The Death of Print, Xanadu and Other Nightmares, or,
Brother, Can You Paradigm?

Walt Crawford
Copyright © 1992 Walt Crawford
Originally Written Summer 1992

Electronic publishing surely dooms books, magazines and daily newspapers. Better yet, we should all be preparing everything in multimedia, making mini-movies rather than writing memos. In any case, an old-fashioned book is nothing but a bunch of paragraphs strung together. Let people get at the paragraphs they really need, independently, through a worldwide universal hypertext system. That will surely improve our lives, at least until we conquer space, time and urban depression through the wonders of virtual reality. There are no dangers in new technology, and old technology doesn’t stand a chance: that’s the miracle of paradigm shifts.

If your response to all this is "I believe, I do," this may not be the right program for you. I sometimes write a column for an electronic journal and serve on its editorial board; I rely heavily on electronic mail. But I have also written ten books and a few dozen articles, serve as editor of the LITA Newsletter, have produced five books on LITA’s behalf this year, and serve on editorial boards of three print periodicals. I believe in electronic publication—as another medium, not as a wholesale replacement for print.

Today, I hope to have some fun with technolust and suggest a few reasons that printed books, magazines and physical libraries aren’t likely to go away any time soon. I also hope to make some serious points in this hour, namely the following:

First, the new is not always better than the old. Introduction of a new technology does not assure its rapid or inevitable success, and new technology rarely sweeps away old technology unless the old technology is inherently flawed and the new technology is overwhelmingly superior.

Second, you should be on your guard when something is described as inevitable—particularly if the inevitable development seems undesirable or questionable to you. Almost nothing in the affairs of humanity is inevitable. If someone has an overwhelmingly strong case, they should state the case. Generally, the inevitability card is played because the factual case is less than convincing.

Third, it’s as pointless and harmful to treat all libraries identically as it is to treat all library users identically. It’s bad enough that many university libraries seem forced to make overall decisions based on specific problems that primarily affect their science, technology and medical collections. It’s even worse when liberal-arts college libraries behave as though the same problems were their only major concern. It becomes totally absurd when people assume that the same concerns should dominate the thinking and planning of community college libraries, public libraries and school libraries. That’s just as unfortunate as suggesting that the information needs of, and best tools for, a working mother, a dyslexic teenager or an unemployed job-seeker are the same as those of a Ph.D. candidate in biochemistry.

Fourth, data is not the same as information; information is not the same as knowledge; knowledge is not the same as understanding; and understanding is not the same as wisdom. Beyond that, libraries, books, magazines and daily newspapers play many roles beyond simply providing data, information, or knowledge.

Fifth, predictions tend to be self-fulfilling if enough people make them or accept them. To a very great extent, we get the futures that we work for; to an even greater extent, we get the futures that we settle for.

Sixth and last, the problem with paradigm shifts (or, to avoid that cliche, the creation of new archetypes) is that they seem to assume going from one stable situation to another (but very different) stable situation—after which we can stop thinking about it. The reality is change: sometimes faster, sometimes slower, always complex and somewhat unpredictable. Libraries and librarians should constantly be redefining themselves in small ways; that’s very different than some wholesale redefinition of who we are and what we do.

That’s more than enough ground to try to cover, particularly as I try to bring in universal hypertext, Captain Picard, the importance of things not wholly read, and dozens of other points. Before moving on, I should add a disclaimer. This talk represents my own opinions. I am not speaking on behalf of the Research Libraries Group, and none of the positions taken here should be assumed to represent RLG positions. I am, of course, speaking as the president of LITA—but even if LITA took positions on issues of this sort (which we normally don’t), the division includes more than five thousand intelligent people and organizations—and there’s no reason to believe that any or all of them would agree with what I’m saying here.


I make my living thanks to computers. I have a personal computer at home. My wife has a notebook computer at home. I use a PC, on a LAN connected to a mainframe at work. I would never want to go back to writing with a typewriter, let alone by hand. I like technology, at least when it solves my problems—and I believe in using technology to solve problems created by technology. But I’m a tool-user, not a techno-junkie. Let’s look at some aspects of technolust.

Personal Data Assistants and Other Wonders

The hottest new items for avid technophiles are Apple’s Personal Digital Assistants, the first of which is Newton: special-purpose computerized devices to make life more wonderful for all of us. PACS-L had a thread about what Newton and other PDAs could mean for libraries, and the true technophiles had wonderful ideas, such as people wandering around with their little PDAs in hand, getting access to the library’s catalog through infrared links, maybe scanning in pieces from books on the shelf (until we get rid of those books...), and so on. One even went so far as to suggest that battery-operated pen-controlled portable computers might be more ecologically sound than all those paper notebooks.

I won’t dwell on the ecological implications of substituting heavy metals in the waste stream such as cadmium (used in nicad batteries) for readily-recyclable paper, which is made from renewable resources. I think I know which way that particular balance goes, but never mind.

