Print

Print



On 9/18/13 2:31 PM, Trail, Nate wrote:
[log in to unmask]" type="cite">

I think there are lots of times when works are bound with other works for good reason (an editor selects favorite short stories?),


Nate, Diane is not talking about items published together, but those that end up in the same binding, usually for reasons having to do with their physical characteristics (e.g. two small pamphlets). So they don't constitute a "work" -- their relationship is physical, not bibliographic. Bibliographically, they are two works that happen to be in the same physical package. They cause great problems for inventory and circ systems, because some systems don't allow two different works to have the same item identifier (barcode). But together they do not constitute any "workness." (This is a practice from the past that I hope is less used today.)

kc


[log in to unmask]" type="cite">

and that would be cataloged as a new Work, with some additional properties such as the editor and notes about the compilation etc. So we would already have a model for those; and "accidentally" being bound together could be modeled in the same manner.

 

Nate

-------------------------------------------

Nate Trail

-------------------------------------------

LS/TECH/NDMSO

Library of Congress

202-707-2193

[log in to unmask]

 

 

 

From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of Diane Hillmann
Sent: Wednesday, September 18, 2013 5:04 PM
To: [log in to unmask]
Subject: Re: [BIBFRAME] Holding as Annotation (was RE: [BIBFRAME] BIBFRAME Annotation Model Draft 2)

 

Juha is absolutely correct in his analysis. We had a similar problem dealing with the traditional library vendors who were looking for a simple solution that didn't solve the problem they described, but instead created new problems. For the history, see: http://www.loc.gov/marc/marbi/1999/99-02.html and http://www.loc.gov/marc/marbi/dp/dp116.html.  The most important point of these discussions are the inherent differences between bibliographic relationships and physical relationships, which should not be confused.

 

On Wed, Sep 18, 2013 at 4:17 PM, Ray Denenberg <[log in to unmask]> wrote:

Hi Juha,

 

From: Juha Hakala

> If multiple instances of different works have been bound together into

> a single book, they all share the same item record in our present

> bibliographic database. If BIBFRAME conversion follows this model (and

> other solutions will not be practical to us), we'll have one Annotation

> which annotates multiple BIBFRAME works.

 

If multiple instances of different works have been bound together into a single book, would you not create a new Work?

 

No, I think that's the wrong solution.  The relationship to bibliographic works is different for each physical item in a bound-with package. New "works" created to reflect the physical package will be unusable and confusing for others.

 

>

> … Finnish libraries

> may have hundreds of copies of a single instance of a work ....The current practice is

> to store in the union catalogue just the library codes of the

> institutions which own a copy / copies of the resource, and check from

> UC the OPAC circulation statuses whenever necessary. When UC harvests

> the data from participating libraries, it would be necessary to check

> Annotation and delete Holding information. Such filtering would not be

> necessary if Holding data is not mixed with Annotation.

 

Hopefully others will weigh in, but my first reaction is that this is something that a SPARQL query could handle effectively.

 

 

I agree strongly with Juha's points, and wonder why it's necessary to think up technological 'solutions' to issues with relationships within a model? If we've learned anything in our many decades of bibliographic data experience is that not paying attention to such issues creates very messy data, expensive to fix and almost impossible to re-use effectively. SPARQL queries are useful things but hardly a Ginsu knife solution for everything.

Diane


-- 
Karen Coyle
[log in to unmask] http://kcoyle.net
m: 1-510-435-8234
skype: kcoylenet