RE: RE: Notes from Banff f2f

Matthias,
in previous work trying to go beyond Dublin Core's information resource centric metadata I have found the so called ABC model interesting, see "The ABC Ontology and Model" http://metadata.net/harmony/JODI_Final.pdf

It was developed within the Harmony international digital library project to provide a common conceptual model to facilitate interoperability between metadata ontologies from different domains. ABC builds on the core entities in FRBR but also criticise the FRBR manifestation notion, see "Towards a Core Ontology for Information Integration" http://www.cs.cornell.edu/lagoze/papers/Core_Ontology.pdf

One of the authors, Jane Hunter, was one of the participants in the HCLS workshop in Banff.

Regards
Kerstin Forsberg
AstraZeneca



-----Original Message-----
From: public-semweb-lifesci-request@w3.org
[mailto:public-semweb-lifesci-request@w3.org]On Behalf Of samwald@gmx.at
Sent: 23 juni 2007 18:56
To: Eric Neumann; danny.ayers@gmail.com
Cc: public-semweb-lifesci@w3.org
Subject: Re: RE: Notes from Banff f2f




> I guess what I'd still like need to see is howthe concepts from our HCLS
> use-cases would explicitly map to these OCLC concepts...
> 
> - the work, a distinct intellectual or artistic creation
> - the expression, the intellectual or artistic realization of a work
> - the manifestation, the physical embodiment of an expression of a work
> - the item, a single exemplar of a manifestation.


First, let us collect the exact definition of the relevant FRBR classes as they are described on the web [1]:

----
WORK
An abstract notion of an artistic or intellectual creation.
A frbr:Work  is something that, amongst other things, is a frbr:Endeavour and is a wordnet:Work~2 but is not a frbr:Expression and is not a frbr:Manifestation and is not a frbr:Item

EXPRESSION
A realization of a single work usually in a physical form. 
A frbr:Expression  is something that, amongst other things, is a frbr:Endeavour but is not a frbr:Work and is not a frbr:Item and is not a frbr:Manifestation

MANIFESTATION
The physical embodiment of one or more expressions.
A frbr:Manifestation  is something that, amongst other things, is a frbr:Endeavour but is not a frbr:Work and is not a frbr:Expression and is not a frbr:Item

ITEM
An exemplar of a single manifestation.
A frbr:Item  is something that, amongst other things, is a frbr:Endeavour but is not a frbr:Work and is not a frbr:Expression and is not a frbr:Manifestation
----

Examples for the relations between entities belonging to these classes:

---
an_item exemplar_or a_manifestation .
a_manifestation embodiment_of an_expression.
an_expression realisation_of a_work .
---

My general thoughts on FRBR:
* The distinction between an_item and a_manifestation is actually a relation between an individual and the class it belongs to. frbr:manifestation is a meta-class in disguise and should be removed. Instead, we should just be talking about classes of frbr:items.
* With the current definition, the distinction between a_manifestation (or rather, a class of items) and an_expression is not that clear. It seems like a mixture of work and expression. Again, I think this could be simplified by defining classes of items, and not much of the expressivity would be lost.

Applying FRBR to our use case:
This is a bit complicated because with the current Demo, as (besides our description of biological reality) we are mainly making statements about *digital* entities like Uniprot and PubMed recor, and not so much on physical objects like books or print-outs. The documentation of FRBR I saw so far does not mention how the fleeting nature of these digital entities should be treated -- are the Uniprot HTML pages I see on my computer screen 'items'?

Anyway, it seems that in the current Banff demo scenario, we are distinguishing between 'works' and 'expressions', and maybe also 'manifestations'. At least that is the best fit I see between FRBR and our demo.

At the moment, the Science Commons PURL system is distinguishing three things (Pubmed example):

1)
http://purl.org/science/article/pmid/11160518
"The article itself"

2)
http://purl.org/commons/record/pmid/11160518
"The PubMed record 11160518, without commitment as to representation"

3)
http://purl.org/commons/html/pmid/11160518
"HTML representation of the record"
or
http://purl.org/commons/xml/pmid/11160518
"XML representation of the record"


(1) certainly is a frbr:work.
(3) would probably be a frbr:manifestation
(2) ...I'm not sure where the database records fit into FRBR.

Any ideas?

[1] http://vocab.org/frbr/core#sec-terms

cheers,
Matthias Samwald

----------

Yale Center for Medical Informatics, New Haven /
Section on Medical Expert and Knowledge-Based Systems, Vienna /
http://neuroscientific.net





.
-- 
Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen! 
Ideal für Modem und ISDN: http://www.gmx.net/de/go/smartsurfer

Received on Monday, 25 June 2007 09:10:21 UTC