Subject Identifiers for FRBR Topics

Good afternoon.

This message is the result of a thread that began on the Ontopia mailing
list but is really more appropriate to this list. I'm included an edited
version of two messages I sent into the Ontopia list, updated based on a
reply from the editor of the FRBRoo specification [1], a joint CIDOC/IFLA
project effectively representing best-of-breed work relating to
interoperable FRBR standards.

For those outside of the library community, FRBR (typically pronounced
"fur-burr") is the name of a library standard called "Functional
Requirements for Bibliographic Records" ([2] for a description; [3] for
the actual specification).

In a nutshell, FRBR is a way of expressing the natural hierarchy of
entity relationships inherent in a library bibliographic record, e.g.,

     Entity:         Content:
     -------         --------
     Work:           Moby Dick (as an abstract concept)
     Expression:     Ukrainian translation (an expression of the work)
     Manifestation:  Mobi Dik : abo, Bilyĭ kyt : roman / Herman Melvill;
                     z anhliĭsʹkoï pereklav I︠U︡riĭ Lisni︠a︡k.
                     trans. I︠U︡riĭ Lisni︠a︡k
                     Kyïv : Vsesvit, 2004. ISBN: 966959684X
                     (a specific manifestation or publication of a
                     specific expression of the work)
     Item:           British Library, System number: 013411859
                     General Reference Collection YF.2008.a.25077
                     (a specific item, "a book sitting on a shelf")

These four (Work -> Expression _> Manifestation -> Item) are just
the "Group 1" entities; FRBR is considerably more complicated (and
interesting from an ontological standpoint).

Within the library world FRBR is still relatively new, and library
software systems are still in the process of being replaced or
retrofitted to handle FRBR-ized records. Intelligently-written
digital library software stores bibliographic information in a
rather "object-oriented" way, where a data record about a specific
Item only includes Item-specific data, inheriting fields up the
tree as appropriate. E.g.,

There's unfortunately been no way of canonically identifying for
purposes of interchange within a computer system the variety of
Group 1, 2 and 3 entities and their relationships, i.e., there's
been no URLs for that purpose. Below are appended two messages
originating on the Ontopia mailing list, indicating some existing
and some future work in this area. These have been edited slightly
but are relatively intact.

In summary, we are soon [4] likely to see publication of a set of
URLs suitable as identifiers for FRBR entities and their inter-
relationships. This is exciting news for those doing Topic Map
and/or digital library-related work within the library community.

Murray

[1] http://www.cidoc-crm.org/who_we_are.html
[2] http://en.wikipedia.org/wiki/FRBR
[3] Functional Requirements for Bibliographic Records,
     Final Report, 1998.
     See:  http://www.ifla.org/VII/s13/frbr/frbr.htm
[4] IFLA 2010, 11-15 August. See:  http://www.ifla.org/

cc: Martin Doerr

---------------------- message #1 follows -----------------------

On 04/08/10 09:38, Kevin Trainor wrote:
 > Hi All,
 >
 > Can anybody explain the history behind the Ontopoly identifier type
 > called item-identifier
 > (http://psi.ontopia.net/ontology/item-identifier)? I am considering
 > using it in the sense that “item” is used in FRBR. That is to say there
 > can be many items associated with a particular “manifestation”. Yet, I
 > don’t want to hijack this type if it has a different meaning. In that
 > case, I would do better to create an occurrence type.

Kevin,

I agree -- it would be better to create four URLs in a new namespace
than hijack one from the TMDM namespace, since this would be a use for
which it was never intended. As a rule of thumb, it's always preferable
to create a new Topic than to abuse an existing one. As for creating an
occurrence type, I'm not sure how this alters the basic issue.

You could of course just make up your own as well, but for interoper-
ability within the library (FRBR) community a common set of URLs would
obviously be preferable. If the project requirements don't require
such interoperability there's not much holding you back from doing what
you like and creating your own URLs/Topics.