Newton is new enough so that it’s only a little obsolete. It isn’t on the market yet, and there’s no firm date set for it. The price will be "something under $1,000"—which certainly suggests that every library user should have at least one of them handy, doesn’t it? Of course, Newton may be a poor example; as I understand it, it’s basically a personal calendar and appointment book with room for note-taking.

My immediate reaction to Newton was that it was the latest Yuppie Scum Toy, designed to restore Apple’s once-enviable profit margin. On further thought, I believe that reaction is sound. It’s safe to assert that devices such as Newton will do very little to narrow the gap between haves and have-nots—except, of course, that they will help to keep well-to-do fools busy buying and learning to use these new toys, thus keeping them out of our hair.

It’s demonstration time. (Pull PocketPal out of pocket). I have here my own personal data assistant, which just happens to be a pen-operated calendar and appointment system. I own two different ones; this is the more portable of the two, although not by much. It’s quite a unit; small, light, compact, with a display as legible as ink on paper, and I have yet to replace any batteries. Then there’s durability, since these portable devices do get tossed around. Let me show you. (Drop PocketPal on ground; jump on it a couple of times.) I just dropped this device from a five-foot height and jumped on it twice. And guess what? It still works exactly the same as before.

The best part is the price: $5.25. Actually, these are cheap enough that some years, publishers send them to me as gifts. Here it is: they call it a PocketPal, and it certainly meets all the calendar and appointment needs that I have. Oh, my bigger one costs almost $20; it’s a DayTimer, and I use it to keep detailed logs of work time. I understand that there’s now an electronic DayTimer program to run on pen-operated portable computers, actually developed with the company’s cooperation. This $200 program, when combined with a $1000 computer, will give you all the functionality of a $20 DayTimer, as long as you keep replacing batteries.

But why?

If you suffer from terminal technolust, the answer is that everything’s better if a computer is involved—or, as one person put it, "I would rather have a thousand technical failures crowding the marketplace than to be content with doing things the same old way." If you hate your present life that much, if pens and pencils revolt you so, then help is on the way—as long as you’ve got the money, of course. But if you do have that kind of money to waste—I’m sorry, spend—then don’t pretend that you’re one of the people. You’re part of an elite, and if you insist on libraries spending their money to accommodate your hot new machines, you’re saying that your needs for my tax dollars count for more than the needs of the disadvantaged and those who aren’t embarrassed by appropriate technology.

Male Fantasies and Virtual Reality?

Ed Valauskas, who is surely no technophobe—among other things, he’s co-creater of Macintoshed Libraries, available both in print and as HyperCard stacks—commented on virtual libraries in his "Letter from the Frontier" in the April 1992 Apple Library Users Group Newsletter. He recounted studies by Cornelia Brunner that show differences in the way men and women view technology. Oversimplifying, you could say that Real Men want to recreate the universe through technology—in other words, they suffer from terminal TechnoLust—whereas technologically-knowledgeable women want to use technology to connect people. I’ll simplify that to the tool-user’s approach: technology provides tools toward ends, not the ends themselves.

As Valauskas points out, most virtual library advocates are male. To quote, "What a better way to ‘correct’ our [librarians’] social image than by trashing entirely the notion of books in paper, reading words sequentially, using indices to locate ideas. Instead, hyperanthropos, Lucian’s HyperMan, running the operation at his mighty computer keyboard..."

Valauskas’ own conclusion: "Give me real books..for the time being, and a real library. I may have CD-ROMs, electronic journals, microfilm, microfiche, books and periodicals on diskette, software, cassettes, computers, and videos in my library, but paper still seems to be a quite popular medium. And the statistics back me up... Let’s just hope that we use technology not as a means to re-invent libraries in a ridiculous self-image, but as a vehicle to reach a broader audience even better than we do right now..."

The true believers call for the American Library Association and other similar associations to spend our resources in furthering this goal, as being the only real future for all libraries. After all, what is a public library but a miniature university library? Nobody really wants to go to libraries, do they—they just want those neat little packets of information to be found there. You say a worker with relatively limited command of English wants a book to help her repair some plumbing at home? Well, she can just tap into her home computer at $12/hour, and get quick access to the latest and greatest information on evolving trends in plumbing. The newest and most authoritative information: that’s what we all need, right? Oh, sorry, that doesn’t leave any money for novels, for Readers’ Digest Do-It-Yourself Books, for English-as-a-Second-Language videos, and for all that other trash. Too bad. Can’t afford $12/hour, and don’t have a home computer? Too bad: you’re not the public for a real, contemporary public library.

Sorry. I got carried away there. Not quite as carried away as the true believers get, to be sure. But read their informal messages; you’ll see that some of them are quite serious in asserting that nobody wants to use libraries except to get little packets of information, that any librarian or association who doesn’t devote full-time efforts to the great digital future is a fool, that theirs is the only way. And yes, there is something remarkably macho about the whole business, even though some women are also involved.

