If someone asked us for a short and concise definition for library, we could say that a "library" is:

What are the implications of these new technologies and of these new ways of life in the library? The purpose of this paper is to raise the discussion and contribute to reach a vision and a model for the future relationship between these two entities, Internet and the library, trying to understand what can it be the "digital library".

From the definition above we are going to retain the keywords "knowledge", "community" and "memory", and from what we said after that we will add "computers" and "networks". These are the words we are going to try do not forget in this paper.

2. DEFINING A "DIGITAL LIBRARY"

Indeed, the dream of the digital library is not new, but now we are starting to see it more clearly, believing that we are at last in condition of turning it into a reality. In fact, with the power of our computers and networks (Internet, WWW, electronic mail, etc.), we have now the technology to convert our memory and knowledge in digital bits, to store it at an affordable cost and, when required, to make it available to everyone at any point in the Earth (or sky). But how should we do it?

Computers were first introduced in libraries to help in the management of the catalogues of that knowledge and memory. In the beginning, they were used almost exclusively by the librarians, but with the vulgarization of the networks and the personal computers the catalogues became online. In fact, it was even coined an acronym: OPAC - Online Public Access Catalog. With the advent of computer networks, sharing catalogues become easier to libraries, and a new kind of services were improved, the Inter Library Lending (ILL).

Meanwhile, the widespread use of personal computers and the success of open networks like the Internet brought also a new reality: the digital publication. It became easy to write a text, to store it in a server, and to have it accessible worldwide. So, instead of just a library of books, we can now also have a library with bytes.

How could we now define this new reality? Just a library with computers? In 1994, and as a result of a workshop, a group of expert took the definition a bit further:

At a first glance, the definition above seems acceptable, but what about if, in a digital world, the only way to get an edition of a novel, such as the well known Tolstoi’s classic "War and Peace", would be in a digital format?

Even if we could print it, avoiding to have to read it on the computer screen, we doubt it’d be pleasant to read (at least with the printing technology we have today at home or in the office). And what about the global value of this edition when compared, let’s say it, with a 50 years old classic edition? The book has a special place in our culture, and these books would tell us not only what Tolstoi thought about Napoleon one century ago, but also what people thought about Tolstoi a half a century ago. Will it be possible, for a huge computer file build somewhere in a digital dictatorship, to keep and transmit also this "knowledge" and "memory"? We doubt it.

Luckily, another group of experts came out, more recently, with a less radical vision:

But if we think that information and knowledge in digital format can be not only text but also sound and images, we can conclude that we are not inventing nothing too radical. Lots of libraries, all around the world, have it already. So, what is the interest of all of this, after all? The real interest comes when we recover, from the Gladney’s definition (but keeping the spirit of this last definition), two new keywords: "services" and "communications". Here we prefer "services" as a more open concept than just information systems, and communications means essentially Internet.

3. WHAT PEOPLE HAVE BEEN DOING

With the intention of creating some context to our discussion, we are now going to present a brief description of some actions claiming to be related with the present topic, the digital library.

We do not intend to present an exhaustive list of all the trials and projects under development. Our purpose is only to refer to initiatives that, by their technical or just historical reasons, reflect from our perspective (and limited by our sources) the generic approaches and areas of intervention.

3.1. From Time Gone By: The Gutenberg Project

Project Gutenberg begun in 1971!

The main mentor of this project was Michael Hart, who started it under the premise that "anything that can be entered into a computer can be reproduced indefinitely" (Hart, 1992). The purpose has been to collect as many as possible copyright free electronic texts, in the most simple possible formats (most of them are in simple ASCII) and turn then available to the public by the simplest possible way.

The project has been running in a volunteer basis, and all the material collected (about 50 GBytes) is available from Walnut Creek in CD-ROM or by ftp from ftp.cdrom.com (path "pub/gutenberg"). The material available is divided in three types: light literature (Alice in wonderland, Peter Pan, etc.); heavy literature (The Bible, Shakespeare, Moby Dick, etc.): and references (thesaurus, dictionaries, almanacs, etc.).

