I've been called an unreconstructed idealist by people I call my friends, and many of you have heard me talking about the Internet, and publishing technologies, and scholarly publishing, in ways that would embarass many. Well, I'm here to do it again, with even better reason.

Several months ago I was asked to give a talk at the OCLC's annual library director's conference. The keynote speaker was the new president of the University of Arizona. He drew a marvelously lucid contrast between the commercial and noncommercial sectors which is worth retelling. "The commercial sector," he said, "tries to make as much money as possible within the constraints of society. The noncommercial sector tries to do as much good for society as possible, within the constraints of their money."

Within an hour, another major speaker, a woman from MegaTrends, Inc., the authors of MegaTrends, spoke stirringly about her MegaPredictions regarding the future of libraries. "Libraries are the holders of so much organized intellectual content," she said, "that you could make a great business digitizing it and selling it online." She challenged them to become more entrepreneurial, and think of themselves more as businesses, because that was the future.

Libraries as moneymaking entrepreneurs? It made my skin crawl. But it fits with the trend of many universities hiring leaders who are "bottom line oriented." It fits with the notion of university presses as "profit centers" for the university. We see this maximum-money perspective infecting the educational superstructure, to its detrement. Its logical conclusion includes charging for every paragraph, every live reference, every glance.

So who are we, anyway? Are we profit centers? If we are, then what are we doing publishing these lame loser monographs that sell only 500 copies? Why aren't we doing more books on, say, the social history of the corset, which sell so much better? If we're entrepreneurs, then what are we doing with copyediting, quality control, and the like--especially for those loser publications with virtually no audience?

I know the answers, and so do you--we are some strange hybrid between entrepreneur and do-gooder. We are both an independent business and a service to scholarship, simultaneously virtuous and venal.

This tension may lead to efficiencies, but the paradoxes may also lead to schizophrenia.

What does this have to do with electronic publishing? A great deal, surprisingly. This tension between the dark and light side of the Force, as it were, is also linked to some dividing lines I'm beginning to see online, and that find some manifestations in electronic publishing.

The tools for electronic presentation are now almost mundanely easy--any of us could have our compositors generate HTML from our typesetting files, for a handful of hundreds of dollars per book at composition time. We all have websites, and know that we can plop whatever we want onto those sites. But we don't put up those electronic files because we're not willing to give something away for free. It goes against all our habits.

Making the material available online isn't the problem-- the problem is in charging for electronic versions of our publications. The transaction costs, in time, necessary expertise and staffing, and developing access-restriction systems, are too great.

When I was at Project Muse, Johns Hopkins University Press's online journals project, I once did some blunt-pencil analysis that indicated that something over half of the costs of the online journals project was attributable to systems for preventing access to the articles. While the specific numbers and definitions can be argued, the basic principle stands: we were spending hundreds of thousands of dollars to ensure some people *didn't* read what we published. The principle stands, and it stands as an absurdity. What could be done if we weren't trying to keep people away from our stuff?

I've been finding that out over the last eighteen months. I now work at the National Academy Press, as many of you know. We publish the reports of the National Academy of Sciences, National Research Council, Institute of Medicine, and National Academy of Engineering--about 200 reports per year. We're unusual in that we're as much a service for our institution as a business. We have to be self-sustaining through sales, but we also occasionally get plums like the book version of the National Science Education Standards dropped into our laps that sell hundreds of thousands of copies. It's worth noting that we also are required to publish such likely bestsellers as Opportunities and Priorities in Arctic Geoscience, or Ruminant Nitrogen Usage.

For the last four years, the NAP has put its entire frontlist and much of its backlist up in page image format, for free. Over the last eighteen months, we've developed and implemented a unified online book access system designed to make our material maximally available online for free. It's an online research tool that makes over 1400 titles available via searchable full text and page image presentations.

By giving this material away, we're succeeding in increasing our sales. People find publications they didn't know existed; they can browse a book they can't find in any bookstore. Backlist titles can be explored online, and then purchased. The book becomes its best advertisement, especially as the search engines--Alta Vista, Yahoo, and the rest--index the material.

Right now we're receiving an average of 60 online orders per day, for over $3,000 per day. Yes, it's science policy, not humanities, and yes, certain titles sell more than others, and yes, we're giving 20% off for online orders, and yes, we historically sell to individuals more than most UPs have. But it's also a striking indication of what new models are possible online.

In this case, customers are paying for the container, not just for the content. People want to read paper, and want to own the book. It's the library uses that we're replacing--that is, the uses for which people don't pay us anyway.

We are putting in place a subscription system to be able to charge for rapidly printable PDF, DjVu, or other high-quality files--and some people will happily pay to be able to print out chapters, articles, even whole books, because they'll want it fast, and clean. We will continue to make the OpenBook page images searchable, browsable, even readable for free--because what better advertisement for these books can there be than the book itself? How costly is it to market a publication on Opportunities and Priorities in Arctic Geoscience, to sell a few hundred copies? Much more costly than making it available online.

What lessons are there for you in all this? Plenty, I hope. I believe that the OpenBook approach is indicative of what more of us should be doing--refreshing our core mission to scholarship, applying what it is that makes us different from the commercial publishers.

I want to hearken back to some themes I began this talk with. This experiment wouldn't have happened if NAP hadn't been following the dissemination mission of the institution. If we had begun by trying to find ways to squeeze every possible dollar out of our material, a whole different model would have developed.

Instead, this model--of free browsing, easy access, and researcher-friendly publication first, and sale second--is proving itself to be much more in keeping with the role of a noncommercial publisher. I like to think it points in the direction that university presses, and other noncommercial publishers, should be looking.

So what are we doing publishing lame losers that sell under 500 copies? We're doing what we should be doing--ensuring that quality research and valuable scholarship is given professional treatment, and is provided to interested readers in the highest form possible, within the constraints of our money.

What is it that university presses do? We add value. We make public. We ensure quality. We select, and validate, and enhance, and market. We take risks on publications, and provide professional publishing services to enhance the likelihood of success. In exchange, we have jobs that are intellectually challenging and make a positive difference in the world.

We're noncommercial publishers, whose mission is to do the most good for society as possible within the constraints of our money. We happen to be in an unusual moment in history, when these new tools can be successfully applied to perform our missions in unparalleled ways.

And here's the kicker--if we don't do it, or if we do it wrong, our authors will do it themselves, one way or another. The MegaTrend here is that scholarly communication and research is increasingly online--and that's where we have to be too, if we want to appeal to our authors.

What the NAP is doing is giving away our content and increasing sales of our print publications; we're succeeding in both our missions. I believe that all of us could be doing that, if we were willing to take a small risk. If we make our content available online, in HTML or OpenBook or some other format, we can help create a collection of resources that helps scholarship, helps scholars, helps education, and helps society, while still recovering the costs of professional publishing, increasing our visibility to our audience, and enhancing the longevity of our profession.

As an unreconstructed idealist, I unabashedly believe that we can do well by doing good, and believe we can change the world in many small, valuable ways, not only by what we publish, but also by how we publish it.