Sweeping Aside Reality

The heart of technolust is an unwillingness to deal with the real world. New is always better; technology is always a good thing; once something works, it’s time to look for the next new wave. Jerry Pournelle, a prime techno-junkie, tells us that the thing we now think is far too powerful or arcane will be commonplace in a year, and that "in two years, you won’t know how you’ve lived without it." And if you don’t know how you’ve lived without, say, full-color page scanners at home (down to around $2,000), high-resolution 19" monitors (only about $2,000-$3,000) with 24-bit true-color graphics cards (another $2,000 or so), a write-once optical disc to back up your work ($3,500-$5,000, but Jerry Pournelle assures us that every serious writer has such a thing as a matter of course) and all the other techno-toys, well, then, you’re just beyond hope.

Some of you probably still use something less than 486 CPUs on DOS machines—or, horror of horrors, use DOS itself rather than Windows 3.1 or OS/2 2.0. Some of you Mac users don’t have Quadra systems (and why not?). My guess is that at least half of you don’t have high-speed laser printers at home, that perhaps more than half don’t have true-color printers; that one or two of you don’t have V.32bis modems; and that oh, ten or fifteen of you haven’t found it necessary to build a local area network for your home computers. Is it possible that one or two of you still live without color monitors at home, or even use something as crude as an AT-class machine, just because you don’t seem to need anything more for home computing? I wouldn’t be surprised if—perish the thought—there are a few of you who get by quite nicely at home with no computer at all.

Yes, I do have a ‘386 at home, albeit only 20MHz, and I do have a laser printer, and a mouse, and a VGA color monitor, and a handheld scanner. No CD-ROM drive, right now, but that might change—and the system might get upgraded, for that matter. But that system is primarily a desktop publishing system; with it, I produced seven books this year as well as five newsletters and a variety of other materials, all on evenings and weekends. The system as a whole is roughly three years old; it’s paid for itself several times over. If it gets upgraded, it will be to serve more advanced desktop publishing purposes—not because I’m particularly in love with the technology. If I wasn’t doing intensive desktop publishing, I’d probably still have a five-year-old AT clone (which is still used every day in my wife’s library) or possibly even an eight-year-old CP/M computer (also still being used).

Of course I suffer from technolust once in a while. I read PC magazines. They try their best to keep readers in a buying frenzy. But given the realities of money, time and other interests, I find it easy to keep under control. So should you.

Electronic Publishing and the Death of Print

One common technolust fantasy is that print is essentially obsolete, and that electronic publishing will sweep it away in a matter of years or decades. For some such advocates, the argument is that electronic publishing is so superior—in terms of economics, access or flexibility—that nobody will want old-fashioned print. That’s a marketplace argument, which I’ll turn to in a moment; while I think it’s a stupid prediction, it’s not inherently dangerous.

The dangerous argument comes from those who recognize that establishing universal full-fledged electronic access would be horrendously expensive, and who therefore call for the elimination of print media (by government action if necessary) so as to clear the way for the electronic future. This is both a totalitarian argument and a profoundly dystopian one; fortunately, there are few such ferocious advocates. More victims of technolust say that librarians and others must wean people away from print. Even though poor deluded readers seem to prefer print, we know better. Which is another way of saying that we know that certain specialized needs can be filled better if we can just siphon off the funds that the majority use to fill their broad requirements.

Either argument basically admits that electronic publishing won’t defeat print, as a whole, in a free marketplace—so they argue for interfering with the marketplace. But you can’t have it both ways. If you’re going to interfere with the marketplace, you’d better be able to demonstrate overwhelming advantages and the lack of any serious drawbacks. "Inevitability" doesn’t cut it; if something’s inevitable, you don’t need to force it.

The fundamental problem here is that some people can’t deal with complex futures. They don’t want to see print and electronic media—particularly if that means that lots of works simply won’t be available for free or subsidized electronic access. But a complex future is the only future that makes much sense, at least if we want to have serious authors and thoughtful writing. New media should and will displace some forms of print, but by no means all or even most.

The Promise of Electronic Publishing

When you’re reading missives by those who predict the death of the book, it’s always useful to know whether they like books in the first place—and, for that matter, whether they’re really readers. Surely some of you have read predictions of the death of libraries and realized halfway through the prediction that the writer doesn’t use libraries, doesn’t understand them, or lumps all libraries into the same category. So too with those who predict the death of print.

Books have never been the best medium for many kinds of data and information. Electronic media make far better sense in some of these areas. The legitimate promise of electronic publishing is to provide better (that is, cheaper, more ecologically sound, more up-to-date, and easier) access to certain kinds of text—replacing books and journals where books and journals have never worked very well.

Librarians would be foolish to ignore this promise of electronic publishing, and I think the record shows that librarians have not been foolish. Libraries were the first major market for CD-ROM. Libraries have been doing online searching—another kind of electronic publishing—for decades, and show no signs of stopping. Libraries and librarians should work together to see that this promise of electronic publishing continues to improve, that real economies can be recognized in library budgets, and that varied forms of electronic publishing are used intelligently.

But this isn’t the future of electronic publishing as some see it. No, they see an electronic cyberspace, replacing bad old print with wonderful new electrons.

