Another Piece of Cake?

Actually, that was really the way it was: a cake for the consumption of a group of people interested in the development of digital libraries. It was in 1991, long before this concept had been implanted into the heart of the library community. Indeed long before most people knew about the Internet.

At this meeting the first seed for what was going to be NetLab [1] was sown - the multidisciplinary research- and developmental group aimed at digital libraries and global network based information systems. The NetLab seed grew in the earth that the explosive development of communications systems (international computer networks) and computer graphics was made up of, together with access to programs for global network-based information systems.

The collaboration between the librarian Traugott Koch and the programmer/computer technician Anders Ardö was water for the tender plant. Enthusiasm and multidisciplinary competence gave the nourishment that made the organisation grow faster and faster. NetLab has always consisted of a distinctive blend, a staff mixture of equal parts librarians and computer programmers for the optimal nourishment. Neither part has dominated, but both have contributed with their special knowledge in an unusually successful cooperation where the whole is greater than the sum of the parts. The moral and practical support from the library management also did contribute to the life-giving light for the growing NetLab plant. The purpose of NetLab was to test and develop a common solution for searching for electronic information using a common user interface based on systems for distribution and searching via computer networks, but also to spread knowledge of the possibilities for electronic information searching that was being made accessible in the early 1990´s. In the beginning a substantial part of the work focused on "missionary work", that is to spread information about the digital library, its advantages and its possibilities. The "missionary workshops" had a very practical content and also a frankdialogue between Traugott the librarian and Anders the computer programmer. At the same time NetLab built a digital library at a local server as a demonstrator of the new technologies. In 1992, a work station for the use of Gopher, WAIS and WWW was placed at one of the libraries in Lund, UB2, as oneof the world´s first Internet terminals freely accessible at a library.

The first experiments with the creation of a digital library was presented in 1992 at the 15th International Essen Symposium. The first NetLab version of a digital library was dominated by local information from Lund university and the campus libraries. Of course, this approach was widened and the digital library extended its linking to the rest of the Internet world. During this first period of time several strategically important projects were initiated. The main focus was the development of tools and methodologies for building digital libraries, e.g. knowledge organisation, metadata and WWW-indexing. A major part of the work was aimed at external projects, both national and international.

These first steps lead to contacts internationally and with like-minded people in the Nordic countries.

Many of NetLab's projects stem from the Nordic WAIS/World Wide Web Project (W4 project for short) [2]. Most of the themes we have worked with later can be seen in embryonic form there. For example the W4 included cross searching between a library OPAC and a web index, automatic classification in UDC and then obviously search and navigation with respect to subject metadata.

The search and navigation themes were further developed in the Engineering Electronic Library, Sweden (EELS) [3], which started in 1994. It was NetLab's first service with stringent quality selection criteria metadata definitions. As such it is a very early implementation of a Subject Based Information Gateway, (SBIG).

The W4 project ended in 1995 and was followed up in 1996 by two new Nordic projects, first the Nordic Web Index (NWI) [4] and somewhat later the Nordic Metadata Project (NMDP) [5], both with partners from the Nordic Library communities, and in the case of NWI also with an academic Internet Supply Provider.

These two concepts, the metadata aware web index and the SBIG, have until recently been the two most important threads in NetLab's work.

Web indexing and metadata

1994 is an important year in Internet development. Apart from that NetLab released its first SBIG that year, Internet itself celebrated its 25th anniversary, and we experienced the first breakthrough of the WWW. Both the first and the second international WWW conferences were held, but perhaps more importantly it was for the first time possible to order pizza on the the web. 1992 the Gopher search tool Veronica was released, 1993 developers released the first spiders on the web and again 1994 the first full scale search engines appeared [6].

The W4 project (see above) released its search engine pilot a year later. None of the American search tools did support eight-bit characters, so the W4 pilot and the full scale NWI were in that respect early implementors of a web search engine supporting European character sets.

However, NWI had a companion project in the Nordic metadata project, NMDP. In 1996 The Nordic Council for Scientific Information, NORDINFO [7] had the vision to support the NMDP. It was the first international project, which made the choice of building Dublin Core-based tools. NetLab participated in this project that would last four years and was responsible for building the Dublin Core Metadata Creator [8] which is still in use.

Having been one of the Dublin Core pioneers has been a somewhat mixed blessing to the project. The Nordic metadata project has become well-known all over the world. On the other hand, the steady and still continuing progress of the Dublin Core has forced the project to constantly modify the tools earlier made available. This has been possible because the project partners were involved in other metadata projects such as EVA – the acquisition and archiving of electronic network publications [9] and took the Nordic metadata tools and developed them according to their own needs. Many of the resulting new services were later incorporated into the basic product.

