2nd European library seminar/3rd Gabriel workshop 4-6 June 2003 at the Biblioteca Nacional in Lisbon

Citation

Woldering, B. (2003), "2nd European library seminar/3rd Gabriel workshop 4-6 June 2003 at the Biblioteca Nacional in Lisbon", Library Hi Tech News, Vol. 20 No. 8. https://doi.org/10.1108/lhtn.2003.23920hac.003

2nd European library seminar/3rd Gabriel workshop 4-6 June 2003 at the Biblioteca Nacional in Lisbon

Britta Woldering

From 4-6 June 2003 the 2nd European Library Seminar/3rd Gabriel Workshop[1] was held at the Biblioteca Nacional in Lisbon, Portugal[2]. A total of 62 librarians and Webmasters from 20 countries participated in the workshop, mainly Gabriel contacts but also quite a number of librarians from Portugal.

Since its foundation in 1995 Gabriel, the World Wide Web service of the European national libraries, has established itself as the official network service of the Conference of European National Librarians (CENL) and has undergone a continuous development. The Gabriel Web site has been redesigned and restructured, and it was relaunched in summer 2002. At present, 41 libraries from 39 European countries are represented in Gabriel. Gabriel offers access to varied information sources and numerous online-catalogues and services of the European national libraries. Gabriel is the only trans-European library service.

The Gabriel service is used more heavily every year: last year it got about 2.5 million hits. Gabriel will also play an important role in the development of "The European Library" (TEL)[3], an integrated search and service facility to be developed after the end of the project phase. The project now enters its final year and this seemed to be a good occasion to bring together all those who are involved in Gabriel and TEL.

The aims of this seminar were: to stimulate international co-operation, to share ideas and experience regarding the World Wide Web in general, and Gabriel in particular, and to give further information about the project "The European Library". At the same time the seminar was one event in a series of "The European Library" workshops. It consisted of one and a half days with short presentations by team members, Gabriel contacts and TEL partners.

The European Library (TEL)

The seminar began with a morning session about "The European Library" (TEL) project. It was chaired by Fernanda Maria Campos, vice-director of the Biblioteca Nacional of Portugal. She gave an introduction to the TEL project and its vision. TEL is a pioneering collaboration between a number of European national libraries, created under the auspices of the Conference of European National Librarians (CENL). The objective of TEL is to set-up a co-operative framework which will lead to a system for access to the major national and deposit collections (mainly digital, but not precluding paper) in Europe's national libraries. TEL's principal aim is to establish a professionally designed and maintained single access point to selected parts of the libraries' holdings, so that the informed citizen in any country can utilise not only the resources of his or her own national library, but also, during the same session, those of any other partner's national library which may hold material relevant to his or her interest. Initially, the feasibility of this venture is being tested, with part-funding from the European Commission as an "accompanying measure" under the cultural heritage applications area of Key Action 3 of the Information Society Technologies (IST) research programme. The project started in February 2001 and will run until January 2004. The TEL partners are the national libraries of Finland, Germany, Italy, The Netherlands, Portugal, Slovenia, Switzerland and the UK, along with ICCU, the national central cataloguing institute from Italy, and CENL. In time it is anticipated that more and more of the CENL member libraries will join the enterprise as full partners.

At the end of the project phase there will not be an operational service. However, the TEL partners have committed themselves to establish a TELService, which will be launched at the Annual CENL Meeting in September 2004. The nature of the TELService will be a portal giving access to the combined resources of national libraries in Europe and allowing free resource discovery for digital and non-digital materials. For this purpose all title level metadata will be provided freely to TEL by the TEL partners and will be searchable freely by TEL users. It will offer integrated cross-collection searching and the delivery of both free and priced digital objects. The availability of legal deposit material via TEL is likely to be very limited, particularly in the short term. Therefore, the real potential will be in (digitised) heritage material, licensed material as well as traditional material. To achieve this service the evaluation of different technical options is now under way. The solution to be found must be very robust, highly scalable and widely accessible. Authentication and multilingual, searching are in the early stages of development and will not be available within the TELService at launch, while the Web site of the TELService will be multilingual, providing all languages of the TEL partners. Regarding metadata a TEL application profile was developed and the concept of a TEL metadata registry was introduced. The TELService will be funded by subscriptions of the TEL partner libraries and the costs comprise the hardware and software needed for the portal, the building and the maintenance of the database as well as personnel costs for a small team of support staff. TEL should be set up under the aegis of CENL and should be integrated with Gabriel at the service level.