Limitations of Electronic Publishing

Not everything is in electronic form. Many things never will be, particularly while they are copyrighted. Just because something isn’t available electronically doesn’t make it unimportant, except to the truly devoted electronic publishing enthusiasts.

There’s more to a book than the text that’s in it—and that’s not a trivial point. The feel of the paper, the heft of the binding, and particularly the incredible convenience of the package: these are all part of what makes books so special. Even mass-market paperbacks, where the binding is truly trivial and the paper is basically high-end newsprint, have special qualities of the book about them, qualities that no electronic medium seems likely to replace. The handout goes into some detail about some of those qualities.

Varieties of Electronic Publishing

    Let’s look at a few curious aspects of electronic publishing.

The "Library-of-the-Month Club"

First, there’s the Library-of-the-Month Club, Jerry Pournelle’s idea for what to do after CD-ROM players are in every household—and how to get them into every household. He was pushing this off and on a few years ago; he’s quieted down somewhat now, off to the next hot new thing.

Here’s the idea. Everyone (well, almost everyone) has CD-ROM players. Now you start up CD-ROM subscription services, delivering the "library of the month"—a new CD-ROM each month, with 500 to 1,000 books on it (that’s roughly the quantity of a CD-ROM, if we assume pure text, with no graphics, sound, etc.) You charge, say, $20/month (he was a little fuzzy about pricing), you sell millions of them, everybody’s happy. Now, can you see a few flaws in this scenario?

Well, for one, I just might not want the same set of books as you would. That’s why book clubs have alternate selections—and why they don’t really expect everyone, or even most members, to buy a book every month.

But let’s make this as plausible as possible. After all, you really only need to press a few thousand copies of a CD-ROM to make the pressing inexpensive. So let’s say we have a hundred different library-of-the-month services: one for general-interest nonfiction with a few good novels thrown in, three or four for different aspects of science fiction, a couple for mysteries, and so on. After all, there are dozens of book clubs, some of them quite successful.

Do you want to get 500 books a month in one particular area, even one you’re very interested in? What would you do with them? Avid readers may read four or five books a week. Most of us are lucky to read one or two, under the best of conditions. And I suspect most avid readers have widely-varied interests, so they’d need three or four of the CD-ROM services. Which means they’d be getting thousands of books every month—stacking up endlessly, never to be read or paid attention to.

Now, let’s look at the price. I chose $20 because it’s about as much as most people would go for. You surely won’t get mass-market acceptance for a subscription that costs $50 or $100 per month. So: the CD-ROM sells for $20. Figure $3.50-$4 for manufacture and distribution—about a buck for the CD-ROM, probably another fifty cents to a buck for the lengthy leaflet containing the contents and descriptive paragraphs, another buck for postage and container, and another buck for very efficient handling (addressing, packaging, etc.) That leaves $16. Let’s assume an incredibly benevolent or non-profit company, one that only gets $2 profit out of this deal. We’re down to $14. Now, even if we assume that there are no editorial or production costs involved in this CD-ROM, that leaves $14 for royalties, for 500 authors. That’s just a bit less than three cents an author.

Would you like to tell an author that her novel is worth three cents a copy? "Well, you’ll make it up on volume." In other words, this book that only 5,000 people would have purchased as a $20 printed volume (returning $10,000 in royalties) will suddenly appeal to, say, 50,000 people because they’re getting so much for their money. Which means the author gets 50,000 times three cents: that’s the munificent sum of $1,500. What a deal!

Of course, these numbers are ludicrous. Realistically, at least half of the subscription cost would go to production costs, distribution, and profit; there’s no way an author could get more than two cents.

I write books now and then. There’s no way I would settle for less than a buck a copy, and I’d be hard-pressed to settle for that. So we’re up to $500 for royalties, which means at least $510 for each copy of the disc. Which, of course, we’re going to convince 50,000 people to pay on a subscription basis—why, that’s only a buck a book!

Right. For an assortment of books 95% of which you’ll never read, in a form that’s hard to read in any case, at an annual cost of more than $6,000. Sure sounds like a popular deal to me! Maybe there are millions of folks out there who want to do textual analysis rather than just read the books, but I somehow doubt it.

Magazines on CD-ROM

One extreme technophiliac in the library field, as part of a treatise on why virtual libraries are absolutely necessary, said that CD-ROM would displace print magazines any day now—since you could buy a Sony Data DiscMan for under $600, and have CD-ROMs delivered monthly that would include "more magazines than you could possibly want." And with the Data Discman, attached to a home computer, you could print it, cite from it, cut and past materials, etc.—"the utility has improved tremendously and the cost is minimal at best."

Now, if most magazine subscribers really wanted to be able to cut and past, print out selectively, etc.—and if the wretched little screen of the Data Discman was really a substitute for the high-quality text and full-color graphics that typify magazines—and if everyone wanted the same set of magazines, so that you weren’t pressing a separate CD-ROM for each one—and if advertisers would go for the new medium...well, then, yes, maybe. Sigh.

