Different page numbers, etc.

In short, a new Instance.

It might not *always* be correct (perhaps it does boil down to a binding 
difference only), but from what we're seeing creating Instances based on 
differing ISBNs is right nearly all of the time.  And, certainly at this 
stage of experimentation, we've chosen to play the averages.

When you start to look closely at this on field, a host of questions emerge.

Stipulating that a book has been bound for libraries, and has its own 
ISBN, can we safely assume it is the hardback version that has received 
this treatment?

Even if the answer to the above is yes, I would still feel wary 
considering treating one type of ISBN in a special way.  There are loads 
of different types of bindings.

Can we always be certain that, within these different binding types 
(identified by ISBN), the physical characteristics (page numbers, e.g.) 
of the book are the same?

Can we always be certain that the one without a qualifier is the 
hardback version?

We haven't attempted logic in the code to deal with multi-volume ISBNs.

The ISBN can be a decent way to identify different Instances of a Work, 
but it contains a fair amount of variation that we've not yet attempted 
to tackle.  At this time, that is why simple splitting rules are 


On 01/25/2013 09:33 AM, Trail, Nate wrote:
> Too bad it's not that simple (publisher binding vs library binding). The second ISBN may be for the paperback, with new introductory material, updates, corrections, better indexing, etc.
> Nate
> ---------------------------------------------------------------
> Nate Trail
> ---------------------------------------------------------------
> Network Development and MARC  Standards Office
> Technology Policy Mail stop 4402
> Library Services
> Library of Congress
> 202-707-2193
> [log in to unmask]
> -----Original Message-----
> From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of J. McRee Elrod
> Sent: Thursday, January 24, 2013 6:23 PM
> To: [log in to unmask]
> Subject: Re: [BIBFRAME] Bibframe and translations from MARC
> Kevin Ford asked:
>> If ISBNs are used as "splitting" points - meaning that two BIBFRAME
>> Instances would be created from the one MARC bib record in the above
>> example - where does the LCCN go?  Neither Instance? The first
> Instance created from splitting the ISBNs from the 020? Both
>> Instances?
> A different binding does not a new manifestation (aka instance) make, so both ISBNs should be associated with a single instance.
> We don't want to get into trying to disguish between publisher binding and library rebinding.  The content is unchanged.
> Let's not confuse manifestations with items.
> If contrary to bibliographic logic two records are created. the LCCN should be associated with both.
>     __       __   J. McRee (Mac) Elrod ([log in to unmask])
>    {__  |   /     Special Libraries Cataloguing   HTTP://
>    ___} |__ \__________________________________________________________