In the TEL project work has been carried out in areas like content, systems, metadata, and business planning. The following presentations reflect the workpackages of the project.

First, Jani Stenvall, systems librarian of the National Library of Finland, reported on Finnish activities on legal deposit and experiences with e-publishers in the TEL context. The current Finnish legal deposit law dates back to 1980. As with almost all European legal deposit legislation, it does not comprise networked electronic publications. Since 1997 the Finnish National Library has been preparing for the new legal deposit legislation, which is likely to include off-line publications and Web publications. The recent development is that TV and radio programs and films will be included in the same legal deposit law. In parallel the Finnish copyright law is reformed and will also state the access rights to digital publications. The Finnish National Library distinguishes two kinds of Web publications: one is the Finnish Web, which will be harvested periodically without any special selection criteria. Only some special selected cultural sites with open access material will be harvested more frequently. The other kind of Web publications are restricted or licensed material, which will enter the Library as deposit and needs close contacts and negotiations with the publishers. The Finnish National Library expects to have the new legal deposit law and the technical environment in place in 2005.

Regarding the e-publishing situation in Finland it is important to state that there are no really big e-publishers. The e-publishing group in Finland consists of commercial and traditional book publishers, public sector publishers and learned societies that publish on the Internet. In 1997 the Finnish National Library started a voluntary deposit of electronic publications and since then has kept close contact with the e-publishing group.

In 2002 TEL workpackage 1 "Publisher Relations" carried out a survey among publishers in every partner country, but in Finland no publisher responded. For this reason the Finnish National Library decided to try again in early 2003 with a tailor-made questionnaire for Finland and managed to receive 42 answers from 90 questionnaires sent out. The answers reflected the whole range of challenges concerning e-publications. The public and the commercial sector of e-publishers have very different perspectives of digital publishing and especially the estimated costs of depositing e-publications show a large variety. This reflects the amount and type of e-publications the publishers are offering: it ranges from a few static documents a year to very complex objects like dynamic or database-driven publications. These are a challenge for the deposit libraries. The situation in Finland is similar to the ones in most TEL partner countries: The deposit law – if existing – is being revised, preparations are being made for deposit systems and terms and conditions for the use of the deposited materials are being negotiated with the publishers. One of the aims of TEL is to provide access to all digital material of the partner libraries, but to achieve this goal the situation of handling e-publications has to stabilise in each country and much more promotion of TEL to the publishers is necessary.

Next, Robert Smith, head of bibliographic services of The British Library, introduced workpackage 2 of TEL, "Business Plans and Models". The business planning process in TEL seeks to answer some fundamental questions:

• •

  What is the nature of the service? The TELService will be a portal giving access to the combined resources of national libraries in Europe. It will provide free resource discovery for digital and non-digital materials. The TELService will deliver digital objects both free and priced, depending on copyright issues. Nevertheless, the legal deposit material available via TEL at launch will be limited, on the one hand, because digital deposit has hardly started and, on the other hand because the regulations for access are very restrictive for the time being.

• •

  What is unique about the TELService? The TELService will bring together disparate collections of European national libraries by providing cross-collection searching and presenting integrated results. In addition it will form a launch-pad for new research.

• •

  Who is it for? The TELService has a high public mission and is mainly aimed at the "informed citizen", but also at the main user group of national libraries, the researchers.

• •

  How will the service be achieved? There are options for technical solutions whose evaluation and assessment are under way. As we expect the TELService to be accessed world-wide with many hits per day, the robustness of the solution, the scalability and wide accessibility of the system (i.e. not only with the newest versions of software) are crucial. Another important factor is cost-effectiveness:

• •

  How much will it cost? The costs of the TELService will depend on the technical solution and the funding available. The key elements of the costs are the hardware/software for the TELService, the building of the databases and their maintenance, and of course the personnel costs for the support staff. It is expected to take on a small team of two to three persons for technical, organisational, and promotional maintenance of the service.

• •

  How will it be funded? The TELService will be funded by subscription not of the end-users, but of the partner libraries. The initial funding is likely to be 250,000 Euro per year in total. So far this subscription model has received committment from the national libraries of Finland, Germany, the Netherlands, Slovenia, Switzerland and the UK.

• •

  What benefits will it bring to its investors/stakeholders and users? The benefits of the TELService for the TEL partners will be the increased use of their collections, the improving and broadening of access to their collections and it will serve as a platform for further collaborative programmes. For the end-users the TELService will be a new and dynamic research tool which provides easy integrated access to digital and traditional resources. This opens new possibilities for innovative research.