From our keyword list we would choose "knowledge" and "computer" to label this initiative.

3.2. Make My Day: Technical Reports Online

The CORE project started in 1991, and its purpose was to build a database of scanned journals published by the American Chemical Society (Entlich et al., 1995). In the end of 1994 they had a database of more than 400,000 pages of full text and graphics (in magnetic tapes and CD-ROM). The text was converted to ASCII and marked-up with SGML (Standard General Markup Language), being the database accessible with dedicated X-Windows interfaces. The other major contributors of this project were the Cornell University, OCLC, Bellcore and Chemical Abstract Service.

The results of the CORE project were very well accepted by the users, but another conclusion was also that "the task of building and maintaining electronic journal databases remains formidable."

A contemporary and also ambitious initiative, the TULIP project, started in March 1991 and concluded in the end of 1995 (Elsevier, 1996). It was sponsored by Elsevier Science, and involved nine universities in the USA (C.M.U., Cornell,, Georgia Institute of Technology, M.I.T., Univ. of California, Univ. of Michigan, Univ. of Tennessee, Univ. of Washington, and Virginia Polytechnic and State Univ.).

The main goal of the project was to research and test systems for networked delivery and use of scanned journals. Elsevier contributed with the scanned page images and OCR generated text, plus bibliographic data, of 43 materials science and engineering journals. The universities provided solutions to deliver these journals in electronic form to their users. The focus of the research was on technical issues, user behavior and organizational and economic problems.

When the project started, the Internet was already a reality, but the Web was still in an embryonic state. Due to that, the delivery technology was based on dedicated graphical clients for X-Windows, MS-Windows and Apple Macintosh, besides alphanumeric clients for mainframe terminals. But soon the maintenance costs were evident, and the project shifted to WWW.

In the final conclusions of the project, it was pointed out that the transition from conventional to digital libraries (defined here as libraries with full digital contents), will take much longer and cost more than commonly thought, mainly due to network bandwidth and storage limitations.

However, and as it was also pointed by the CORE project, we think that this conclusion can not be dissociated from the approach taken: to scan the original material. For example, it was estimated in TULIP that a typical journal issue, with 20 articles and 200 pages, requires approximately 17 Mbytes of storage, with 16 Mbytes for the scanned pages (in TIFF format). The ASCII information resulted from the OCR process requires only 800 Kbytes, and the indexing and bibliographic information (in SGML format) requires about 200 Kbytes.

More pragmatic approaches were taken in a series of projects in the Computer Science Reports area, where some of the most representative are UCSTRI - Unified Computer Science Technical Report Index (Van Heyningen, 1994), NTRS - NASA Technical Report Server (Nelson et al., 1995), WATERS - Wide Area Technical Report Service (Maly et al., 1994)(French et al., 1995), and CSTR - Computer Science Technical Reports(Anderson et al., 1996).

A common goal of these projects was easy installation and maintenance of the server sites and support for heterogeneous collections. The idea was not to provide only scanned and OCR material, but also to take advantage of the fact that today it is normal to produce, in the source, these materials already in digital formats (ASCII, MS-Word, PDF, HTML, GIF, etc.).

In April 1995, the teams of the WATERS and CSTR projects joined efforts, and they opened, in November, a new service: NCSTRL - Networked Computer Science Technical Reports Library (Davis, 1995). NCSTRL is a network of servers providing three kind of services: repository, indexing and user interface. At this moment NCSTRL is a worldwide service, with more than 50 universities and research centers across the world.

If we had to label this kind of initiatives, we would choose from our keywords list "knowledge", "community" and "networks".

3.3. Working Against Time: The Telematics for Libraries Program

Telematics for Libraries is a part of a vast Telematics Program launched in 1991 by the European Community.

The first phase run from 1991 to 1994, and supported about 80 actions, of which 51 are projects (one third already finished, mainly studies and pragmatic approaches, the rest being scheduled to finish until the end of 1996 or during 1997). The major part of the projects are centered in the library services and library networking (an area where Europe was relatively logging behind), but a large variety of other topics have also been addressed (such as electronic publishing, delivery of electronic publications, copyright, etc.).

