-----BEGIN PGP SIGNED MESSAGE-----
Am 19.09.2013 17:42, schrieb Ford, Kevin:
> Dear Juha,
> About your first paragraph:
>> 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....
> -- Not only does this represent something for a test case but the paragraph
> represents a use case. The question I see is: Is it possible to take BIBFRAME
> holding/item data and parse it into your current system's framework. I'm not
> going to say this would be a requirement (I've not been keen to restrict
> ourselves to what is possible today versus what is possible in the future) but
> it is worth considering. It is worth noting that a Holding resource would have a
> relationship with an Instance, not a Work. You mentioned "work" repeatedly, so
> there may be a misunderstanding there.
The more general question appears to be one about the appropiate place
for this situation in the bibframe model (and I don't think that FRBR
will ever have an answer for this):
Several prints have been individually bound together at some distant
time in the past. Not only for "holdings" purposes of the /item/ but
also for recording provenance information or describing the physical
and artistic features of the binding this is to be considered as one
physical /object/ of high cultural interest (worth "catalogueing" in
its own right).
Thus the compound object itself should be represented by more than a mere
BF annotation, but considering it as a BF instance there is the
requirement to denote exactly one corresponding BF work. Such a
retrospectively construed "work" from singular aggregation of individual
instances is very counter-intuitive (and definitely not a "creative"
work) and furthermore provoces the question of the exact kind of relation
between this "compound work" and its individual constituent works
(those, whose instances had been bound together) and how to model it.
This might become easier to grasp, when general objects (collection items)
are dealt with by Bibframe: They often have physical components but
must not be decomposed, dissasembled, partially summoned in reality.
(Or cannot, if you consider a scratchbook with individual works on
recto and verso of a certain page).
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.13 (Cygwin)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
-----END PGP SIGNATURE-----