"Lessons Learned" suggests a collection of papers dedicated to failure, so including an article about the Internet Public Library, is in some ways surprising. The IPL has achieved all its initial goals and lived up to its creator's hopes. However, in the high-stakes, high-cost world of Web publishing, that may not be enough.

In the Beginning . . .

Joe Janes, an assistant professor at the School of Information and Library Studies at the University of Michigan[1], wanted to put a new spin on "ILS-726: Information Technology, Impacts and Implications," a graduate-level survey course that he had already taught several times. For the spring semester, 1995, he wanted to try to integrate library studies with the World Wide Web. He had been involved in the University of Michigan Digital Library project[2], and he wanted to further explore the merger of networking and libraries by planning, building, and running a digital library on the Internet based on the public-library model.

His idea was that he could do more than just replicate the functions and processes of a real public library, or add to the long lists of digital resources that were then available on the Internet — resources that had little intellectual control or input from the library community. He wanted to create a hybrid that combined the strengths of both public libraries and the lists of links that attempted to categorize Internet resources. His Web-based library would feature such standard public-library services as reference, cataloging, educational outreach, exhibits, a children's space, and popular reading. It would also have some features less common in public libraries, such as government documents, special collections and archives, and serials, as well as online-only services, such as a reference MOO. [3]

Undazzled by the technology, Joe vowed his new library would never lose sight of the audience it served. Answering such questions as "Who is our public?" and "How will we best serve them?" would always be the library's primary mission.

When ILS 726 was offered in the spring, the construction of an "Internet Public Library" was the class project. Students at the School of Information and Library Studies had been invited to apply to take the course, and Joe selected thirty-five applicants. For Joe, dedicating a class of top-notch graduate students and an entire semester to forming a public library on the Internet was exciting, challenging, and frightening. One problem that arose immediately was the possibility of success. If the class succeeded in creating that new entity, what would happen to the library when the semester ended? Sara Ryan, an IPL staff member and former ILS-726 student remembers,

We approached this project not as an exercise, but as something very real. There was so much energy involved in the IPL's creation that we couldn't just end it when the class was over. Early on in the semester we had started a listserv about the IPL and this had built a community of believers with high expectations. We simply had to find a way to keep it running or risk disappointing all those people.

I had met Joe more than a year earlier when we were both involved in a collaboration between the University of Michigan Press, where I worked, and the Library School. I offered my support when he first broached the idea of the library. Shortly before the IPL made its debut in March 1995, Joe told me that he and the students had invested so much energy in creating the library that they were willing to go to great lengths — including working for free — to keep it running. They decided that when the semester ended they would continue with their development work and look for funding to keep the library going.

From the moment it went online, the IPL was an astounding success. The first week it had over 10,000 hits and as the semester drew to a close the accolades poured in. The library was such a success that the library school awarded Joe a Kellogg grant of $150,000 over three years to help sustain it. From that grant, and others that came later, he hired five students from the ILS 726 class, Schelle Simcox, Nettie Lagace, Sara Ryan, Michael McClennen, and David Carter. The problem of how to continue the IPL — at least in the short term — had been solved.

Building the Business

In the summer of 1995 the IPL staff continued expanding the library. They offered constantly updated electronic versions of classic library resources, such as magazines and serials, online texts, newspapers, and online searching. They created a unique online reference department where anyone could send a query by e-mail and have it answered within 48 hours. They opened a "children's room" where kids could read and listen to original stories, interview their favorite book authors, find resources to help them with their schoolwork, and pose questions about science to Dr. Internet. And the teen division featured online resources such as the A Plus Research and Writing Guide, and Career Pathways: How to figure out what to do with the rest of your life. The IPL also created an "exhibit hall" where they featured such multimedia displays as Pueblo Pottery and Music History 101: A Basic Guide to Western Composers and Their Music. The IPL had all the resources of a local public library as well some that were unique to the Web.

"Even their most enthusiastic patrons became less enthusiastic when it was suggested that they might need to pay for services."