The second phase started in 1995, and at this moment there were accepted 21 projects (some of them still under discussion). The last call asked for projects that encourage the European market to demand and supply information technology (IT) products and services and the entry of new players. It also challenges the libraries to "continue and accelerate the move from collection-based to access-based services, by means of resource-sharing, interconnecting and networking". On the other side, it gives priority to "projects and applications based on real needs and requirements, leading to concrete results" (Telematics, 1995).

An a first glance, the progress between the first and the second calls is not evident. Projects still concentrate in the development of traditional library services, such as catalogues, OPAC and ILL, with a strong presence of public libraries.

One reason is that, due to the fact that a great part of them are still running, the impact of the projects of the first phase is not yet evident. Another reason is the "pragmatic fever" that seems to have attacked the European Commission, which encourages only proposals with "concrete results", making it difficult to propose (and to have funded) true research projects". If we think that all the projects have to be multinational initiatives, and taking into account the broad technical and cultural European diversity, then maybe we have some explanations to the possible problem of lack of progress. As a final conclusion, we could say that it is an interesting engineering program, where everyone can show its programming skills, but a week support research initiative. But, off course, it is only an opinion...

From our keywords list, we could choose "networks" and "memory" to define this program. With some good will, maybe "services" could be a third label.

3.4. Putting It to Work: The Electronic Library Program

The Electronic Library Program (eLib) is a national UK program, launched in October 1994 with a budget of about UK £15 million over 3 years (Rusbridge, 1995). This program follows the recommendations of the Follett Report, a report of a joint commission of several higher education and library funding institutions in UK (JISC, 1993). The purpose of this commission was to investigate "how to deal with the pressures on library resources caused by the rapid expansion of student members and the world-wide explosion in academic knowledge and information."

The eLib approach is to support a large number of small and pragmatic projects, with 59 projects already approved at the end of August 1996 (eLib, 1996). This program shares some perspectives with the Telematics program (mainly in what concerns the pragmatic approach), but its UK national focus offers a real context and maybe better chances of success.

Due to their small focus, no single project explores a complete vision of the digital libraries problem. However, we could label all the eLib program with "knowledge", "community", "computers", "networks" "services" and "communications" (we should reserve "memory" for more institutional contexts, maybe with the direct involvement of national libraries).

3.5. Looking at the Future: The Digital Library Initiative

The Digital Library Initiative (DLI) is a four-year USA program, started in October 1994 and managed jointly by NSF - National Science Foundation, ARPA - Advanced Research Projects Agency, and NASA - National Aeronautics and Space Administration (Schatz & Chen, 1996).

Six research oriented projects (chosen from 73 proposals), have been funded with $24.3 million, in a total budget of roughly US $75 million. Each project, lead by one university, will develop a testbed for research and prototyping purposes. The program focus is to advance the means to collect, store and organize information in digital form, and to make it available for searching, retrieval and processing via communication networks (such as the Internet). The funded projects are:

Carnegie Mellon University: interactive on-line digital video library system to enable users to access, explore and retrieve science and mathematics materials from video archives. Research issues are integration of speech; image and natural language understanding; human-computer interaction; pricing and charging for digital video use; and privacy and security (Wactlar et al., 1996)(Cristel et al., 1995).

University of California, Berkeley: digital library with a focus on environmental information. Research areas include automated indexing, intelligent retrieval, and search processes; database technology to support digital library applications; new approaches to document analysis; and data compression and communication tools for remote browsing (Wilensky, 1996; Wilensky, 1995).

University of Michigan: large-scale multimedia digital library for earth and space sciences. The main research areas are scale, heterogeneity and software agents. The testbed will consist of a distributed environment, populated by three classes of cooperating agents: user interface agents, mediation agents, and collection interface agents (Atkins et al., 1996; Crum, 1995).