At the last TEL Steering Committee meeting in April 2003 in The Hague it was agreed that the TELService will be set up following the completion of this accompanying measure. The major impetus for TEL is not based on market considerations, but is political: fulfilling a cultural role in a united Europe. It is planned to set up the TELService under the aegis of CENL and to integrate it with Gabriel at the service level. Some specifics of the Hague Agreement are that authentication and multilingual searching will not be available at launch, while a multilingual user interface in every TEL partner language will be provided. Further, as mentioned above, the availability of legal deposit material will be very limited, while real potential is in digitised heritage material, licensed and conventional material. Another important agreement is that all title-level metadata will be provided free to TEL by the contributing partners. It was also agreed to integrate Gabriel and TEL at the service level, because it would be difficult to maintain a distinction between the two services which are complementary to each other, when looking at the mission statements[4]. How this integration can be achieved will be worked out in the near future.

After this outline of the business plan which takes more and more shape, the TEL metadata application profile and technical aspects of TEL were introduced. Theo van Veen from the R&D department of the Koninklijke Bibliotheek (The Netherlands) and leader of the TEL metadata workpackage, presented the metadata development in The European Library and the metadata model, which was defined to support shared access to digitised and printed material provided by the TEL partners or by publishers. The main objective of the metadata development, to be carried out in the metadata workpackage of the European Library project, is to define a datamodel to support shared access to digitised and printed material provided by the TEL partners or by publishers.

The first stage in this workpackage consisted of a review of the state-of-the-art based on the survey of current practices among the partners and on desk research. This survey of the current practices was the result of a questionnaire that was sent to the project partners. An outcome of the questionnaire was that there were no common data model, no common bibliographic format and no common terminology. There is also some dissimilarity in the digital content which is offered by the different partners. The partners have agreed on the next steps to solve these issues, the first step being the installation of a TEL metadata working group.

It was agreed to use XML as record syntax with Dublin Core as a starting record schema for the initial testbeds. However, the TEL working group also concluded that the TEL operational service should eventually make use of the Library Application profile, which is a combination of Qualified Dublin Core and terms originating from other application areas. Additionally it was decided to analyse the functionality of TEL in order to check whether the Library Application profile was sufficient to meet the TEL functional requirements. By mapping functions against metadata needed for those functions the working group reached the conclusion that the Library Application profile was not sufficient to meet the requirements of TEL and that a TEL Application profile is needed.

During the course of the project it was acknowledged that TEL will further evolve after the end of the project and that it should take into account future functional requirements. In order to meet – at this moment unknown – future functional requirements TEL should allow for a controlled evolution of the TEL Application profile. For this purpose the concept of a TEL metadata registry has been introduced. TEL participants may introduce new metadata terms and define a number of characteristics for those terms, which will be put in the metadata registry before entering a formal acceptance procedure. The application profile will be derived from this registry.

Special attention has been paid to digital collections and collection level descriptions. TEL is a collection of collections and it was concluded that TEL would become a much more powerful search tool when collection level descriptions were offered, integrated with metadata describing other objects. This should result in a functionality that allows a presented collection description record to be transformed into a searchable target for the portal.

A test portal, developed to test the access of XML resources via http, has been successfully used to test and demonstrate the relationship between metadata and functionality.

Next, Annette Siegenthaler from Die Deutsche Bibliothek (Germany) and leader of the TEL workpackage 4 "Interoperability Testbeds" presented the analysis of a "Call for information", which was carried out to support the decision for a portal software for the TELService. The TEL project had been initiated to conceptualise an infrastructure for a European Virtual Library. National libraries in Europe have already powerful information systems and OPACs. Workpackage 4 of the project has taken on the task to explore how the existing services could be integrated into one European service. A dual approach was chosen: Libraries already running a Z39.50 target should be incorporated by using the protocol Z39.50. Others should provide the metadata of their collections as data providers via the OAI protocol. A test bed was implemented that gathers the metadata via OAI, builds a central index, and offers those data via an SRU[5] service.

The idea was pursued to integrate both approaches in one central portal. Several portal software products were spotted that would suit this purpose and an elaborated list of requirements had been developed. These requirements were used in carrying out a so-called "Call for information": nine vendors of appropriate portal software products were asked to check the list of requirements and answer the questions: to what extent their product would cover the required features, what initial costs would be incurred and which maintenance costs would have to be calculated.