And Internet users loved it. The IPL won awards from MacUser and PCWeek magazines, it was chosen as Librarian's Site of the Week, and letters of support rolled in. The IPL staff was pleased, but they knew that they couldn't continue developing the library at their current funding level. They had to find more support.

The University, while happy to provide a server, an Internet connection, and some seed money, had made it clear that long-term support was not an option. So with the Kellogg grant covering their short-term expenses, the IPL staff turned its efforts to making the IPL self-supporting.

Schelle Simcox, assistant director and head of fundraising, says that although she was moderately successful at raising cash, she had no idea how difficult it would be. A business plan that the staff drew up in early 1996 was sufficient to win them a $200,000 grant from the Mellon Foundation, but that was all they had to show for the many letters, phone calls and presentations they'd made. Many funding agencies were interested in the IPL's unique online library, but felt that there was not enough potential return on investment for them to actually write a check. Unfortunately, funding agencies weren't the only ones who couldn't bring themselves to pay: The IPL staff found that even their most enthusiastic patrons became less enthusiastic when it was suggested that they might need to pay for services.

As it turns out, the IPL staff had stumbled onto one of the most difficult problems facing any business operating on the World Wide Web today: Users believe that information on the Internet is supposed to be free, no matter how much it costs to put that information online. Any number of promising Web publications — some with much better financial backing than the IPL — have been laid low by that notion. For instance, Web Review, an online publication launched in 1995 by Songline Studios was a hip, polished, weekly 'zine[4] dedicated to covering what was new on the World Wide Web. It garnered rave reviews from both the media and its readers for its timely and thoughtful reporting produced by a staff of 20 writers, editors, and designers. When online advertising space didn't sell as expected, Web Review realized that it needed to charge a modest subscription fee to stay in business. Not surprisingly, even the most avid readers balked, and Web Review was forced stop publishing. It was only through a partnership with Miller-Freeman, a well-established journals publisher with a solid financial base (and a desire for an online magazine) that Web Review was resurrected — without subscription fees.

That is the dilemma facing everyone who wants to publish on the Web: People don't see how much it costs to publish quality content and how much skill goes into preparing that information. Print publishers never directly confront that problem because people assume (erroneously) that what they are buying in a book or magazine is represented by the paper on which it is printed. People cannot assign appropriate value to online information.

For the IPL, the dilemma of "information valuation" was compounded by the fact that the IPL was based on a model that didn't completely apply. It called itself a public library, invoking visions of free services; but those services, as Joe Janes explains, are never really free.

People think the public library is free, because they don't have to hand over any money whenever they use its services. But it didn't used to be that way; public libraries were actually subscription libraries. People paid a set fee for each service and that went on for several hundred years. In the late 1800s Andrew Carnegie stepped in, donated a lot of money for buildings, and got the government to support these new public libraries with tax revenues. And this system has been in place — virtually unchanged — since the 1920s. So now what happens is you have all these librarians saying that everything is free.

These services only look free. In reality, everything is tax-supported. Of course we'll answer your questions, we'll make these photocopies for you, we'll order whatever you want through interlibrary loan — if you live in this community. People start to believe that the library is free because they don't see any fees; they don't see the transaction; even though everyone in the community pays a fee once a year, on April 15. And people don't realize that the library has to pay for stuff too; all these costs have been hidden — even from librarians themselves!

The IPL made a tactical error in embracing the public-library model. The founders, caught up in the excitement of creating services for the Web public, neglected to remember how real public libraries are funded. Community libraries get money from each member of the community through town or city taxes. They are more like health clubs that charge an annual fee than like grocery stores that charge for each item. The health-club dues allow each member unlimited use of the facilities and certain priviliges at no additional charge. Instead of buying treadmills, the library uses its share of the city's tax revenue to pay a staff, maintain buildings and collections of intellectual properties, and acquire new materials. Any donations to the library go toward extras like multiple copies of books, compact discs, or specialized programs such as field trips for children with special needs. Unlike its real-world counterpart, the IPL serves the entire World Wide Web — a virtual community that has never voted to have a library, let alone support it with tax revenue.