Back in 2005 I had a FRBR Topic Map (authored in LTM) with the following
Topics:

     /* concept types .................................................. */

     [Work : FRBRConcept = "Work" / FRBR
       @"http://purl.org/ceryle/psi/frbr/#Work" ]
     {Work, Description, [[A work by one or more authors.]] }

     [Expression : FRBRConcept = "Expression" / FRBR
       @"http://purl.org/ceryle/psi/frbr/#Expression" ]
     {Expression, Description, [[An expression of an FRBR work.]] }

     [Manifestation : FRBRConcept = "Manifestation" / FRBR
       @"http://purl.org/ceryle/psi/frbr/#Manifestation" ]
     {Manifestation, Description, [[A manifestation of an FRBR expression.]] }

     [Item : FRBRConcept = "Item" / FRBR
       @"http://purl.org/ceryle/psi/frbr/#Item" ]
     {Item, Description, [[An individual item or instance of an FRBR manifestation.]] }

     /* relation types (association templates) ......................... */

     [ExpressionOf : FRBRRelation = "is an expression of"
       @"http://purl.org/ceryle/psi/frbr/#ExpressionOf" ]
     ExpressionOf( [Expression] : From, [Work] : To) / Template

     [ManifestationOf : FRBRRelation = "is a manifestation of"
       @"http://purl.org/ceryle/psi/frbr/#ManifestationOf" ]
     ManifestationOf( [Manifestation] : From, [Expression] : To ) / Template

     [ItemOf : FRBRRelation = "is an item of"
       @"http://purl.org/ceryle/psi/frbr/#ItemOf" ]
     ItemOf( [Item] : From, [Manifestation] : To ) / Template

Not that you'd necessarily want to use the Ceryle namespace, but I'm
happy to share the frbr.ltm file if anyone wants a copy. But in the
last five years there has been plenty of newer work.

There is a set of FRBR URLs developed by the RDF community, available
at:

    Expression of Core FRBR Concepts in RDF
    http://vocab.org/frbr/core.html

e.g.,

     http://purl.org/vocab/frbr/core#Work
     http://purl.org/vocab/frbr/core#Expression
     http://purl.org/vocab/frbr/core#Manifestation
     http://purl.org/vocab/frbr/core#Item

Though this is, again, not the FRBR community.

You may also be aware of the publication of the FRBRoo Model, a formal
expression of FRBR concepts, intended by CIDOC and IFLA as a harmonisation
of their respective bibliographic models:

   http://www.cidoc-crm.org/frbr_drafts.html

Version 1.0 was published in 2009, version 1.01 in January 2010. I think
it's relatively stable. A sad but altogether-too-typical lost opportunity
is the fact that they've gotten so far as 1.01 and have no URL identifiers
for any of the defined concepts.

I have contacted the editor, Martin Doerr, the Research Director of the
Center for Cultural Informatics in Heraklion, Crete, and he has confirmed
that they are in the process of including a set of URLs for this purpose
and expect to publish an update "within the next months" that would
include URLs suitable for use as public identifiers. I will post any
information I hear to the list. This would be an appropriate set of URLs
for interchange of FRBR information and I look forward to their
availability. Then someone could publish a Topic Map merging all the
known FRBR Topics that have been used up until that time.

---------------------- message #2 follows -----------------------

On 05/08/10 09:51, Murray Altheim wrote:
[...]

Just an update: the International Federation of Library Associations and
Institutions (IFLA) is holding their annual conference on 11-15 August
and I have just been informed by Martin that they plan to "create a
respective namespace for FRBRoo, and that the topic will be discussed
and decided on the IFLA conference the next days."

This is good news for FRBR users in the RDF and Topic Maps community,
and for the library community in general, as this will promote the
development and use of interoperable data, documents, stylesheets and
tools. I will report anything I hear back to the list, but that will be
the last on this subject.

As this is not directly related to Ontopia I'll move this discussion
over to the TopicMapMail list unless anyone objects, given it's an
open list and shares many participants.

Murray

...........................................................................
Murray Altheim <murray10 at altheim dot com>                       ===  = =
http://www.altheim.com/murray/                                     = =  ===
SGML Grease Monkey, Banjo Player, Wantanabe Zen Monk               = =  = =

       Boundless wind and moon - the eye within eyes,
       Inexhaustible heaven and earth - the light beyond light,
       The willow dark, the flower bright - ten thousand houses,
       Knock at any door - there's one who will respond.
                                       -- The Blue Cliff Record
_______________________________________________
topicmapmail mailing list
[email protected]
http://www.infoloom.com/mailman/listinfo/topicmapmail