Mass-Market CD-ROM

The library-of-the-month club is just one pipe dream among the many that make up the mass-market CD-ROM. And boy, have there been projections of mass markets for CD-ROM. The question still is, just what do people want to buy in 600 megabyte quantities—and want badly enough that they’ll set up a computer system to run it?

I keep reading about the wonderful new CD-ROMs and wonder whether I’d pay to have them at home—after all, I already have the computer, and a CD-ROM drive is now as cheap as $199. Would I pay $900 for Compton’s Encyclopedia? Good lord, no. What about $90 for Magazine Rack, with year-old text from scores of popular magazines? Probably not; of the magazines I actually want—and I read quite a few magazines—only about $90 worth are in there—and the way I get them now, they’re not a year late and I get full color and the ability to read them anywhere, at any time, lying down, in the bathroom, whatever. Audubon’s Guide to Birds? The complete Sherlock Holmes? Why? The CIA World Factbook? Well, that’s what libraries are for...

Now, here’s the thing: I really would like to find a solid reason to buy a CD-ROM drive. Most people couldn’t care less. And why should they? What we have, therefore, is all these special deals: you buy the drive, and your first five discs are essentially free. Will you ever buy another CD-ROM? After you’ve oohed and ahed over the VGA graphics, will you use the ones you already have? Have you ever compared VGA graphics to the graphics in mass-market magazines, to say nothing of art books?

The newest hype is multimedia. Of course, you need to upgrade the old PC or Macintosh—adding suitable sound equipment, having a fast enough CD-ROM player to make it work, and so on—but now you can get moving pictures and stereo sound! You can also spend $80 for a Beethoven symphony with lots of things to look at while you’re listening to it. If you’re studying music, this may be a great deal. For the average classical music lover—well, after all, $80 will only buy five to seven full-priced CDs or about a dozen reissues, but they don’t give you stuff to look at. You’re left with just the music; what fun is that? Oh, I could buy a good book about the composer and his music—or borrow one from a library—but that isn’t the same as multimedia.

No, it isn’t. Oh, let’s not forget video. Check out the quality of the video in multimedia systems sometime—in comparison, say, to ordinary broadcast television. An unfair comparison? Really? Well, then, why am I spending so much more for the CD-ROM than I would for a cheap videocassette with near-CD-quality sound (which almost all contemporary videocassettes have)?

Don’t get me wrong. I love CD-ROM, for any number of niche markets. Even multimedia has its place, although I’ll wager it’s a lot smaller place than the enthusiasts predict. But as a mass-market success? Well, anything could happen.

Electronic Broadsides

One printed form that had become almost obsolete, for a variety of reasons, takes on new life within the electronic elite: the broadside.

If you follow a reasonably vigorous and sometimes literate listserv such as PACS-L (with its astonishing 4,000 members in 30-odd countries), you’ll see lots of little questions and answers, some personal attacks, and a lot of what’s been called water-cooler talk. You will also, now and then, see a coherent, thoughtful, assertive statement, either of principle or on a particular issue: an electronic broadside.

They’re easy to spot, particularly in retrospect. They tend to begin threads of discussion, or move existing discussions in different directions. They tend to be quoted heavily—both in direct responses and on slightly divergent topics. They tend to be misquoted as well, and misinterpreted; that’s always been true of broadsides. They tend to be more carefully written than typical submissions—one usually has the sense that they were composed off the net, not off the cuff.

Electronic broadsides have one or two defects as compared to their traditional, largely obsolete, print equivalents, but they also have several advantages. The defects are that you don’t get the wonderful typographic elegance that typified the best of yesteryear’s broadsides (you also don’t get the typographic horror-shows that typify today’s more deranged leaflets), and that you can’t reach the population as a whole with electronic broadsides.

The advantages are equally clear and, for broadsides affecting primarily the network-using elite, quite convincing. First, electronic broadsides are legal and require effort only to create, not to distribute; with print broadsides, you’re usually forced to hand out leaflets these days, since posting on walls or telephone polls is likely to be restricted or forbidden. Second, the creator of an electronic broadside can see its effect quite rapidly, usually within a day or two. Rarely, in non-revolutionary times, was this true for traditional broadsides. Third, responses can not only be posted as readily as the original, they can be tracked and linked into threads of discussion; other than community bulletin boards, this was never possible with print.

Should involvement in electronic discussion groups count toward tenure? By and large, absolutely not—no more so than taking part in staff-room debates. Should refereed articles in electronic journals count toward tenure? Absolutely, if the articles are worthwhile and the refereeing process is legitimate. Electronic broadsides fall into a gray area—although it can certainly be true that an electronic broadside can be widely cited and can have some small effect on the field.

Project Gutenberg

Those of you who deal with Internet/BITNET can hardly have escaped mention of Project Gutenberg. I don’t want to spend much time on this project, and I don’t really want to attack it, particularly after seeing the response of its director to any hint of criticism. I do want to say something about the English language, however, and particularly the phrase "given away." This summer, a PG missive proudly announced that they had already "given away" 2.6 billion electronic texts, a step on their path to one trillion such gifts in the next nine years. Wow...2.6 billion! That’s pretty impressive. To you or to me, that would presumably mean that there have been 2.6 billion occasions on which someone has actually made use of, or at least taken possession of, PG e-texts. Right? Don’t you usually assume that "given away" implies "to a willing recipient"—as opposed to "thrown away" or "littered" or "strewn across the landscape"?