"If you want to be an information provider you need to have someone on your team with the ability to bring in a steady source of income."

For that reason, the future of the IPL is uncertain. At one point this past summer the existing grants could no longer cover staff salaries, and it appeared that the project would have to scale back radically or shut down completely. Anticipating the end, several staff members left the IPL to take positions at public and university libraries. In August 1997 the library school once again came to the rescue and gave IPL a supplemental grant, which was enough to employ two administrative staff members. That money will allow the library to stay open until April 1998.

The Lessons Learned

Schelle Simcox, who left the IPL staff when funding ran out in July, says that the primary lesson of IPL is the need for strategic marketing and planning. "If you want to be an information provider you need to have someone on your team with the ability to bring in a steady source of income. That person should be a marketer who can sell your ideas to an audience that isn't quite ready for them; someone with the ability to pull people together, get them excited about ideas that are unproven, and convince them to offer financial support."

Joe Janes thinks that the IPL was trying to operate in an arena that was not ready for what it could offer:

The whole world of information is up for grabs. Gutenberg fixed things — in a sense made them stand still — for a very long time in the fairly slow-moving but nonetheless fluctuating world of print. We'd been living that way for 500 years when literally, almost overnight, the Internet arrived. And now we have a Website, a CD-ROM, virtual reality, a book — sometimes all four or more rolled up into one product! And it moves, it throbs, it vibrates, it sings; it's altogether different, and yet it's still the same stuff underneath. I don't think that societally or culturally we've got the right handles, yet, to think about digital information. We're still in the translation phase, working toward our first real understanding of what the issues are. We're neither a generation nor a technology sophisticated enough to grasp what it really is we can do with what we've got. Electronic text creates all these new realities for how you use information. But until the technology is sufficiently developed to fully support digital information, and the public understands and figures out what to do with it, progress will be pretty slow.

And in the End

The Internet Public Library has been ahead of its time; the Internet is not yet ready to support it. But Joe Janes and his students met their original goals:

  • serving the public by finding, evaluating, selecting, organizing, describing, and creating quality information resources;
  • developing and providing services for our community with an awareness of the different needs of young people;
  • creating a strong, coherent sense of place on the Internet, while ensuring that our library remains a useful and consistently innovative environment as well as fun and easy to use;
  • working with others, especially other libraries and librarians, on projects that will help us all learn more about what works in this environment;
  • upholding the values important to librarians, in particular those expressed in the Library Bill of Rights.

The fact that the IPL has been unable to raise enough capital to support itself is disappointing but not surprising, given its circumstances and the climate of the Internet as a whole. Yet its accomplishments are impressive: Very few public libraries can claim that they've served more than five million people in two years with a staff of six and a budget of less than $450,000. And very few libraries of that size have invented a computer-based method of accurately answering more than five hundred reference questions a week, or a MOO dedicated to reference. Certainly few other libraries can look at newly created Web sites and see tools, architectures, and information strategies that they created. The IPL was the first digital library of its kind, and being the first of anything is always risky. But by taking that risk the IPL can serve as a lesson, and perhaps make the road easier for digital libraries in the future.

Lorrie LeJeune is Product Marketing Manager for Technical Publishing at O'Reilly & Associates. In previous incarnations she was an Electronic Publishing Specialist at the MIT Press and the University of Michigan Press. At Michigan she cofounded The Journal of Electronic Publishing. At the time she was also a graduate student in Information and Library Studies.

These days she spends most of her time monitoring Internet and Web technology, analyzing the computer-book market, and explaining why O'Reilly puts animals on the covers of its books. A B.S. in Animal and Veterinary Science from the University of Maine helps her make those explanations more colorful.


1. The School has since changed its name to the School of Information.return to text

2. The University of Michigan was one of six universities awarded funding by the National Science Foundation to build collections of digital documents with sophisticated search engines, user interfaces, and other advanced features.return to text

3. MOO is an acronym for Multiple-user, Object-Oriented environment.return to text

4. World Wide Web slang for an online magazine.return to text