This note by the COST G9 chairman summarizes main achievements of the COST action G9: Modelling Real Property Transactions, from its start in March 2001 until August 2002. The summary includes an assessment whether the achievements so far matches the research plan of the Technical Annex. The Technical Annex was the base for the COST Committee of Senior Officials for adopting our proposal as a COST action. It thus sets the criteria against which our deeds are assessed.

A version of the note was e-mailed to the deputy chairman and the 3 nominated working group leaders August 15th for critical comments and additions. With minor editorial changes, the present version was circulated August 22nd to the COST G9 Management Committee for approval at the 4th MC meeting in Delft, October 12th 2002.
 

2. Summary

The research is highly relevant, and the progress reasonable, taking the resources available into account. The structuring of work into working groups, as foreseen by the Technical Annex, was not achieved. This implies that the direction of the research is probably too dependent on only one person (the chairman). This note suggest an alternative to working groups, as a means of sharing responsibility for giving research direction. Following an account of urgent research, it proposes Management Committee action in terms of stated commitments by members of the MC to address specific parts of the G9-research.
 

3. Main achievements during first third of action activity

a. Book manuscript
A 181 page manuscript for a book: THE ONTOLOGY AND MODELLING OF REAL ESTATE TRANSACTIONS, has been prepared, based on contributions from the G9 researchers and further participants of the well organized Bremen workshop, November 2001. Each of the contributions have been reviewed by about two G9 researchers, cf the Table of contents. The manuscript comprises a 3½ page glossary that relate to the contribution of Rados Sumrada. The publishing company Ashgate is now reviewing the manuscript. Publishing is expected in 2003.

b. Expression of Interest (EOI)
An EOI on the Network of Excellence: Modelling Real Property Institutions, was prepared and submitted in response to the Call EOI. FP6. 2000. It is available at http://www.i4.auc.dk/~est/

The EOI develops from the G9-action in the following way: G9 aims at developing the concepts, the ontologies, that mirror the similarities of real property transactions in participating countries. Assuming that this is done during the next year or two, we should be able to 'reconstruct' the institution of real property rights, as the institution is practised in different countries. That is: we use our developed modelling language to express the real property institutions of diverse countries, and explain why and/or to what degree and at what costs they are working.

The EOI was structured to accommodate for indications, which were given by staff of Danish research units, regarding where our research field fit into the 'world view' of the EU's 6th research framework. This fits with presentations given at the Vienna workshop by Harry Uitermark, Netherlands, and Marjan Ceh, Slovenia, on work on the ontology of topographical databases.

While the outcome of the EOI effort is pending, the support for submission is a stated fact: Of the geodetic surveying departments, all G9 members joined, as Armands Auzins, Latvia, after deadline expressed support for participation. Furthermore the Institute for Geoinformatics, Westfaelische Wilhelms-Universitaet Muenster, Germany, agreed to join.
The participation by computer science/ knowledge engineering departments saw changes, some reflecting the move by Heiner Stuckenschmidt from Bremen, Germany, to Department of Artificial Intelligence, Vrije Universiteit Amsterdam, Netherlands. However, also here a new department agreed to join, namely the Institute of Applied Informatics and Formal Description Methods, University of Karlsruhe (TH), Germany, while Christian S Jensen, Department of Computer Science, Aalborg University, signed up after deadline.
The economic departments of G9 apparently contributed as COST midwives. Important is that Benito Arrunada, Department of Economics and Business, Universitat Pompeu Fabra, Spain, after deadline agreed to join. He has published intensively on economic issues within the field of G9.
Summarizing, the continued G9-support for the EOI and the interest of new departments to join indicate that our new research field is being recognized at the European level.

c. Workshops and MC meetings in Brussels, May 2001, Bremen, November 2001, and Vienna, May 2002

The workshops have primarily functioned as a forum for the G9 researchers. The presentations have facilitated a common understanding of the research issue, but also provided the base for outlining the research field as evidenced by the book manuscript, cf. 3a above.
The Brussels meeting agreed on the setting up of working groups, as suggested by the Technical Annex. However, the subsequent events were not structured accordingly. Workshop presentations may be classified into categories like:

• broad description of property rights within a country (Latvia)
• detailed, semi-formal description of property transactions (Denmark, Finland, Hungary, Slovenia)
• theoretical and methodological issues (property rights; conceptual languages, methods for ontology development)
• other, e.g. assessment of transaction costs (Finland), XML and graphs (Austria)

d. Presentation of initial findings

Initial findings were presented at the geodetic surveyor's FIG XXII International Congress, Washington, D.C. USA, April 19-26 2002 (Erik Stubkjær: Modelling Real Property Transactions, Joint Session 14: Cadastral Appraisal, Land Markets and  Valuation) and published accordingly (http://www.ddl.org/figtree/pub/fig_2002/Js14/JS14_stubkjaer.pdf ).
 

4. G9 research: What's the next steps?

The following does not attempt to review the G9 domain as such. Reference is made to pertinent chapters of the book manuscript. The further development of reviews and research issues count among the research tasks to be accomplished next.