If you believe that PG has placed 2.6 billion e-texts into the virtual hands of willing recipients, or even any visible fraction of that number, I hear there’s some ocean-front property outside Phoenix that’s selling quite cheaply. What this claim actually means is quite simple: PG had posted 26 e-texts at that point. PG projects that, by the year 2001, some hundred million people, or devices, or some such thing will have access to the Internet/Bitnet and whatever grows out of it. Thus, presto chango, multiply 26 times a hundred million, and you get 2.6 billion.

What? You mean that 100 million people aren’t currently linked to Bitnet/Internet—perhaps a tenth that many, at best? And it might just be true that the overwhelming majority of those users haven’t the slightest interest in downloading e-text versions of widely-available books, books they can buy for $4 or $5 in easy-to-read paperback editions? Well, that’s beside the point; in the true virtual world of e-text distribution, it’s still 2.6 billion strong.

Looking for New Forms

I’ve read the suggestion that Baby Bells and others will be offering virtual libraries to the average household at cable rates, perhaps $20 a month for basic services. Delivered right to your home, without the annoyance of librarians, etc. Who will actually pay the price of four paperback books a month for a relatively limited online information service?

Electronic bulletin boards represent a surprisingly widespread, surprisingly populist and charmingly anarchistic form of electronic communication. The Internet and BITNET host thousands of electronic discussion groups. Yes, there will be other new forms. Those that find suitable niches should do fairly well. Those that don’t, won’t. That’s true of all publishing, to be sure.

Right now, electronic journals appear to be the most promising form of true electronic distribution, with library-oriented CD-ROMs as the most successful form of digital publishing—except, of course, for the only true mass-market digital success, Compact Discs. But the most successful e-journals are those that have no subscription price and are produced entirely through volunteer labor. They’re interesting, but they’re not models for electronic distribution as an economically sensible path. The question remains: when will an electronic journal actually become economically feasible—that is, make economic sense as an alternative to print publication, without large hidden subsidies? So far, there’s no answer. I very much hope that there is a positive answer, and soon—academic libraries desperately need that development.

Economics of Print and Electronic Publishing

The handout includes an extensive section on specific economic tradeoffs of print vs. electronic publishing. Please do read it; it’s some of the most important material I have to convey, even if it’s a bit detailed.

    To summarize that material:

Only a fraction of the cost of publishing represents printing and distribution; electronic publishing just doesn’t displace the bulk of real-world publishing costs or prices except in some special cases.

For narrow-circulation scientific, technical and medical journals, there are real potential economies, and we can only hope that electronic publishing will have an impact here. For short-lived, bulky reference works, particularly abstracting and indexing, electronic publishing (whether online or CD-ROM) is clearly the way to go, and that’s already happening.

For garden-variety books, one rough figure I’ve heard is seven to one: that is, a hardbound book sells for about seven times as much as the costs of printing and distribution. That’s more than can be saved through electronic publishing, barring other revolutions in the publishing industry.

You can produce a high-quality, permanent-paper, 208-page paperback in a moderate run (2,000-5,000) copies for $1.97 a copy, and mail it for $0.33 to $0.70 cents a copy; that doesn’t leave a lot of room to cut costs, particularly since you’d spend $5.55 or more to print the same thing out on a laser printer, using three times the material in the process. For mass-market paperbacks, the numbers are wildly different: they cost so little to produce that publishers don’t even want them back—dealers just send back the covers.

I also discuss the economics of mass-market magazines and daily newspapers, and the important social functions of the daily newspaper that would disappear with personalized electronic newspapers. Please do read this material.

And, Not Or

When the only tool you own is a hammer, everything looks like some variation on a nail. When the only tool for distributing text was print, everything looked like a possible book, pamphlet, newspaper, magazine or poster. Now we’re adding electronic access and new media: call them drills and screwdrivers. Well, when you add a drill and a screwdriver to your toolkit, you’ll use the hammer less often—lots of things can be dealt with better using bolts or screws. Lots of information sources make more sense on CD-ROM or through online access than as books. We’ll find in the future that many specialized publications make more sense using electronic distribution than as print journals.

But you don’t throw away the hammer just because you’ve added the drill and screwdriver. Many things can be dealt with most efficiently either with the hammer alone or with a combination of tools. So it is with publishing and distribution: many things still make the most sense in traditional print form, and there’s no reason to believe that will change.

And, not or: that’s the only reasonable way to deal with the dichotomy between electronic media and print. Today’s libraries and libraries of the future need to deal with both. Libraries, now and in the future, also need both strong local collections of printed and nonprint materials and better ways to gain access to a wider range of materials than they can hold: and, not or. Libraries are doing this. Few of you lack CD-ROM and some level of online access, and surely none of you are getting ready to disassemble your book collections. Keep it up.

