Hi all,

Here is an email summarising some work I did on the RDA cataloguer
scenarios about a month ago, I'm forwarding here in case it's of
interest.

Cheers,

Alistair


----- Forwarded message from Alistair Miles <[log in to unmask]> -----

Date: Sat, 11 Oct 2008 16:37:18 +0100
From: Alistair Miles <[log in to unmask]>
To: Gordon Dunsire <[log in to unmask]>, Tom Baker <[log in to unmask]>,
	Karen Coyle <[log in to unmask]>, Diane Hillmann <[log in to unmask]>,
	Mikael Nilsson <[log in to unmask]>
Subject: New analysis of RDA cataloguer scenario 1

Hi all,

I've spent a day working on the RDA cataloguer scenarios, here's a
summary of what I did and some of the issues raised.

I rewrote the RDF expression of cataloguer scenario 1, and updated the
page at:

http://dublincore.org/dcmirdataskgroup/Scenarios/1

My goal was to use the RDA elements schema at [1], the RDA roles
schema at [2], a FRBR schema based on Gordon's work at [3] [4] and
[5], and the various RDA value vocabularies also in the metadata
registry.

In preparation for this, I reviewed [3], [4] and [5] and compared them
with Ian Davis' work on a FRBR ontology [6]. I understand Gordon's
work was done before there was support for schemas in the metadata
registry, so I made some assumptions about the intention.

A comparison of Gordon's and Ian's FRBR schemas is published at:

http://spreadsheets.google.com/pub?key=pWXJnkpaKdGk4kKVbATNgGg

Ian's schema looks to have complete coverage of FRBR entities and
relationships, uses a systematic and concise naming convention, and
has domains and ranges expressed, so I decided to use this vocabulary
in constructing the scenario. 

If IFLA is at the stage of minting URIs for FRBR entities and
relationships I strongly suggest they adopt the same class and
property names that Ian has used. Ian's naming convention makes
working with the vocabulary very convenient, straightforward and
predictable.

Coming back to the cataloguer scenario, I found the available RDA
schemas were sufficient to express all of the information present in
the scenario.

The most important issues emerging at the level of the whole schema
are inconsistent naming conventions for constructing property names
(e.g. workTitle vs. formOfWork), and a number of properties which may
turn out to be redundant (e.g. identifierForTheWork doesn't tell you
anything more than simply identifier if you already know the resource
is a work). I'll have more to say about these I think after working on
the other scenarios.

Here are some property-specific issues & questions the scenario
raised. I'm not sure it's worth rushing in to address these just yet,
before further analysis is available, but here they are anyway for
your interest:

 * Is rdarole:author a subproperty of frbr:creator?

 * What controlled vocabulary should be used to express the form of a
   work (e.g. novel)? Or should it be an uncontrolled literal value
   (e.g. "novel")?

 * What is best practice for representing a language as a controlled
   value in RDF?

 * What schema should be used for expressing properties of a
   frbr:Person?

 * What's the difference between rda:languageOfExpression and
   rda:contentLanguage?

 * What schema should be used to describe properties of a frbr:Place?

 * Found rda:Publisher as a property name, expected rda:publisher.

 * What schema should be used to describe properties of a
   frbr:CorporateBody?

 * For extent of a novel, is it correct to use rda:extentOfText? With a
   literal value?

 * Found both rda:identifierOfWork and rda:identifierForTheWork.

 * When should you use rda:title? What property is right for
   expressing title of a frbr:Manifestation?

A complete record of today's work is available at:

http://dublincore.org/dcmirdataskgroup/AlistairMiles/AnalysisNotes20081011

Cheers,

Alistair

[1] http://metadataregistry.org/schema/show/id/1.html
[2] http://metadataregistry.org/schema/show/id/4.html
[3] http://sandbox.metadataregistry.org/vocabulary/show/id/49.html 
[4] http://sandbox.metadataregistry.org/vocabulary/show/id/90.html
[5] http://sandbox.metadataregistry.org/vocabulary/show/id/64.html
[6] http://vocab.org/frbr/core

-- 
Alistair Miles
Senior Computing Officer
Image Bioinformatics Research Group
Department of Zoology
The Tinbergen Building
University of Oxford
South Parks Road
Oxford
OX1 3PS
United Kingdom
Web: http://purl.org/net/aliman
Email: [log in to unmask]
Tel: +44 (0)1865 281993

----- End forwarded message -----

-- 
Alistair Miles
Senior Computing Officer
Image Bioinformatics Research Group
Department of Zoology
The Tinbergen Building
University of Oxford
South Parks Road
Oxford
OX1 3PS
United Kingdom
Web: http://purl.org/net/aliman
Email: [log in to unmask]
Tel: +44 (0)1865 281993