The University Licensing Program (TULIP), which started in early 1991 and concluded at the end of 1995, has been very rewarding for all partners, who have learned many valuable lessons in the course of the project that will enable them to better face the long transition phase towards digital libraries. Participants in this collaborative project were Elsevier Science and nine leading universities in the USA: Carnegie Mellon University, Cornell University, Georgia Institute of Technology, Massachusetts Institute of Technology, University of California (all campuses), University of Michigan, University of Tennessee, University of Washington and Virginia Polytechnic Institute and State University.

The goal of the project was to jointly test systems for networked delivery to, and use of journals at, the user’s desktop. In the TULIP project, the scanned page images plus bibliographic data and unedited, OCR generated, “raw” ASCII full text of 43 Elsevier and Pergamon materials science and engineering journals, were provided by Elsevier to the universities, which developed or adapted systems to deliver these journals in electronic form to the desktops of their end users. The focus of the research was on technical issues, on user behavior and on organizational and economic questions. The major focus of at least the first half of the project, was at the technical side of TULIP. When this project started, there were very few institutions willing or able to bring up large scale implementations, aimed at bringing primary information to the desktop on their entire campuses. Many of the participants did just that. The lessons learned here, have already had an important impact on the implementation of the digital library (components) at the participating universities, as well as on the directions Elsevier is taking. Technical conclusions:

• Most universities decided to “shift to the Web”, thereby abandoning X-Windows and MS-Windows applications, as the advantages plus the sheer user-pull outweigh any disadvantages such as fewer possibilties to provide “real-time” functionality (e.g. image-zooming).

• Large-scale Internet FTP transfer is not feasible whith the current transmission schemes and restricted bandwidth. Scalability of TULIP-like systems, will also be hindered by the limits of current massive storage technology. It is expected that a “staged” approach to electronic collection building will emerge, which is composed of local servers for primary relevant material and remote (perhaps regional) servers for material of secondary importance to the particular institution.

• Speed is crucial for image viewing on the screen. The components of the system, influencing the (perceived) speed, are the server/storage speed, the network speed, the client machine speed and application software image caching “smarts”.

• Printing page images is an important concern, but becomes viable with the advent of printers, that understand compressed Group IV fax images, and require careful attention to setup. Printing of images from other than directly IP-attached laser printers, should be avoided.

The objective of the user behavior research, was to obtain specific feedback about TULIP from endusers, to guide future developments of delivering journal information to the desktop, in order to get insights on the requirements for electronic services to be attractive and valuable, both from the content provider’s and infrastructure provider’s side.

Two types of research were done: quantitative and qualitative. The quantitative research consisted mainly of analysis of the logfiles, i.e. records of user actions. Penetration, defined here as a function of repeat usage, rather than just number of actions, varied widely among the universities. The qualitative research, which consisted mainly of focus groups and one-on-one interviews, using basically the same interview guides for each site, was aimed at answering the questions raised by the quantitative research. For instance, what gets (and keeps) users interested in electronic products; which requirements should electronic productus meet to be attractive and valuable; how should we bring electronic (full text) information to the desktop?

Users seem to have the following requirements, regarding functionality, ease of use and critical mass of electronic information services:

• ease to use: as intuitive as possible, and preferably using a familiar interface;
• access to all information from one source;
• effective search capabilities;
• high processing speed (downloading and printing);
• high publishing speed (timeliness of the information);
• good image text quality;
• sufficient journal and time coverage;
• linking of information.

This is the enduser’s definition of convenience.

Other conclusions to be drawn from the user studies, include the following:

• The general concept underlying the TULIP project is very well received by students and faculty. This concept consists of desktop access to full text/image articles; fast and easy to search, to read and to print.
• Hardware and software are serious obstacles for convenient use of the TULIP information at most sites.
• Most users consider the coverage (in number of titles and in time) of the journals in the TULIP project to be insufficient. This insufficient coverage (not all core journals - those not published by Elsevier - are available in the database) requires endusers to search additional information elsewhere, which is considered time-consuming and redundant; not contributing to an increased convenience.
• Graduate students use TULIP more frequently than faculty.
• There is enthusiasm about the concept of desktop access to electronic information, but the end of paper products seems to be far away still. Besides some practical benefits of paper products, there also seem to be “emotional” ties with paper and the library.
• Although there clearly are a whole suite of conditions for a service such as TULIP to become a regularly used tool, such as overcoming the technical hurdles mentioned above, meeting user expectations, sufficient coverage, etc., promotion does play an important and continuous role in the degree of success that can be achieved. Recognizing the fact that meeting user needs is primary; promotion and training are crucial for a service such as TULIP, to develop a base of regular users.

The organizational and economic issues summarized below, have been derived mainly from a series of interviews with key players at the universities.

There seem to be four major factors which can make or break a project like TULIP:

• the (project) organization, that is: dedicated project management, cooperation among the parties involved and having the right resources available at the right time. Politics, lack of priority and lack of responsibilities can cause long delays and, have all but killed the project in a few of the TULIP universities;
• understanding your user community’s possibilities and needs, is also a key; a few universities stated, that they felt, in hindsight, that the first step of the project should have been to consult and involve their endusers;
• if there is no adequate infrastructure in place, a project like TULIP can not be implemented, even with committed team members, who know what their users want. Infrastructure includes systems and systems development (most universities stated they had underestimated this), networking and (non-local) printing infrastructure, the ability of the (inter-)campus networks to deal with bandwidth-consuming graphical data, and user desktop systems;
• promotion, which was not viewed as a major issue at the beginning of the project at most universities, is clearly important for electronic information services. At sites, which did more extensive promotion, we see a significantly higher use of TULIP.