Eight vendors sent answers to this query and provided us with detailed information about their product. The work-package members chose the following features as being the most important ones and analysed and assessed the very comprehensive material the vendors sent according to these features:

• •

  SRU support: search and retrieve via http, a protocol using http/XML still under development and being used for the TEL http/XML testbed.

• •

  Dublin Core support and extensibility.

• •

  Collections are considered to be a major feature in TEL. The user interface should offer collection level descriptions of those collections available through the portal.

• •

  Authentication by the portal, so that the user has to register only once for the use of restricted services in TELService.

• •

  https: allows transmission of encrypted data via the Web.

• •

  Different versions: multilingual interfaces of the TEL portal.

• •

  De-duplication: nice to have, but low-ranking feature.

• •

  Control of functions: the TEL administrator can change things, not only the portal vendor.

• •

  Support of OpenURL: to provide a standard mechanism to link to resources.

• •

  Support of Unicode was not demanded in the requirements. It was required that the supported character set should be indicated as an option to the user. Unicode will not be used in the search, but should be a possible character set for the presentation.

• •

  ZTHes profile describes an abstract model for representing and searching thesauri (semantic hierarchies of terms) and specifies how this model may be implemented using the Z39.50 protocol or other protocols and formats. This profile is considered to be a worthwhile approach for implementing translation services. The support of the ZTHes profile was not demanded in the requirements but, because many vendors indicated that their products support it, it was included in the analysis.

• •

  Security is a very complex issue relating to the security of the server as well as the client. A comparison between the different portal software systems regarding security was left out because of this complexity.

• •

  Scalability.

In the meantime the TEL Steering Committee decided to refrain from buying portal software for the time being, but to enhance the SRU portal used in the testbeds to get the service started.

The Final Conference of the TEL project was to take place on September 24, 2003, at the National Library of Lithuania in Vilnius as a pre-conference to the annual meeting of the directors of the CENL libraries. The project phase ends on January 31, 2004. The launch of the TELService consisting of the actual TEL partners is scheduled for September 2004, on the occasion of the 2004 meeting of the directors of the CENL libraries.

The digital library

The afternoon session of the first day of the workshop dealt with various aspects of digital libraries: the national digital library, digital deposit and digitisation.

The first presentation was given by José Luís Borbinha, director of the Services for Innovation and Development of the National Library of Portugal (BN). He presented some of the projects in which the BN is involved, for example, Linking and Exploring Authority Files (LEAF)[6], funded by the European Commission under the 5th framework of the Information Society Technologies (IST) programme. It started in March 2001 and will run for three years. The scope of LEAF can be summed up under the question: How can millions of existing authority data originating from very different sources be used together by everybody (librarian, archivist, museologist, public user) in such a way that there will be no loss of information, no data accumulation without quality check but an automated linking between high quality information driven by actual user needs? LEAF will try to enhance search and retrieval facilities by providing high quality access to authority information for everybody. For this purpose LEAF is developing a model architecture for collecting, harvesting, linking and providing access to existing name authority information, independent of their creation in libraries, archives or museums and independent of national differences. The scenario will be to establish a Central European Name Authority File using authority files about personal names. A LEAF demonstrator will be built, and the results of LEAF will be integrated into the MALVINE search engine and thus extend MALVINE into a global multilingual and multimedia information service and maintenance agency about personal names.

Another very interesting but regional project is the connection of name authority files of the BN to the Portuguese Internet search engine "tumba!" If a query consisting of an author's name is submitted to the search engine, a link to the OPAC of the BN is provided at an exposed position of the result list. If the user clicks on that link, he is connected to the OPAC and a search query with this author's name is carried out automatically. The user gets a results list from the OPAC and can use all services offered, for example, in the case of a free accessible digital resource he may view the material. Further, the BN is developing a digital deposit system for the National Digital Library for which the prototype is expected to be ready at the end of 2003.

The Koninklijke Bibliotheek (KB) of The Netherlands has no prototype but was the first national library in the world to have an operational system for long-term preservation of digital publications[7]. Marco de Niet, head of the Innovative Projects department of the KB and Gabriel Team member, presented this system, which is called Digital Information Archiving System (DIAS) and gave an overview of current research activities on digital preservation of and permanent access to digital objects. The challenge of digital publications consists of finding a solution to create, maintain, manage and exploit an archive for digital objects, which can guarantee that the contents will remain stored securely and that they can be accessed and retrieved from any place, at any time, now and in the future. Since 1995 the KB has carried out various activities to create such a secure place for digital objects and to be able to offer permanent access. Finally, the KB developed a secure place together with IBM, which became operational in December 2002. This system follows the NEDLIB main recommendations:

• •

  to follow a common approach regarding infrastructure, workflow, terminology etc.;

• •

  to create the e-deposit as a self-contained system for archiving and preservation;

• •

  to make a distinction between the bitstream (the data object) and the representation based on the decoding of the bitstream; and

• •

  to model the e-deposit system after the reference model for an Open Archival Information System (OAIS, ISO 14721:2002).

