RE: Content vs. Carrier

Karen,

> I, too, have no doubt that for some functions we will exchange records
> -- for example, for cataloging purposes, we may have a record that
> combined the essential data of the Work, Expression, and Manifestation
> (or some future version thereof).

The way I see it, "record" is just an old fashioned term for what Web standards call "information resource" or "representation" depending on how we choose to look at it. Take a MARC record for example. They can be represented in a variety of media-types including application/marc, application/xml, application/xhtml+xml, application/octet-stream, etc. Each representation can be separately identified and deliverable using HTTP without assuming the information is stored in any particular way. An important "trick" is to coin a separate URI to identify the record as a generic document. It's easy to believe this generic resource is what librarians mean when they refer to a "record".

http://www.w3.org/2001/tag/issues.html#genericResources-53 
http://www.w3.org/2001/tag/doc/alternatives-discovery 

> However,  I think that our records will not be uniform in the way they
> are today. I can imagine that for many functions records may be
> created "on the fly" from a variety of resources. This is where tying
> the provenance to the record falls down. Although I guess that depends
> on what you mean by record... but a bibliographic description with
> authors, titles, subjects, publishers could be made up of many
> resources, each with relevant administrative data.

The triples used to express "bibliographic description" can be coupled with a URI that identifies their source (called a quad). We should encourage the use of generic resource URIs for this purpose since they satisfy human, machine, and semantic use-cases.

> The big question in
> my mind is what is the ideal level at which to record that
> administrative data? Trying to do so on a statement-by-statement basis
> is probably too costly and unwieldy. I suspect we will arrive at the
> best point through some trial and error.

If we created a "record" ontology, we could use it to define concepts needed to express "administrative data". Quads can do the rest.

Jeff

> Quoting Antoine Isaac <aisaac@few.vu.nl>:
> 
> 
> > In the LLD approach a record (which is an information resource)
> without
> > bibliographic data (which described non-information resources, even
> > though in the "real world our books may be information resources but
> > containing other data ;-) ) is meaningless, while bibliographic data
> > could still be used independently of the record that initially
> > contained it.
> 
> Yes, I agree, we have to assume that bibliographic data will be used
> outside of any context that we provide for it. But perhaps our own
> uses will retain the provenance, even if others do not?
> 
> kc
> 
> 
> --
> Karen Coyle
> kcoyle@kcoyle.net http://kcoyle.net

> ph: 1-510-540-7596
> m: 1-510-435-8234
> skype: kcoylenet
> 
>

Received on Monday, 26 July 2010 21:40:35 UTC