University of California, Santa Barbara: access to large and diversified collections of maps, images and pictorial materials. The research focus is in the area of spatially-indexed and graphical information (Smith, 1996; Smith & Frew, 1995).

Stanford University: shared environment for a virtual library of heterogeneous networked information sources and collections (from personal information collections to on-line collections of traditional libraries). The research issues are heterogeneous databases; information sharing and communication models; client information interfaces; and user finding services (Paepcke et al., 1996; Stanford, 1995).

University of Illinois: digital library of journals and magazines articles in the engineering and science literature. Research areas are scalable systems, semantic retrieval and sociological evaluation of the testbed (Schatz et al., 1996; Schatz, 1995).

It is easy to classify the DLI program from our modest point of view, since all of our keyword seem to be addressed.

4. BUT WHAT REALLY IS THE INTERNET?

Now, after this small tour, let’s return to our vision.

If we want to define the Internet, we need to understand what people have been doing with it. Until May 1995, the major backbone service in the Internet was assured in the USA by the NSFNET (after this date the service was decentralized in a new architecture, where traffic is exchanged at several interconnection points). As a central point, this service had the chance to collect interesting numbers about how the different services were used, as we can see in Table 2 (NIC, 1995) (Berghel, 1996).

From these numbers we conclude that the Internet is used for communication (Email and News); searching and browsing (WWW) and what we could call transfer of goods, such as software and information transferred by WWW and ftp.

To complete this picture we must point out that local and Intranet traffic, mostly Email, is not covered by these numbers, so we should consider the importance of the Internet as a communication medium as underestimated. This comment becomes even more important if we add that, for the over 4000 worldwide distributed News groups, the accounted traffic is the traffic transferred between the News servers, each one being typically accessed by several users in local networks or through POP (Point Of Presence) services.

On the other hand, it has been broadly pointed out that the information technology (IT) in general, and the Internet in particular, has been supporting the existence of virtual communities, something that we could define as communities of individuals that have strong common characteristics (cultural, professional, political, etc.), but are not geographically confined. In the words of Tim O’Rally:

This vision has already a field of concrete experiences in the scientific community. It was first identified by Steven Harnad, that coined the expressions "skywriting", "esoteric publishing" and "preprint continuum", and discussed the implications in the publishing process and relationship with esoteric communities (Harnad, 1990; Harnad, 1991; Harnad, 1995). Harnad presents an interesting perspective on the evolution of the human communication, with the phases of speech, writing, printing and now skywriting, both a new medium and a new model of communication, interactive, independent of the space and more suitable to the human cognitive process. In this context, esoteric communities can be defined as real virtual communities (an apparent contradiction?), living and working around their knowledge and memory and using Internet for their skywriting activity.

5. A VISION FOR THE DIGITAL LIBRARY

Finally, in the previous point, we discussed the Internet and we found all of these keywords again.

With this reflection we can now complete the view of the Internet as the mean (the "ether") that can bring the library, now the digital library, to the community. In fact digital libraries can be important not only for the geographically defined communities (that have already the traditional communication structures), but even more important for the non-geographic communities, as an active member of the process of development and creation of knowledge and memory.

In conclusion, we can finish with a vision and a definition for the concept of "digital library". A "digital library" is defined not only as a repository for data and information, with the traditional mission of preserving knowledge, but as a system with the mission to stimulate, support and register also the process of creation of that knowledge. This mission can be performed by a correct configuration and usage of the Internet technology (discussion lists, annotation services, etc.), providing this new library with functions that will turn it in an alive and interactive partner for the different virtual communities that are born and living in this new world.

We can point as concrete examples of this vision projects such as the "Internet Archaeology" journal (http://intarch.york.ac.uk/). It is an eLib project of an electronic journal that will incorporate multimedia and virtual reality reconstruction of sites, links to other external material (such as the excavation databases) and will give to the readers the chance to attach their own analysis and opinions to the articles. Another eLib example can be the ESPERE project (http://www.ulst.ac.uk/espere/), dedicated to electronic submission and peer review. Together, these kind of initiatives will make it possible to have a scenario where "electronic publishing unavoidably leads to a situation best described by the metaphor of the permanent seminar" (Guedon, 1994).

