DESIRE: Development of a European Service for Information on Research and Education

Desire is a large, pan- European co-operative development project to promote the use of the World Wide Web (WWW) in the European research community. The project has attracted substantial funding from the European Union's Fourth Framework Programme.

The project is co-ordinated by SURFnet bv, the academic network provider in the Netherlands and has a total of 23 partners in 8 countries. Other major contractors are the University of Bristol, Lund University Library, Netlab (Sweden), UNINETT/AS (Norway), Origin bv (the Netherlands), the University of Newcastle upon Tyne, the Queen's University of Belfast and the Joint Research Centre of the European Union (Italy).

The aim of Desire is to solve the problems which currently hamper the use of the World Wide Web as a means of giving researchers access to research data. This is done by improving the technical infrastructure, by enhancing the users knowledge through training and by creating well-maintained collections of relevant research data.

In establishing this service, Desire builds on existing projects like SOSIG, ROADS and Netskills in the UK, The Nordic Web Index project in Sweden and SURF-ACE in the Netherlands. The reason for co-operation on a European level is to make the developed services more general and to gain knowledge and raise mutual awareness.

History

In the ISUS work group of TERENA, devoted to Information Systems and User Support, there were a lot of thoughts on the future of the World Wide Web as a means for researchers to disseminate research results. On a conference in December '94, people involved in ISUS decided on writing project proposals for the Telematics Applications Programme. Erik Huizer of SURFnet, now managing director of the SURFnet ExpertiseCentrum, wrote project proposals for an e-mail, a conferencing and an Information Services project.

The Information Services project plan later resulted in the Desire project proposal, nominated by the EC as the best proposal in the Research for Telematics area. Negotiations on the contract with the European Commission proved to be extremely difficult. The start of the project was because of these problems delayed by half a year, thus serious threatening the state- of-the-art properties of some initial anticipated deliverables. The official start of the project was January 1, 1996

The Issues

The World Wide Web has grown dramatically over the last few years. However, the amount of data available is enormous. The problems have grown equally.

It has become increasingly difficult to locate relevant resources. It is not uncommon to get over 10,000 hits when searching for a single word in search-engines like AltaVista, Lycos and Webcrawler. Even general subject based catalogues like Yahoo are becoming less useful because of the enormous amount of data.

Another problem with these tools is the fact that they are US-oriented thus inducing a lot of expensive transatlantic traffic whereas the bandwidth of transatlantic connections is scarce.

As opposed to the time before the 'Web revolution' it has become extremely easy to put information on the Internet. The ease of doing so has meant that everybody, indeed, does put information on the web. However, the quality of the information in terms of availability, relevance, uniformity and timeliness is on average very low, and in fact unacceptable for serious research activities.

A last problem is the lack of tools to provide access to information on a restricted basis, whether because of commercial interests or because of matters of confidentiality.

The Works

The goals of Desire are stated as to deliver the following:

• A sustainable approach to making searching, finding, browsing, accessing and retrieving multimedia resources as easy as possible.

  To make it as easy as possible to find resources on the Web, Desire takes a dual approach. To cater for users trying to locate arbitrary information in the European information space, Desire proposes a European Web Index (EWI). The EWI builds on the Nordic Web Index, a project carried out by Lund University Library. The EWI will enable researchers and others in Europe to discover and locate information resources from all over Europe in a coherent way.

  Other users need to locate information in a narrower set of identified categories. For this purpose Desire will provide the foundation for this kinds of services by developing subject-based information gateways (SBIGs) through which mediated, quality controlled and well-catalogued data sources may be accessed. An example of an existing service like this is the SOcial Sciences Information Gateway (SOSIG) in Bristol.

• Tools for the creation, maintenance, quality, cost-recovery and description of information resources.

  For the creation and maintenance of information resources Groningen University in the Netherlands extends a modular set of tools called Webmanager developed at the Royal Dutch Library of Sciences. This integrated set of tools allows groups of users to collaborate. The tools specifically focuses on link checking and the integrity of the information base.

  Description of the information resources plays an important role in the indexing and cataloguing which takes place in the EWI and SBIGs. On one hand, mechanisms to incorporate standard ways of describing information resources will be developed, on the other hand attention will be paid to the feasibility of automatically classifying resources.

• Security mechanisms to control access to information resources for reasons of confidentiality or charging.

  The increasing adoption of networked information services raises many issues of access control and confidentiality, for example:

  • An information system that provides feedback (such as answering a questionnaire or providing a registration service for a conference or course) may need to validate the identity of the user to prevent false or duplicate entries.
  • An information system may contain a mixture of public information and private information with access permitted only to appropriate groups of users.
  • An information provider may provide a public service, but only wishes to permit access to subscribers who pay part of the cost of the service.
  • An information provider may wish to provide access to information via both public and private networks whilst wishing to keep those networks separated for security reasons.

  The security part of Desire is devoted to the investigation of the usability of firewalls to separate public and private networks and to the usability of charging and authentication mechanisms on an individual basis. Authentication for access to information resources will be based on tables and identity controlling mechanisms like chipcards.

• Quality checking mechanisms to permit the quality of information and information services to be monitored and reported.

  To make sure that information sources are always available, a group of partners (with as main contractor the Joint Research Centre of the European Union) will develop enhancements to existing Management Information Bases (MIBs). Goal is to make it possible to use generic network management protocols like SNMP and SNMPv2 to monitor and manage information servers.

• Performance features to reduce the international bandwidth requirements of access to multimedia information across the European continent.

  The impact on the use of network bandwidth of the simplicity with which networked information can be explored and retrieved through the use of the World Wide Web has been enormous over the past few years. One way of reducing the use of bandwidth is to keep copies of recently or frequently requested information close to the requester of this information. This process is called webcaching. Special attention will be paid to find an efficient and effective way to implement hierarchical caches and co-operating caches.

  The creation of the European Web Index is anticipated to further reduce the use of, mainly intercontinental, bandwidth.

The Results

In general, taking on a project the size of Desire means taking risks. It is anticipated that some parts of the project will fail to deliver useful results, while others will not deliver the desired results. Large parts of the projects will hopefully end in useful contributions to the European research community.

Several of the pilot services which will be established, are planned to continue as production services. Component technologies (caching, indexing methods) have clear exploitation possibilities by providers of information services and network operators, although the legal problems of providing secure information services, particularly across national boundaries, will remain.

The participating service providers in Desire (SURFnet, UNINETT, Origin) will exploit the technical infrastructure to deliver an information service infrastructure to their customers.

The ongoing monitoring and review and the provision of full documentation and training packages means that the benefits of Desire will be clearly and careful recorded and that the project is ready for transition to the exploitation phase.

Tangible results so far are the deliverables presented in three of the workpackages:

• The caching workpackage delivered 'Caching Requirements Survey/ Specification for a Prototype'.
• The information tools workpackage delivered 'Overview of HTML Authoring Tools' and 'Specification for Information Provider Tools'.
• The indexing and cataloguing workpackage delivered 'Specification for Indexing Approach'.

Further deliverables will be presented on the Desire deliverables page, http://www.nic.surfnet.nl/pr ojects/desire/deliver/deliver.html.

Conclusion

The co-operation in a project of the scale of Desire remains a hot issue. It is hard to keep coherence between the workpackages. The evaluation and training workpackages and the project management will need to keep a close eye on the evolvement of the project. When this can be achieved the total of the project can have a great impact on the use of the World Wide Web as a means of communication between researchers and others in Europe.

More information on the Desire project can be found on the Desire webpages, http://www.nic.surfnet.nl/surfnet/projects/desire , or by e-mailing the author, Klaas.Wierenga@sec.nl.