The next challenge after having created an operational secure place is to develop an operational module to guarantee permanent access to the contents in those systems. There are two main threats for permanent access to deposited digital objects:

1. 1.

   evolution of application formats and technology standards (e.g. Wordstar, Wordperfect, Lotus, dBase etc.); and

2. 2.

   loss of the logical/intellectual information and authenticity (for example, through current conversion tools in MS Word etc.).

The KB and IBM work together in various activities to tackle these threats: on the one hand, they are designing and building the preservation subsystem of the DIAS and, on the other hand, they are experimenting with the Universal Virtual Computer (UVC). The preservation subsystem of the DIAS can be summarised by the following three objectives:

1. 1.

   to identify the digital objects in danger of becoming inaccessible due to technology changes;

2. 2.

   to implement the activities associated with technical preservation, i.e. migration and emulation strategies; and

3. 3.

   to register the technical metadata needed to generate and validate the software and hardware environments required to render the digital object.

The UVC is a method for preserving digital documents using emulation, being developed by IBM. A distinction is made between preserving the original data and preserving the behaviour of a computer program to render the data. The UVC enables the creation of programmes that are written to decode specific digital object types for the UVC. If it is possible to emulate the UVC in the future, all documents which can be viewed using the UVC can be emulated.

Pilot systems of the DIAS preservation subsystem and the UVC for coloured JPEGs should be finished in December 2003. The installation of the pilots in the DIAS should take place in the first half of 2004.

The last presentation of the first day of the workshop was given by Olga Barysheva, head of the division of digital resources of the National Library of Russia (NLR)[8]. She gave an overview of the organisation and the development of the digital collections in the NLR. In 2000 the NLR decided to create a digital collection consisting of digitised objects of important treasures held by the NLR. Digitising objects has many advantages: the objects can be made accessible for many users at the same time, independent of place and time. In addition, the original can be preserved by reducing its physical usage dramatically. Furthermore, digitised objects and digital collections contribute to the global cultural heritage being made available to everybody world-wide.

After having decided to build a digital collection, the NLR started with equipment procurement and installed a digital resources division in 2001. After having decided on the data model, the development of the program for the digital collection management was carried out. When the technical environment was established, the digitisation could begin.

The first project was the creation of an electronic collection of heterogeneous objects consecrated to the St Petersburg anniversary. Support for multi-user access via a common interface, effective search facilities, and the possibility for the creation of a CD-ROM special series could be provided. The collection consists of different resource types: graphic and art materials, maps and atlases, manuscripts and rare books. The project was concluded successfully.

The first project after the pilot project was the digitisation of 90 illustrations from the Great French Chronicles, an edition from the fifteenth century. Another project was the digitisation of manuscripts from the archive of the famous Russian poet Anna Akhmatova. It consists of 4,500 leaflets and was a great challenge to digitise because the texts had to be made more readable than the original because the ink had faded.

The NLR digital collection system now comprises five collections (St Petersburg, art graphics, maps and atlases, newspapers, manuscripts and typescripts) with more than 5,000 digital copies. A total of 485 CD-ROMs with 308 gigabytes are held as depository copies. Further digitisation and expansion of the digital collection are planned.

Gabriel and digital libraries

The second day of the workshop was opened by a presentation about Gabriel by the Gabriel editor, Olaf Janssen from the KB of The Netherlands. He gave an overview of the development of Gabriel from its start in 1995 up to now. Gabriel is the official www-service of the CENL libraries and the only genuine pan-European library service. A total of 41 libraries from the 39 countries which are members of the Council of Europe participate in the service. It is hosted and maintained by the KB in The Netherlands and is mirrored in Finland, Germany, Slovenia and in the UK[9]. It offers standardised information about the 41 national libraries and their services. Furthermore, the Gabriel Web site offers information about Gabriel and CENL, about European co-operation in projects and partnerships as well as news and updates. In addition it presents the online exhibition "Treasures from Europe's National Libraries", a guestbook and a search engine which allows the user to search all Web sites of the European national libraries participating in Gabriel.