The ideas presented in the following aim at identifying the research that is needed to establish an initial ontology. We may draw upon the experience made already within the action; namely that when a model has been developed, e.g. of the subdivision process in Slovenia, it is fairly easy to prepare a parallel description of subdivision processes in other countries. An initial ontology is not to be 'repeated' in the same sense as a model of the Slovenian subdivision process, but the draft ontology provides for a frame of reference for our discussions of alternative ways of expressing the abstract structures of real property rights. This frame of reference has both a technical aspect: what 'format' and tools are chosen, and a conceptual aspect: what formal language and conceptualization is applied. The following is a rather sketchy account of these issues.

As regards tools, the OntoEdit version 2.0 was tested in February 2001. One outcome is a snapshot of the user interface, after the entry of basic property right concepts in the form of concept hierarchies:

The Onto-Edit tool invites to tree structured accounts of the universe of discourse. However, it was not obvious how more complex (>2-ary) relations are to be rendered. Examples of such relations include:

• prepares (Agent /GeoSurveyor, Message/ Map), that is: the surveyor prepares a map
• models (Message/Map, GeoObject/ RealEstate), or: a map models the geographical aspect of real estate
• requests (Agent/ AssetHolder, Agent /Notary) to restructure (Asset / RealPropertyRight), or: an owner requests the service of a notary for performing a transaction
• redistribute (Activity/Transaction, Asset), or: a transaction redistributes assets, as money is traded for rights.

OntoEdit applies a XML-based format (.oxml) for storage and exchange of ontology details. Like HTML, the format is readable for humans. A part of the tree structure that is rendered on the image above (namely Agent with sub categories AssetHolder and CadastralAgency), is thus available in the following form:
 

It may be an open issue, whether a general XML editor (e.g. XMLwriter) is more appropriate, as it provides a larger flexibility in defining relations and other compound structures. A problem is that while having a larger flexibility, "[a]t the same time, however, you want your "categories" to be grounded in sound representation, say in a reference ontology, in order to perform automated reasoning, for example, to assist search, etc." (Leo Obrst <lobrst@mitre.org> Subject: Re: Ontology "views"/"perspetcives" 14. dec. 2001)

Leaving the aspect of tools and addressing the issue of conceptualization and 'reference ontology', a point of departure may be the paper presented by Degen, Helle, Herre & Smith at 2nd International Conference on Formal Ontology in Information Systems, FOIS'01: GOL: A General Ontological Language.

An account of the GOL-paper is outside the scope of this note; in stead, reference is made to an annex of this note: A tentative application of GOL for the domain of real property. An important point of the GOL paper is that knowledge is better rendered as bound to specific 'situations' than as bound to abstract 'laws' or 'norms'. The annex subscribes to this idea by structuring concepts around 'situids' and 'configurations', for example as follows:
 

A final contribution to the eliciting of core ontologies of real property rights is the detailed rendering of Slovenian transactions on real property, cf. the corresponding Glossary (Annex B).

It is crucial that an initial ontology is outlined and that appropriate tools are selected. All G9 researchers are therefor urged to indicate how they intend to contribute to the field indicated above. This field is not the only one within the scope of the G9 action, but those who can need to push forward to establish a frame of reference for further refinements.
 

5. Action development relative to the research plan (deliverables) of the Technical Annex

As requested by the Technical Annex, the Working Party on Land Administration (WPLA) has been informed of the G9 action. The WPLA organizes European managers of cadastral agencies and will be represented by the Delft workshop, where time is allocated to discuss issues of mutual interest.

As mentioned already, working groups were set up as stipulated but did not influence research. For example, terms of reference for the three working groups were not developed. Research commitments were not specified, e.g. relative to the plan of the Technical Annex, but followed informal e-mail dialogues. Similarly, a staffing of tasks was not accomplished. However, research contributions to the book manuscript does refer to neighbouring research as requested.

The Technical Annex stipulates one seminar a year and a further MC meeting a year, combined with working group meetings. So far, the praxis has been combined workshops in plenum and subsequent MC meetings. Hands-on exercises, including the drafting of alternative ontologies, have not been accomplished yet.

The Technical Annex mentions further activities for 2nd and subsequent years, including the preparation of a preliminary report around newyear 2003- 04. These activities are not covered here.

Evaluation and proposal

The lack of formality in research organization can be perfectly appropriate in new research fields as the present. However, this flexibility has to be animated and directed, and the directions shared. When unexpected events occurs (e.g. relocation of workshop, preparation of Expression of Interest), the resources left for direction of research appear to be reduced to a critically low level.

The Management Committee is urged to address this situation.

It is proposed that MC members commit themselves in writing for rather specific research tasks for the subsequent 12 months period. Such commitments may include staff from other departments; that is, another way of establishing 'working groups' and carrying responsibility for research direction. Draft commitments shall be circulated before the MC meeting, and discussed, adjusted and decided at the MC meeting.
 

6. Summary

The research of COST action G9 has so far progressed well, taking the available resources into account. Deviations from the proposed organization of research of the Technical Annex have been addressed. The presented account of research issues is meant as contribution to a more complete account, as decided by the Management Committee. MC members are invited to commitment themselves in writing to address specific parts of the thus stated research issues.

Annexes

A. A tentative application of GOL for the domain of real property. ESt, Aug18, 2002
B. Glossary  [COST G9 – Rados Sumrada – Version Aug.20, 2002]
C. CadastralActivitiesConsolidated. ESt, March 7, 2002

The annexes are available by e-mail request to est@i4.auc.dk