Like the Dublin Core, our tools now have solid kernel functionality that will not be radically altered. These tools - especially the metadata template but also metadata harvesting and indexing software and URN generator - are in widespread use. It is important to support metadata creation with user guides and appropriate tools. The resulting metadata must be properly indexed to motivate the providers to do the job properly. The experience from the NMDP has shown that the applications and documentation required for Dublin Core production and use can be developed very effectively in international collaboration-operation, and then adapted according to the local needs for national and regional projects.

The intention was that NMDP, among other things, should provide guidelines and the infrastructure needed and NWI the harvesting infrastructure for building of a Nordic national search services, which we envisioned should have search facilities approaching those found in library OPACs -- enabling users to search for creator, title etc. Through the information retrieval protocol Z39.50, these services should be possible to cross search, and it should also be possible to cross search them with OPACS.

Even though the vision did never become reality, we did actually implement that metadata search engine. But the metadata needed was never produced in the quantity needed. However, within the DESIRE [10] projects we developed a new harvester, the Combine [11] which was a substantial improvement. The vision was fruitful in another way: It spawned new projects. The nationwide metadata enhanced search engine proved infeasible, federated metadata initiatives were not. Coalitions of organisations were formed within different communities. Within one of these, the providers of information about Swedish academic research commissioned such a search engine from us, SAFARI [12]. It used the NMDP's metadata creator together with the Combine harvester to provide advanced search interface to its users.

What we envisioned proved infeasible. These early projects appearing in Australia, UK and Germany as well as here in Scandinavia, are to be regarded as predecessors of initiatives as RSS and OAI. Also, only in Sweden tens of thousands of web pages has been meta-tagged during the last 5 years or so, and this cataloging effort is most likely the second largest after the libraries' cataloging of books.

Apart from the above mentioned, NetLab has been using Dublin Core Metadata in a series of projects such as EUC [13], EULER [14], EUN [15], NISBIG [16], ETB [17], Renardus [18], studera.nu [19] and Skånewebben [20].

People at NetLab have also taken part in the development of metadata by being members in boards within this issue. Traugott Koch and Sigfrid Lundberg are members of the Advisory Board of the Dublin Core Metadata Initiative and Traugott Koch is also a member of the Dublin Core Metadata Initiative Usage Board (maintenance of DCMI standards).

Subject Based Information Gateways

Subject Based Information Gateways (SBIGs) are subject entrances to quality assessed Internet resources. Most of these services have scope policies declaring what subjects they are indexing as well as defined target group, for instance academics and researchers. The resources in SBIGs are selected in accordance with an officially published list of quality criteria. The resources are then manually given rich descriptions that are fed into a database as records. This work is done by librarians or subject specialists and the records provide metadata about the resources, containing information like title, URL, keywords and abstract. The SBIGs are often limited to a specific subject, but may be universal as well. Furthermore, the resources in SBIGs are classified and given classification codes. These codes can be extracted automatically from the records and used to organise the browsing structure of the service. The use of open standards, in order to support co-operation with other services, is widely used.

NetLab has played a part in the development of SBIGs since 1994 when EELS started. Within the European Union funded projects DESIRE, Renardus and ETB, NetLab has been taking part in work and project-related research concerning quality assessment, classification and metadata but also the additional robot-based indexing. The development within these projects have been of use for SBIGs, such as EELS, Länkskafferiet [21], Skånewebben and NISBIG.

One can say that the DESIRE project focused on how to build and maintain SBIGs, using existing and developing SBIGs as inspiration and test beds. The goal was to show how to make an SBIG, no matter what subject should be covered. Since several of these existing SBIGs collaborated-operated within the DESIRE project a common wish arose about making a cross-searching functionality available to the users. The main attraction for this is the possibility to catch the different perspectives that various disciplines have on a certain subject. If different SBIGs apply the same quality criteria, utilise the same information in their records and use the same software for their database construction they can be cross-searched. As with the integrated search the result from the cross-search will distinctly show which SBIG the records are from.

The Renardus project has taken this a step further. The aim of the Renardus project is to provide users with integrated access, through a single interface, to SBIGs and other Internet-based, distributed services. The collaborative approach has implications for organisational, technical and metadata standards. This service will be available to the end users in June 2002.

Another European project which focuses on sharing and integration is ETB. The aim here is to create a dynamic metadata network through which the subject gateways can share their educational resources´ metadata. The project will enable and encourage trans-cultural and trans-national co-operation and communication and will enable individuals and workgroups to produce, handle, retrieve and communicate information in the languages of their choice, and to combine information resources from different regions and countries, and of different levels.

