CIDOC Documentation Standards Working Group
Reference Model interim meeting report

An interim meeting of the DSWG took place in London, hosted by the V&A February 10-14, 1997. The meeting was open to all members of the working group, but in the event those attending represented a core group with a specific interest in Object Oriented data modelling. The meeting was co-chaired by Nick Crofts (DSI, Geneva) and Pat Reed (Smithsonian, Washington). Participants were as follows:

Anne Serio (Smithsonian, Washington)
Martin Doerr (ICS Forth, Crete)
Ifigenia Dionissiadou (Benaki museum, Athens)
Lene Rold (National museum, Copenhagen)
Louise Smith (MDA, Cambridge UK)
Martin Stiff (MDA, Cambridge UK)
Siegfried Krause (National Museum, Nuremberg)
Jennifer Trant (Archives & Museum Informatics, Pittsburgh)
David Bearman (Archives & Museum Informatics, Pittsburgh)

Costis Dallas, Per Vestbostad and Alice Grant were unable to attend. Frances-Lloyd Baynes (V&A) was unfortunately not able to participate at all the sessions.

The purpose of the meeting was to continue work on the Object Oriented version of the CIDOC data model. (For those of you who haven't been following this, the group decided in March 1996 to develop an OO data model, the decision was formally ratified in Nairobi). 

The London meeting proved important in three ways. 

Firstly, the model was improved and a number of outstanding issues were resolved. The group agreed upon the appropriate scope for the model. Our immediate objective is to finish modelling the information categories contained in the CIDOC guidelines. However, it was recognised that the reference model should ultimately support all aspects of museum documentation: collections management, research and analysis, communication and presentation, even though many existing standards concentrate on just one or two aspects of museum activity.

The group sees the Reference model as providing a reference data structure which has the potential to support all the information categories needed for documenting cultural objects. Documents such as the current CIDOC Guidelines (formerly MICMO), the Africom, Archaeological sites and Ethnography standards define sets of information categories required for adequate documentation. It is not the role of such documents to define data structures.

Secondly, the level of understanding of OO modelling techniques and issues was greatly increased within the group. The general level of competence is now such that discussing and working on the OO model by email is a real possibility - which we intend to exploit. An initial task involves creating and circulating scope notes for all classes which have been defined.

Finally, and perhaps most importantly, some fairly intense discussion revealed an underlying conflict inherent in some of the objectives which the group had established. This divergence turned on the question of whether the model was intended essentially as a purely conceptual reference - in which case the semantic content should be as explicit and detailed as possible - or whether it should be a more practical implementation reference - which would entail taking system constraints into account. The group realised that, treated separately, the two approaches are complementary. Problems arise from trying to do both in one model. Consequently, in line with design methods such as Merise, two types of model may in fact be needed, one conceptual the other more practical.

Although well advanced, it is unlikely that the OO model will be complete in time for the Nuremberg conference; however, a full presentation of work in progress will be made.

Tentative arrangements were made for a follow-up meeting in Washington sometime in November. Anyone wishing to participate should indicate their interest during or before the Nuremberg conference.

Finally, we would like to thank all those who took part for their commitment and enthusiasm and the V&A for providing facilities for the meeting and for dealing so patiently with our incessant demands for photocopies, hard copy, coffee and biscuits.

Nick Crofts
Pat Reed

Co chairs