In 2000 the Gabriel Board decided to modernise the Gabriel Web site (www.bl.uk/gabriel/). Because of the growth of the Web site in the three years since its launch, it became too large to handle and updates were too time-consuming and left room for errors. The back-end file structure was poorly organised. Above all there was a lack of navigation after the homepage and the look-and-feel, being from 1995, had become old-fashioned. From 2001 to June 2002 Gabriel was restructured and redesigned with the goal to build a clearer file structure (Gabriel must be easily transferable to other hosts), to offer a better navigation, to upgrade the graphical lay out, and to implement a database for data that are edited frequently. This database contains the practical library information, e.g. collections, opening hours, address etc. and the online services information (more than 150 services, e.g. OPACs, national bibliographies, Web sites etc.). The rest of the Gabriel Web site remains static, because fewer changes occur there. The advantage of the database is that it makes sure that all libraries provide the same information in the same format and it offers a Web-based front-end with Webforms, the Database Input Tool (DIT). Via the DIT the libraries can view, modify and add information about them and their services easily and have more control over their own data. The changes in the database will not be reflected immediately in changes on the Gabriel Web site (non-dynamic front-end), but will be visible after the next update, which is done once a month. After each update the Web site is mirrored.

The plans for Gabriel for 2003 besides the continued work on and promotion of the Web site are the analysis of the joint Gabriel/TEL user survey, which was online in May and June 2003, the publication of the results and the establishment of an online photo gallery of national library buildings on the Web site.

In the second part of his presentation about Gabriel the Gabriel editor presented the intermediate results of an ongoing joint Gabriel/TEL action: an online user survey. This survey was set up because the Gabriel Team and Board recognised that they knew very little about the users and the usage of Gabriel. Furthermore, the users of Gabriel are considered to be one of the main user groups for the eventual TELService and the TEL project group considered a data analysis of the Gabriel usage or a user survey as being extremely valuable. For the survey a questionnaire was put together, which was online on the Gabriel Web site in May and June 2003. The invitation to fill in the questionnaire was disseminated through the Gabriel contacts, posted to European and world-wide Internet discussion lists, Internet periodicals and via a TEL newsletter. The questionnaire consists of five sections with questions about:

1. 1.

   the respondent;

2. 2.

   the use of the Internet (frequency, duration, interests);

3. 3.

   the use of libraries (frequency of visits, usage of library services, type of library user);

4. 4.

   the use of the current Gabriel Web site; and

5. 5.

   requirements from the future Gabriel Web site.

Olaf Janssen presented intermediate results of the first 330 responses he received in the first three weeks of the questionnaire being online:

• •

  most respondents are between 36 and 50 years old and two-thirds are female;

• •

  most respondents hold a Master's degree;

• •

  the main areas of interest on the Internet are bibliographic search, e-mail, messaging etc. and news services;

• •

  most respondents would put themselves in the categories of higher education, academic or personal researcher, when using library services;

• •

  when visiting the current Gabriel Web site the pages with information about online services, about library information and the Gabriel search engine are used most and the users often find what they are seeking;

• •

  the preferred language on Gabriel is English for about 70 percent of the respondents;

• •

  the average rating of some aspects of the Gabriel Web site, like information quality, usefulness, comprehensiveness, ease of navigation etc. was about 3.5 on a scale from one to five, with information quality being rated best and style and design being rated worst;

• •

  if Gabriel offered the facility to search all catalogues of all European national libraries at the same time, almost 90 percent of the respondents would use it very often, and they would use it for locating where publications are as well as finding and downloading publications or checking book descriptions;

• •

  almost every respondent would be mainly interested in digital text documents, followed by digital images (one-third of the respondents); and

• •

  if the respondents had to pay for downloading a publication of 20 pages, the majority would consider 1-5 Euro a fair price, but about 30 percent would not pay.

The final analysis of the responses was to be carried out in summer and the publication of the results was expected by September/October 2003.

The third part of Olaf Janssen's presentation was a proposal of the Gabriel Team and two workpackage leaders of TEL for the integration of Gabriel and TEL at the service level. It is anticipated that after the end of the project phase of TEL the Gabriel and TEL project Web sites will be maintained separately until the launch of the TELService. CENL member libraries will get access to the interoperability testbeds developed in TEL for testing purposes. After the expected launch of the TELService in September 2004 a common homepage for both services at the URLs of the Gabriel mirror sites will be created and the integrational work will be started: one entrance for all services and collections of all CENL libraries, a uniform look-and-feel, one solution for searching collections and Web sites, one name for the whole service, probably a completely new one. The agreements about financial, organisational and technical arrangements will be reached between the CENL libraries and the TEL partners.