The development within these projects have been of use to our own SBIGs. Especially EELS has been constantly developed and improved due to the outcomes of the projects, since it has been used as a testbed both in DESIRE and Renardus. One of NetLab tasks in DESIRE was to do some testing of different automatic classification methods and to develop a pilot service using EELS integrated with the robot-generated "All Engineering" [22] database. The user can choose to search either only the SBIG, both SBIG and automated index, or the index only. The hit list from a "mixed" search will give the SBIG-hits first, since they are quality-assessed, followed by the results from the automated index.

Dissemination

Besides the SBIG and the web indexing using metadata, NetLab has also been something of an educational center for teaching internet skills to librarians. The "missionary work" by Traugott Koch and Anders Ardö in the beginning of NetLab´s history has led to a lot of things. Many NetLab members, and in particular Traugott Koch, have given speeches, lectures and workshops all over the world on many different occasions, concerning aspects of the digital library.

NetLab have had a history of cooperation with the Danish organisation DTV in the form of the Nordic Net Center (NNC) [23]. This cooperation has mostly consisted of common projects, such as NISBIG. One of the most successful projects for the NNC group was the organisation and realisation of the online course "From Internet surfer to web master" in 1996-1997. Many librarians from all over the Nordic countries signed up for the course and took part in the discussions about digital libraries that evolved.

There have also been courses aimed at librarians in foreign countries. 1997 there was a workshop series in Vilnius and Lund for Lithuanian university librarians. The workshop dealt with both sides of the Internet, that is searching as well as producing material. In 1998 NetLab held a conference, The Electronic Library – Gateway to Information [24], about the electronic library and its possibilities for librarians from ten developmental countries around the world. This has been followed up by the CELI workshops [25], who started in 2001, aimed at ten university libraries in the south of Africa.

The present and the future

NetLab´s project portfolio has contained a variety of projects durig these ten years. There have been local and national as well as international ones. They have gathered around the main focuses just mentioned: SBIGs and metadata, sprung from each other, and given ideas for new projects.

We feel that we have achieved quite a lot during this decade. Perhaps the most important thing is that NetLab has trained some 30 or 40 librarians and software engineers in digital library practices. Around a dozen of those are still active within our organisation. Many of the others have reached positions they would not have got unless they had the NetLab experience in their CVs.

NetLab started as a project and evolved into a development organisation. As such it lived its own life in a partly self-imposed isolation from its owner, Lund university Library. NetLab was more involved in international collaboration than in its owner, which has been reorganised and matured to meet the challenges of the new developments in scientific communication. The library developed from a traditional monolithic university library into a network of libraries [26], which just recently for the first time acknowledged the need for a research and development department. It realised that it owned one. With some of its original freedom traded against a much higher local funding, NetLab has itself been integrated into the Library head office, and got the project many of us desired the most: to build the infrastructure for our own digital library service.

References:

1. NetLab http://www.lub.lu.se/netlab

2. Nordic WAIS/World Wide Web Project http://www.lub.lu.se/W4

3. Engineering Electronic Library, Sweden (EELS) http://eels.lub.lu.se

4. Nordic Web Index (NWI) http://decweb.ethz.ch/WWW7/1900/com1900.htm

5. The Nordic Metadata Project (NMDP) http://www.lib.helsinki.fi/meta

6. Hobbes' Internet Timeline http://www.faqs.org/rfcs/fyi/fyi32.html

7. NORDINFO http://www.nordinfo.helsinki.fi

8. The Dublin Core Metadata Creator http://www.lub.lu.se/cgi-bin/nmdc.pl

9. EVA http://www.lib.helsinki.fi/eva/english.html

10. DESIRE http://www.desire.org

11. Combine http://www.lub.lu.se/combine

12. SAFARI http://safari.vr.se

13. EUC http://www.medianet.org/euc

14. EULER http://www.emis.de/projects/EULER

15. EUN http://www.eun.org

16. NISBIG http://www.lub.lu.se/nisbig

17. ETB http://etb.eun.org

18. Renardus http://www.renardus.org

19. studera.nu http://www.studera.nu

20. Skånewebben http://www.skanewebben.nu

21. Länkskafferiet http://lankskafferiet.skolverket.se

22. “All” Engineering http://eels.lub.lu.se/ae/index.html

23. NNC http://www.lub.lu.se/NNC/swedish.html

24. The Electronic Library – Gateway to information http://www.lub.lu.se/sida

25. CELI http://netlab.lub.lu.se/sida/celi

26. Lund University Libraries http://www.lub.lu.se

 

Author Details