All these initiatives are now new, but soon they will be part of our society and knowledge, and no one doubts that it’s memory and preservation will be a mission for the libraries.
 
 

REFERENCES

Anderson, G.; Lasher, R.; and Reich, V. (1996). The Computer Science Technical Report (CS-TR) Project: A Pioneering Digital Library Project Viewed from a Library Perspective. The Public-Access Computer Systems Review, 7 (2).

(http://info.lib.uh.edu/pr/v7/n2/ande7n2.html).

Atkins, D. E.; Birmingham, W. P.; Durfee, E. H.; Glover, E. J.; Mullen, T.; Rundensteiner, E. A.; Soloway, E.; Vidal, J. M.; Wallace, R.; and Wellman, M. P. (1996, May). Toward Inquiry-Based Education Trough Interacting Software Agents. IEEE Computer, 28 (5): 69-76.

Bergel, H. (1996, January). The Client’s Side of the World Wide Web. Communications of the ACM, 39 (1).

Berners-Lee, T.; Cailliau, R.; Loutonen, A.; Nielsen, H. F.; and Secret, A. (1994, August). The World Wide Web. Communications of the ACM, 37 (8): 76-82.

Bush, V. (1945, July). As We May Think. The Atlantic Monthly. (http://www2.theatlantic.com/Atlantic/atlweb/flashbks/computer/Bushf.htm).

Christel, M.; Kanade, T.; Mauldin, M.; Reddy, R.; Sirbu, M.; Stevens, S.; and Wractlar, H. (1995, April). Informedia Digital Video Library. Communications of the ACM, 38 (4): 57-58.

Crum, L. (1995, April). University of Michigan digital Library Project. Communications of the ACM, 38 (4): 63-64.

Davis, J. R. (1995, September). Creating a Networked Computer Science Technical Report Library. D-Lib Magazine. (http://www.dlib.org/dlib/september95/09davis.html).

eLib. (1996). eLib Project Details. http://ukoln.ac.uk/elib/projects.html

Elsevier Scienc. (1996). TULIP Final Report. Elsevier Science Edition (also available at http://www.elsevier.nl/locate/tulip).

Entlich, R.; Garson, L.; Lesk, M.; Normore, L.; Olsen, J.; and Weibel, S. (1995, April). Making a Digital Library: The Chemistry Online Retrieval Experiment. Communications of the ACM, 38 (4): 54.

French, J. C.; Fox, E. A.; and Maly, K. (1995, April). Wide Area Technical Report Service: Technical Reports Online. Communications of the ACM, 38 (4): 45.

Gladney, H. M:, Fox, E. A.; Ahmed, Z.; Ashany, R.; Belkin, N. J.; and Zemankova, M. (1994). Digital Library: Gross Structure and Requirements: Report from a March 1994 Workshop. Proceedings of Digital Libraries ’94, College Station, Texas, USA.

(http://www.csdl.tamu.edu/DL94/paper/fox.html).

Guedon, J. (1994, May). Why are Electronic Publications Difficult to Classify?: The Orthogonality of Print and Digital Media. Directory of Electronic Journals, Newsletters and

Academic Discussion Lists. Association of Research Libraries. (gopher://arl.cni.org/00/scomm/edir/guedon.94).

Harnad, S. (1990). Scholarly Skywriting and the Prepublication Continuum of Scientific Inquiry. Psychological Science, 1: 342-343.

Harnad, S. (1991). Post-Gutenberg Galaxy: The Fourth Revolution in the Means of Production of Knowledge. Public-Access Computer Systems Review, 2 (1): 39 - 53.

(http://info.lib.uh.edu/pr/v2/n1/harnad.2n1).

Harnad, S. (1995). The PostGutemberg Galaxy: How to Get There form Here. The Informa-tion Society, 11 (4): 285-291.