The next presentation was given by Elisabeth Freyre, the Gabriel Team member of the Bibliothèque Nationale de France. She presented another survey, commissioned by CENL and carried out in 2001 among the 41 CENL libraries, of which 33 replied. The aim of the survey was to highlight European and international projects in which the CENL members are involved jointly with institutions from their own countries and/or foreign countries, as well as to achieve a better overview of this cooperation. The projects concerned could be within or outside the framework of the European Commission. In the replies, 66 projects were mentioned, of which 19 were outside the scope of the survey. The remaining 47 projects were examined. What emerged from this survey is the extent to which European national libraries are involved in joint projects. Most of these joint projects are European-wide, followed by regional ones within a certain geographic area such as the Nordic/Baltic countries or German-speaking countries. The project themes and results are spread among eight categories in the following order: union catalogues, access to catalogues and library services, digital library, preservation, cataloguing and retroconversion, management, music and others. Most of the projects deal with several of these themes. The partners in these projects comprise, apart from libraries (mainly university or special libraries, very rarely public libraries), associations of librarians and professional networks, universities and technical centres of research or applications, national government ministries and administrations, museums and archives, publishers and private consulting or technical firms. The majority of the projects have a one-to-four-year timescale, most of them covering a period of three years. The funding sources are varied: 20 of the 47 projects are funded by the European Commission, 15 projects are funded by national ministries and administrations and the funding of the rest is split among public research institutions, private foundations, ad hoc professional consortia, professional associations and universities.

As the findings of this survey were so fruitful, CENL decided to put them online on the Gabriel Web site. In addition, the significance of such a survey depends on it being updated on a regular basis. An online form will therefore be developed to allow each CENL member to report and/or update an ongoing project.

The third presentation on the second day of the workshop was given by Luciano Ammenti, CED director of the Vatican Library, about the new local library system, the new WebOPAC[10] and the use of radio frequency identification (RFID) technology for reading rooms.

Since 2001 the Vatican Library has carried out various projects of digitising and cataloguing non-book material. The catalogue comprises, besides monographs and journals, descriptions of engravings, prints, drawings, photographs, medals and coins, musical scores, soundtracks and CD-ROMs. On the WebOPAC of the Vatican Library the digital images of prints and engravings are connected to their descriptions in the catalogue: the user can view the objects, enlarge and rotate the images. The images are protected by watermarks. It is planned to offer digital images with links from the catalogue for the medals and coins collection and also for the manuscripts.

Every librarian knows the problem of misplaced books in reading rooms and the trouble of finding and relocating them. The Vatican Library put a lot of time and personnel resources into this task every year, and therefore decided to introduce a system that allows the regular monitoring of the location of the books, so that they can be found at any time. Another advantage of the monitoring is to get a statistic of the usage of each book, which allows a better planning of space in the reading rooms. After two years of testing and assessing the feasibility of various methods, the Vatican Library decided to assign a chip in RFID technology to every inventory item. This technology uses radio frequency as a vehicle to allow an electronic identification from a distance. It was chosen because the RFID system is not structurally invasive for either the surroundings or the people, it is permanent, it does not entail a lot of maintenance, and the data conserved in the chip are protected by a 128-bit code and by an unequivocal 16-digit numeric address created by the producers during the production of the chip. With this technology it is possible to survey the chip and conduct a count directly from the shelves. "On the spot" identification with the possibility of checking for missing and mislocated volumes and periodic revisions – all these functions can be carried out with the RFID chip's palm survey monitor and the wireless network card, with online database inquiry possible, so that the actual systems data are always available. Every time a chip is read the relevant bibliographic data appear on the display. Regarding the glue for the chip labels to be put in the books, the nature of the different documents had been taken into consideration: different glues for old and modern material were used. Thanks to the use of the RFID technology the Vatican Library is able to have perfect control of the material offered and used in the reading rooms.

Next, Robina Clayphan, Metadata Development Analyst from The British Library, introduced the InterParty project[11]. The project ran from April 2002 until June 2003 and was funded under the European Commission's Information Society Technologies (IST) programme. Its aim was to design and specify a network to support interoperability of party identification (for both natural and corporate names) across different domains. "Parties" are either individuals or organisations involved in the creation and dissemination of intellectual property (or "content"). InterParty built on the work of the <indecs> project, one of whose recommendations was the specification for a Directory of Parties. InterParty was not proposing the development of an identification scheme to replace existing schemes for the identification of participants in the intellectual property domain (for example, national library name authority files or systems oriented towards the needs of rights licensing) but a means of effecting their interoperation. InterParty aimed to deliver the following:

• •

  Analysis of existing data models.