Although the universities and Elsevier Science have not resolved one critical issue, that of how to make the transition to digital libraries work ecenomically, TULIP still helped us a great deal in the development of ideas about these issues, specifically about cost. All participants have been confronted with the harsh economic realities of building even a prototype of an electronic journals system: on Elsevier’s side, the costs of conversion and distribution, on the universities’ side, the costs of implementation of their respective systems. These experiences guide us in determining what we want to and can do in a full scale production. Some universities referred to high costs, especially of storage, as an important factor inhibiting the development of their TULIP systems, and view these costs as a major roadblock for further development. However, at MIT the effort hase been characterized as a “relatively inexpensive effort with high return on investment”.

As part of the process of evaluating the TULIP project, university participants were also asked their views on what they have learned form the TULIP experience.

“Too often, work on digital libraries, not to mention much theoretical discussion, proceeds without a thorough grounding in the realities of cost. There are certain assumptions, which precede this state of affairs, among them the notion that digital libraries somehow will be cheaper than print libraries, perhaps even free. One suspects that this arises from the misplaced hope that digital libraries will liberate us from the difficult cost dynamics of print libraries.” TULIP proved to its participants, beyond a doubt, that building digital libraries will be a costly and lenghty process. Also, we can say, that making additional funds available for electronic content, will not be a trivial issue for these universities.

The TULIP participants’ view on the role of the libraries in the development of digital libraries reflects this focus on a lengthy transition period. Many information specialists said that the crucial role for libraries is the role of agents for the universities’ information needs. “What I see, is moving from content to context. Right now, we deliver page images, but in the future, libraries will be delivering an array of information in a lot of different digital media, that provide an information environment for the user”.

All see the role of libraries increasing instead of decreasing, fulfilling the following roles: finding, selecting and providing the information needed by the community, and leading people to the right information, as well as protection of holdings.

“ If there are no libraries, users are going to be confronted by a lot of inconsistent interfaces, financial arrangements, delivery vehicles ... and the library can, at least potentially, add value, by making that coherent”.

However, the endusers are seen as the driving force behind a lot of what libraries are going to do, and they will often be one step ahead. “It all depends on how individuals value information”.

Most see a continued role for publishers: “Publishers enhance the credibility of information”. More informal and preprint publishing is expected, but in the overflow of information, pulbishers can help in selecting the quality information. “I don’t see the publisher’s contribution to the scholarly process changing drastically. I think the way they do it, is dramatically changing. But they remain the same in terms of managing a process and adding value to content”. There does not seem to be a consensus among participants about the role of publishers in the archiving of the electronic content. While some universities are quite willing to let the publisher be the ultimate archive for electronical material, others are much more apprehensive to leave this role up to the publisher, either for reasons of continuous availability (publishers can go broke) or for reasons of continuous access.

A common view, which all TULIP participants share is, that the transition to a digital library will go slower than they had expected before starting the project. No one can predict the actual transition speed, but TULIP generated some more important insights, concerning this question:

• At the moment, managing large digital collections locally, is harder and more expensive than managing a comparable print collection.
• Not everyone is ready for digital collections, nor will they be soon. Saying it is harder and more expensive, is to say that the leading edge market is still small. Whether one talks of large local stores of data, or regional networked collections, or single remote hosts, the number of academic libraries, really ready to support digital collections, is still small.
• Users will only move to electronic publications when they find the content they need in sufficient quantity. Having journals in electronic form and bringing them to the desktop, are necessary but not sufficient conditions for the scholarly user. You must deliver a certain “critical mass” of needed information, to warrant learning a new system or accessing information in a different way.
• For the publisher, expanding electronic publishing (internationally) offers challenges very different from paper publishing. Elsevier Science has been delivering a uniform product, which required little local training or support all over the world. With electronics, this is not the same; the process is more complicated and requires a different involvement from the publisher.

--------------------------------------------------------------------------------

A personal note

Almost exactly five years ago, I started out on an adventure with a group of innovative and stimulating university librarians and university computing center directors. We decided to take a number of highly-visible risks in search of learning more about moving large scale journal delivery from paper to electronic and from the library shelves to the users’ desktops. We started and stayed within the supportive ambiance of the Coalition for Networked Information and I want to thank Paul Evan Peters for his steady interest and support.

Some of my original colleagues in the participating universities are now in other institutions. Others have been with the TULIP project from the very first meting. My own day-to-day role diminished after the first years, due to the very capable management provided by the Elsevier Project Manager, Jaco Zijlstra, and the equally dedicated supporting teams in the universities and within Elsevier.

Karen Hunter
Elsevier

--------------------------------------------------------------------------------

Introduction

This is the concluding report on The University Licensing Program (TULIP), a collaborative project of Elsevier Science and nine leading universities in the USA. This report reflects the experiences gained and research conducted in the project. All universities have contributed a report on their implementation and experiences. These reports, combined with input from a series of interviews conducted with key participants at the universities, as well as input from the Elsevier Science staff involved in the project, are the basis for this report. The university reports are attached as appendices I-IX. The first four chapters of this report contain the report per se: describing the project and its findings. This part starts with a description of the participants (chapter I), followed by a description of the technical aspects of the project (chapter II). The third chapter deals with promotion. The fourth chapter contains a detailed account of the user behavior encountered in the project as well as of the research done to try to explain and learn from user behavior. The second part of the report contains the conclusions drawn by participants at the end of the project and the recommendations that can be derived from the TULIP experience, which could be of value for others contemplating a “digital library” project. These experiences have furthermore been summarized in appendix X (a checklist for institutions contemplating a TULIP-like implementation, and a summary of aspects to be considered). In the fifth chapter, organizational and economic issues are discussed. Finally, in chapter VI, the implications of this project for the future of digital libraries is addressed.