Hart, M. (1992). History and Philosophy of Project Gutenberg.

http://www.promo.net/pg/history.html

JISC, Higher Education Funding Council for England. (1993). Joint Funding Council’s Libraries Funding Review Group: Report. (http://ukoln.ac.uk/follett/follett_report.html).

Lyman, P. (1994, June 9). Designing the Global Reference Room. The Follett Lecture Series. (http://ukoln.bath.ac.uk/follett_lectures/global_ref_room.html).

Lynch, C.; and Garcia-Molina, H. (1995, August 22). Interoperability, Scaling, and the Digital Libraries Research Agenda: A Report on the May 18-19, 1995 IITA Digital Libraries Workshop. (http://www-diglib.stanford.edu/diglib/pub/reports/iita-dlw/main.html). Also In Microcomputers for Information Management, 13 (2), 133-164, June 1996.

Maly, K.; French, J.; Selman, A.; and Fox, E. (1994). Wide Area Technical Report Service. Final Report for NSF Grant: NSF-CDA-9308259.

Negroponte, N. (1996). Ser Digital. Editorial Caminho (Being Digital - Portuguese ed.).

Nelson, M. L.; Gottlich, G. L.; Bianco, D. J.; Paulson, S. P.; Binkley, R. L.; Kellog, Y. D.; Beaumont, C. J.; Schmunk, R. B.; Kurtz, M. J.; Accomazzi, A.; and Syed, O. (1995). The NASA Technical Report Server. Internet Research: Electronic Network Applications and Policy, 5 (2): 25-36.

NIC Services. (1995). NSFNET Statistics. Gopher://nic.merit.edu:7043/11/nsfnet/statistics.

O’Reilly, T. (1996). Publishing Models for Internet Commerce. Communications of the ACM, 39 (6): 79-86.

Paepcke, A.; Cousins, S. B.; Garcia-Molina, H.; Hassan, S. W.; Ketchpel, S. P.; Rˆscheisen, M.; and Winograd, T. (1996, May). Using Distributed Objects for Digital Library Interoperability. IEEE Computer, 28 (5): 61-68.

Rusbridge, C. (1995, December). The UK Electronic Libraries Programme. D-Lib Maga-zine. (http://www.dlib.org/dlib/december95/briefings/12uk.html).

Schatz, B. (1995, April). Building the Interspace: The Illinois Digital Library Project. Communications of the ACM, 38 (4): 62-63.

Schatz, B.; Chen, H. (1996, May). Building Large-Scale Digital Libraries. IEEE Computer, 28 (5): 22-26.

Schatz, B.; Mischo, W. H.; Cole, T. W.; Hardin, J. B.; Bishop, A. P.; and Chen, H. (1996, May). Federating Diverse Collections of Scientific Literature. IEEE Computer, 28 (5): 28-36.

Smith, T. R. (1996. May). A Digital Library for Geographical Referenced Materials. IEEE Computer, 28 (5): 54-60.

Smith, T. R.; Frew, J. (1995, April). Alexandria Digital Library. Communications of the ACM, 38 (4): 61-62.

Stanford Digital Libraries Group. (1995, April). The Stanford Digital Library Project. Commu-nications of the ACM, 38 (4): 59-60.

Telematics. (1995). Telematics for Libraries: Call for Proposals 1995. http://www.echo.lu.

VanHeyningen, M. (1994, October). The Unified Computer Science Technical Report Index: Lessons in Indexing Diverse Resources. 2nd International World Wide Web Conference, WWW’94. pp. 535-543.

Wilensky, R. (1995, April). UC Berkeley’s Digital Library Project. Communications of the ACM, 38 (4): 60.

Wilensky, R. (1996, May). Toward Work-Centered Digital Information Services. IEEE Com-puter, 28 (5): 37-44.

Wractlar, H. D.; Kanade, T.; Smith, M. A.; and Stevens, S. M. (1996, May). Intelligent Access to Digital Video: Informedia Project. IEEE Computer, 28 (5): 46-52.