Data, Information, Hypertext and Virtual Libraries

The title of this talk mentions Xanadu, by which I mean the project, not the poem. That’s probably too bad. I wrote the title some months ago—but when I was writing the talk, I found that I really don’t have much to say about Project Xanadu. Basically, I think it’s the extended foolishness of one particular prophet, and that it is so far from any reality that there’s little point in dwelling on it. I’ve included a few brief notes in the handout—but generally, I think you can treat Project Xanadu as a dystopian pipe-dream.

    The handout also deals with full-text retrieval, albeit briefly.

Data is Not Information; Information is Not Knowledge

Proselytizers of electronic everything keep telling us that the world’s knowledge, or the world’s information, or something like that, is doubling every five years; only electronic methods can possibly keep up.

It may well be true that the volume of data in the world is doubling every five years. I’ve seen that interpreted as "the amount of knowledge in the world is doubling every five years." That’s nonsense—and I suspect it’s also nonsense to say that the amount of useful information is doubling every five years. A cynic might even suggest that the amount of knowledge is declining, as so many people seem caught up in haring after more and more data, failing to turn it into either useful information or worthwhile knowledge.

Information overload—or its evil cousin, data overload—is nothing new. It has been pointed out that a large insurance company churns out enough paper every six weeks to fill the largest national library. All of that paper contains data; if you want to stretch a point, you could say that some of it contains information. Virtually none of it is relevant to anyone but the company and a particular client; virtually none of it belongs on a library’s shelves or in its electronic connections; and virtually none of it expands the knowledge or enriches the culture of mankind. That doesn’t make it bad; it just makes it different.

Hypertext and Linear Text

Here’s what one technophile, who also happens to write fairly well, had to say about hypertext as a substitute for books. "I will not be browsed through. The essence of writing books is the author’s right to tell the story in his own words and in the order he chooses. Hypertext...completely eliminates what I perceive as my value added, turns this exercise into something like the Yellow Pages, and totally eliminates the prospect that it will help fund my retirement." (Robert X. Cringely, Accidental Empires)

I agree. For well-written linear text, the order is significant and hypertext represents an unwarranted interference. But that’s not the only problem. Hypertext isn’t free-text searching; it requires that links be established. That’s not an easy task—and it limits the reader to the links that someone has prepared. If that’s the author, then the author’s job has become much larger. Not only must text be cut down to bite-size chunks, but the author must prepare multiple sequences, presumably with some sense that they will all be readable. If it’s not the author, then someone else is assuming considerable authorial presence. That’s a new form of editorial interference, one that raises significant questions.

I’m not saying that hypertext is worthless. It’s extremely useful for some kinds of information—but it’s not a universal solution, and it’s a particularly inappropriate treatment for many books.

Big Government and Virtual Libraries

My heading here reads "big government and virtual libraries." That could be taken as yet another chorus of "where’s the money"—since only big government seems likely to have the resources needed to build truly large-scale virtual libraries. But that’s only part of it.

Tom Vogl noted in a February 20, 1992 PACS-L message: "As virtual libraries are created, there will be increased pressure on hard-copy libraries to relinquish their precious space and with it their physical collections, in whole or in part. Inevitably, both the soft and hard copies will be extant in fewer and fewer places, in a few well endowed libraries in hard copy and on a relatively few hard disks, all clearly identified as to location. Such a potential for thought control and effective book burning has not existed since Gutenberg, or possibly since the burning of the library at Alexandria."

The more probable repository would be government-run. Fortunately, as we all know, no government entity would ever think of rewriting history, concealing information or in some other manner interfering with the cultural record. Don’t worry, be online: your government knows what’s best for you.

Just In Time and Just In Case

Some futurists say that every library should move to a "just in time" model as opposed to the "just in case" model that leads to large collections. You get the information someone needs, but only when they need it—by borrowing it from somewhere else, printing out something that’s delivered electronically, or sending the user to a terminal.

Quite apart from the horrendous idea that libraries are just like carmakers, a number of carmakers and other industrialists have been strung up by their own principles this year: just in time only works when your suppliers are dead-on reliable. If there’s a breakdown in distribution, or if they want to hold you up for a better price, you’re dead in the water. We saw it this year with the brief rail strike. We’re seeing it when many just-in-time companies rely on single suppliers: the supplier plays favorites and companies go under.

It’s worse for libraries, of course. Who provides the supplies when everyone is busy being hyperefficient? If we don’t trust big government—and we shouldn’t—do we then trust big publishing to see to it that we get a fair shake on those one-off books we need?

The "just-in-time" model must deal with this commentary by Charles Bailey (February 18, 1992 PACS-L message): "The annual cost of licenses for local electronic databases and access fees for remote commercial resources will grow as the library and its users become more dependent on these electronic resources. In the long-term, if libraries opt for access instead of ownership (e.g., they don’t buy selected journals, but pay per article as it’s needed), libraries will collectively hold fewer new materials and be more dependent on commercial suppliers." Who will, presumably, all be benevolent, just as they’ve worked so diligently to keep print prices down.