• •

  Identifiers and party metadata model.

• •

  Report on privacy and security mechanisms.

• •

  Specification for working demonstrator.

• •

  Demonstrator (alpha system).

• •

  Business model, exploitation plan and governance proposal.

The InterParty value proposition is to offer access to common metadata held in partner databases and the creation of a body of links between identities in those databases to make explicit the relationship between identities that may have different attributes. These functions could help reduce the costs of data creation and maintenance, enable the improvement of data quality and provide a basis for other services. There is an underlying assumption that all parties whose data appear in the database will have an active interest in disambiguation and in a greater certainty of identification in circumstances where this is advantageous to them. In the context of InterParty, disambiguation is the most significant aspect of the metadata. Therefore sufficient metadata are needed to allow disambiguation between parties with shared or similar attributes or to allow collocation of the same party with different attributes. InterParty will create value to the extent that it enables new relationships to be expressed and recorded (InterParty links), for example, "Person X in namespace A is person Y in namespace B". But "is" is only one example of the relationship that might be established; "might be" and "is not" could be equally valid and valuable. Such claims of relationship within InterParty will be made by the members of the InterParty network. A centralised editorial service is an unrealistic option, because the establishment costs are too high. For personal metadata InterParty distinguishes between "publicly known", "not publicly known" and "sensitive" data. It is important to recognise that InterParty is concerned with "public identities" and not with real "persons", and that a public identity is not the same as a name (more than one name may be associated with the same public identity). Within the InterParty network, each public identity will require a Public ID Identifier or 'PIDI' (a combination of a namespace and a unique identifier within that namespace). InterParty links will be expressed as relationships between PIDIs. Regarding the holding of the common metadata, InterParty is likely to require some centralised indexing. For the demonstrator the InterParty links will be the only data held centrally to support automated processing.

To summarize, InterParty was a project designed to demonstrate that interoperability of party identifiers is a real possibility. Interoperability is essential to allow meaningful, unambiguous communication between people and machines, and machines and machines, to fulfil the promise of the network. The problem with party identification is that of ambiguity because simple names are not the same as identities. The basis of InterParty functionality is access to party metadata and a growing body of linked identities. These offer the possibility of disambiguation of public identities and the opportunity for participants to reduce the costs and to increase the quality of their own metadata.

The workshop concluded with a spontaneous presentation of the Web site of the Bibliotheca Alexandrina[12], given by Carlos Bernardo, a representative of the Casa de Portugal Bibliotheca Alexandrina.

Conclusion

The participants judged the workshop to be very fruitful and informative and agreed on the continuation of the European Library Seminars/Gabriel Workshops every two years. The next one is expected to be in 2005 after the launch of the TELService and at the beginning of the integration of TEL and Gabriel.

Notes

1. 1.

   Web site for the TEL/Gabriel workshop in Lisbon: www.kb.nl/gabriel/2003workshop/index.html

2. 2.

   Homepage of the National Library of Portugal: www.bn.pt

3. 3.

   Project Web site of "The European Library" (TEL): www.europeanlibrary. org

4. 4.

   For the Gabriel Mission Statement and the TEL Vision see: www.kb.nl/gabriel/; www.europeanlibrary.org

5. 5.

   SRU: Search and retrieval encoded within the URL

6. 6.

   Project Web site of LEAF: www.leaf-eu.org

7. 7.

   For further information about the Dutch activities in digital archiving and preservation see: www.kb.nl, for Expert centre, Digital archiving and preservation.

8. 8.

   Homepage of the National Library of Russia: www.nlr.ru

9. 9.

   Gabriel Web site and mirrorsites: www.kb.nl/gabriel/; www.ddb.de/gabriel/; www.lib.helsinki.fi/gabriel/; http://nuk.uni-lj.si/gabriel/; www.bl.uk/gabriel/

10. 10.

    Homepage of the Vatican Library: www.vaticanlibrary.vatlib.it

11. 11.

    Project Web site of InterParty: www.interparty.org

12. 12.

    Homepage of the Bibliotheca Alexandrina: http://www.bibalex.org

Dr Britta Woldering(woldering@dbf.ddb.de) is the Gabriel Board Secretary and is based at Die Deutsche Bibliothek, Frankfurt, Germany.