 'see' RDF data - its not very
human readable and was not designed to be, but RDF aware software loves
it and can do wonderful things with it.
Richard Wallis
Technology Evangelist, Talis
> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Ross Singer
> Sent: 22 May 2007 15:14
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] Next Gen Catalog and FRBR
>
> The problem with this argument is that it's difficult to make
> a radical change to the infrastructure when the plumbing is
> still all MARC based.
>
> -Ross.
>
> On 5/22/07, Bernhard Eversberg <ev_at_biblio.tu-bs.de> wrote:
> > Ross Singer wrote:
> > >
> > > Well, Richard's examples show the underlying RDF, but, at
> the end of
> > > the day, it's still MARC in the background (or DC or whatever the
> > > institution supplied).
> > And why not RDF?
> > >
> > > The point, though, is the /potential/ of RDF, not the existing
> > > practice in libraries.
> > >
> >
> > Sure, and I do think I understand that much. The theory is
> wonderful.
> > But my question was, where is it demonstrated (after 10 years) that
> > the potential is more than a theoretical one? What we learn about
> > Talis still doesn't make that abundantly clear. An answer
> would need
> > to be such that it at once grips you by its crispness and
> clarity so
> > you can have no doubt it can sweep MARC away.
> > And even from there to a universally recognized and applied
> standard
> > would still be a long way to go. Which is, however, what we
> need, for
> > interoperability will be key.
> >
> > OTOH, quantum mechanics is better than Newton's mechanics although
> > there's no quick and easy eye-opener for its beauty...
> >
> > B.Eversberg
> >
>
Find out more about Talis at www.talis.com
Shared InnovationTM
Any views or personal opinions expressed within this email may not be those of Talis Information Ltd. The content of this email message and any files that may be attached are confidential, and for the usage of the intended recipient only. If you are not the intended recipient, then please return this message to the sender and delete it. Any use of this e-mail by an unauthorised recipient is prohibited.
Talis Information Ltd is a member of the Talis Group of companies and is registered in England No 3638278 with its registered office at Knights Court, Solihull Parkway, Birmingham Business Park, B37 7YB.
Received on Wed May 23 2007 - 06:18:29 EDT