The problem here is, that not all possibilities of BIBFRAME is used.
With the users perspective, the metadata about Work and Instances shall be used for presentation in an user interface.
To make it simple: The user search for Hamlet and the data base have to works:
WORK 1: Hamlet
WORK 2: Shakespeare's complete writing
With the 1:1 principle from the Bibliographic Framework as a Web of Data: Linked Data Model and Supporting Services (November 21, 2012) you can only present for the user:
WORK 1: Hamlet - with a link to first page of the instance
WORK 2: Shakespeare's complete writing - with a link to first page of the instance (where Hamlet is on page 512)
But the reality is a bit more complex:
WORK 1 have an instance, and WORK 2 have one instance - but inside this instance you can find another instance Hamlet. And this Instance inside an Instance have a Instance-to-Work relation to WORK 1.
If we forget the 1:1 principle, the answer can be:
With link to page 1 Instance of WORK 1
With link to page 512 in Instance of WORK 2 - which also is Instance of WORK 2 with page interval 512-589
Of course you can present relevant pages according to search terms, but you can't express that pages 512-589 of the instance of WORK 2 also is an Instance of WORK1.
So: If we see a multivolume instance as an container for instances, we can use the relations to show that WORK 1 have instances both as individual instances and as parts of a multivolume - and the multivolume on the same time as a whole is instance of another work.
And of course the two works have a relation: one work as part of the other work.
We have a problem with the November 2012 document - but not with the BIBFRAME Vocabulary - all the needed relations is part of the Vocabulary.
Det Kongelige Bibliotek
Nationalbibliotek og Københavns Universitetsbibliotek
Chefkonsulent | Adviser to the Director
Det Kongelige Bibliotek | The Royal Library
Nationalbiblioteket - Stab | The National Library of Denmark - Staff
P.O. Box 2149 | DK-1016 København K
tel +45 9132 4366 | Fax +45 3393 2218 | [log in to unmask] | www.kb.dk
Besøgsadresse | Visiting address | Søren Kierkegaards Plads 1
Leveringsadresse | Delivery address | Christians Brygge 8 | 1219 København K
EAN 5798 000 79 52 97 | Bank 0216 4069032583 | CVR 28 98 88 42
IBAN DK2002164069032583 | Swiftcode DABADKKK
Fra: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] På vegne af Nicolas Prongué
Sendt: 14. oktober 2014 13:54
Til: [log in to unmask]
Emne: Re: [BIBFRAME] Conceptuality of Work in Multi-volume Resources
"If a library is collecting an entire complete manga series and cataloguing it as such rather than individual volumes, then I think having the series act as a Work with each volume being an Instance is the logical conclusion."
As I understand it, I would say that the whole series (published by Publisher XYZ in English for example) is a BIBFRAME Instance. The various volumes would be registred as Holdings. And the series, regardless of any languages or publishers, would be the Work.
Haute école de gestion de Genève
From: Senan Kiryakos
Sent: Tuesday, October 14, 2014 3:33 AM
To: [log in to unmask]
Subject: Re: [BIBFRAME] Conceptuality of Work in Multi-volume Resources
Thanks for the replies (and to the split-off discussion with Reinhold and
Minjie, FRBR Document 3.3 is just what I'm talking about, at least in regards to FRBR. Looking for a similar solution in BIBFRAME is where I am at the moment.
With regards to the conceptuality level of a 'Work', I understand that BIBFRAME's has less of a 'sky is the limit' view than FRBR, which is a good thing in my opinion. Going back to my One Piece example, in FRBR the Work level for One Piece could act as an umbrella entity for the manga, TV series, etc., but in BIBFRAME this overall conceptual level is not necessarily present, with connections between resources being made with relationships instead.
I think the core issue in the manga case is how resource specific should the BIBFRAME Work level represent. Perhaps the 'right' answer may depend on an institutions resources and their cataloguing practices. If a library is collecting an entire complete manga series and cataloguing it as such rather than individual volumes, then I think having the series act as a Work with each volume being an Instance is the logical conclusion. BIBFRAME certainly would allow each volume to be represented as a Work, with relationships being made between each one, but I don't know if this is a better solution (both conceptually and practically with regards to cataloguing practices.)