DCMI Report

Working Group: DCMI Registry WG

Period: October 2000-September 2001

Name of chair: Rachel Heery and Harry Wagner

Date: 29 September 2001

Number of mailing list subscriptions: 70

Short description of activities in reporting period

Registry WG activity over the last year, in time honored style, saw bursts of activity followed by prolonged lulls. At the beginning of the year, inspired by feedback from the Ottawa meeting we moved on with the prototype that had been demonstrated by Eric Miller at DCMI-8. There was liaison with other groups regarding purpose of the registry, and discussion of RDF schemas. However our reliance on the EOR software meant that Eric's move from OCLC, shortly followed by that of Tod Matola, brought a halt to further progress on the software side. However we have been fortunate in recent months to have Harry Wagner at OCLC devote time to EOR and the Registry, and we now seem to be back on track. The Registry prototype is now available at http://wip.dublincore.org:8080/registry/Registry and will be demonstrated at the DCMI-2001 Tokyo workshop.

Further work has taken place recently on functional requirements, this has been assisted by prototyping done by Harry. The complexity of constructing an effective user interface to an RDF database should not be underestimated. We are seeking to provide a human readable interface to DCMI terms for an 'ordinary metadata aware' person in which we hide the RDF constructs, as well as an 'RDF-style' human readable interface, and lastly a machine readable interface. This means we require a fairly detailed functional specification which needs to take account of the interaction between search options and the structure of the Dublin Core RDF schemas. The latest version of the Functional requirements are available at http://www.ukoln.ac.uk/~lisrmh/DCMI-registry/funreq.html and will be discussed at the Tokyo workshop.

Traffic on the mailing list has included detailed discussion of aspects of RDF schemas and best ways to approach multilingual interface. We acknowledge ongoing contributions and continued interest from Eric Miller and Tod Matola.

Outstanding Issues

The following issues need to be resolved and the WG will put them on the agenda for a WG session for discussion at Tokyo

1. Multilinguality

- provision of translations for definitions and UI

- priority of multilingual functionality

2. Dependency on canonical expression of DC terms in RDF schemas

- who is responsible for RDF schemas

- what is most effective expression of structure of DC terms

3. Place of Registry in DCMI workflow

- relation of registry to Usage Board,

- role of registry in provision of documentation

4. Agreement on functionality

- finalise functional requirements

Record of deliverables

Name: Purpose and Scope of DCMI Registry (assigned Rachel Heery)

URL: http://www.dublincore.org/groups/registry/purpose-20010511.shtmlScheduled delivery: delivered 2001-05-11

Name: Policy and process: to recommend qualifier approval process and criteria for approval (assigned to Tom Baker)

Scheduled: November 2000

Comments: This task was moved to the Usage Board and now comes under their remit

Name: Prototype DC Registry: (assigned Harry Wagner)

http://wip.dublincore.org:8080/registry/Registry

Scheduled: in process Comments: expected completion by DC2001 Tokyo workshop

Name: Functional requirements

http://www.ukoln.ac.uk/~lisrmh/DCMI-registry/funreq.html

Scheduled: ongoingComments: This is a more detailed exploration of scope and purpose of registry, outlining detailed functional requirements and relating these to the DCMI RDF schemas and DCMI data model.

Short description of future activities

Future Charter

The WG will seek approval of the following charter:

Providing authoritative definitions of DCMI terms is a priority task in order to manage and promote the Dublin Core vocabulary. The DCMI Registry Working Group aims to establish a metadata registry in order to provide effective machine and human readable access to the DCMI vocabulary.

The Registry will fulfill several purpose:

- To assist DCMI to manage the evolution of DC vocabularies

- To provide authoritative definitions of recommended DC elements, qualifiers and controlled vocabularies

- To identify DCMI recommended names for schemes

- To express these 'controlled metadata sets' and the relationships between them in machine readable schema language and in human readable mode.

- To provide a user friendly interface to the registered metadata ( e.g. search and browse facility, browseable element set lists, links to annotations and guidance on use of DC elements and qualifiers)

- To manage multilingual aspects of DC.

The WG will identify requirements for a DCMI registry, recommend policy and process and implement a solution. The intention is to have a phased delivery, starting with basic functionality and content.

The Registry WG will produce the following deliverables:

1. Finish DCMI registry phase 1

1.1 Revised functional specification

1.2 Working multilingual registry

2. Canonical RDF schema(s) for expressing DCMI terms

Review current RDF schemas to ensure they effectively express the structure of the DCMI data model. This is required as base line for Registry. Liaise with Architecture WG?

3. Agree process for updating RDF schema(s)

This is required to take account of errors, new RDF terms etc

Liaise with Usage Board

4. Agree process for updating registry

Propose workflow to ensure Registry updated following amendment or addition to DCMI terms

Liaise with Usage Board

5. Registry Phase 2

Update functional specification as required

Produce documentation and web pages using RDF schema (and registry?)

Include domain specific application profiles ?

Apply revisions as required to working Registry