Libraries: Not Just Reference Centers

Most library users expect libraries to have a variety of resources, and to serve a variety of needs. I guarantee that formal reference, as such, doesn’t make up most of the use of my branch library—and that branch, part of this year’s Library of the Year, is a vital part of the neighborhood.

    Information is not all that librarians and patrons care about.

Patrons value immediate over remote access: that’s been demonstrated time and time again. Unreasonable? Not at all. For most patrons on most occasions, currency and definitiveness pale beside immediate availability. That tends to be true even for scholars—although it shouldn’t be—but keep in mind that scholars are a small fraction of all library users.

More to the point, patrons want resources (information, entertainment, enlightenment) delivered in a manner that suits their needs and wishes. For many people, far into the future, that means a book sitting on the shelf: even if it’s not the most current book, even if the material in it is also available online. The user may not have ready access to a terminal (or be unwilling to pay communications charges—and those aren’t going away next week!); the user may simply prefer print on paper; and the user almost certainly knows what he or she can understand and use most readily. As Sandra Ballasch said in a March 17 PACS-L message, "It is well to remember that sometimes the user of an item (physical or not) really does know what suits him or her best. It may be wise not to assume that we experts always know just how people will respond to the changes we propose for their benefit.... The object is always to get to the user what they want in a form they can afford and can use. If we go so far as to provide information (for want of a better word) in forms and at places no one wants, what have we accomplished—for them, for us and for the larger community?"

The Future of the Library: Paradigms and Projections

We hear of paradigm shifts, the non-library, the library without walls, the library as electronic switching center, and so on. Talk is cheap. Most people who make absolute projections do so in the comfortable knowledge that nobody will blame them, or even remember, when those projections don’t come true. Clever people make projections decades in the future; with luck, they’ll be dead before the projections fail.

In my preferred future, books as we know them, and libraries as we know them, will be fundamentally important as long as I’m alive and quite a while beyond that. A century from now? That’s harder to say—for one, it assumes that there will be civilizations a century from now. If there are civilizations, I suspect there will be printed books.

Some new thinkers are convinced that library clientele only want that hot electronic information; either libraries change to meet their needs, or libraries become obsolete. This presupposes rather a lot about the clientele—and is a profoundly elitist and negative view.

Jean-Luc Picard

There’s a quiet little assertion about the future, several centuries from now, made in several episodes of Star Trek: The Next Generation. A visitor comes to see Captain Picard; he sets down the book he’s reading to greet them. The show doesn’t make a big deal of this; Picard isn’t a medievalist, and the book isn’t some sacred object. It’s a book, a printed, bound book, and he’s not the only one on the show that reads them from time to time.

The producers of this show are pretty careful about what they’re doing, and I happen to agree with them. Picard is a thoughtful captain, who appears to be interested in expanding his understanding in various ways. He can get information any time he wants, simply by asking the voice-activated supercomputer. But he recognizes the virtues of books as well—certainly not to plot the course to a new planet, but for contemplation and enjoyment.

Working for What You Believe In

The future is what we make it, or what we allow it to be. Perhaps not totally, but to a surprisingly great extent, we determine our own destinies. That’s important, particularly when you’re told that such-and-such (the death of print, the elimination of libraries, etc.) is "inevitable" and you must "adjust to the future." Such claims are designed to be self-fulfilling prophecies, but they will only be fulfilled if people accept them as true.

If libraries stop buying books, then libraries will cease to be places of books; that’s pretty clear. If libraries determine that their future is as virtual phenomena, then those libraries will cease to be useful as physical libraries—because they will have sealed their own fates.

Most absolutist predictions are made for economic reasons: they are made because the predictors want the predictions to come about. Any massive shift to electronic publishing involves incredible shifts in resources; the only way to get those resources is to convince people that the shift is either beneficial or inevitable. Since the anti-print cadre seems unable to make a convincing case for their course being beneficial to everybody, or even to most people, inevitability is the only out: how can you argue with the inevitable?

Your personal death is probably inevitable. I’m certain that mine is. But the death of print? The death of traditional libraries? I don’t think so. More to the point, if you think those are bad outcomes, you should do your best to see that they don’t come about.

"The MTV Generation" and the Printed Word

Some observers say that libraries must abandon the printed word because today’s generation, the MTV Generation, has itself abandoned the printed word. They assert that print literacy is dying out, and that future generations won’t have the attention spans needed for books or magazines. Instead of books, presumably, libraries should be stocking up on videos and saving money for virtual reality systems.

Illiteracy and aliteracy—knowing how to read, but not reading—are both problems, and they are problems that libraries should be directly concerned with. There’s a difference between concern and passive acceptance, however. The portion of society that doesn’t read, that can’t or doesn’t deal with coherent linear text, is the portion of society that dooms itself to being the underclass, subject to easy manipulation and fraud thanks to that unwillingness and short attention span. Libraries and librarians should be part of the effort to overcome aliteracy and illiteracy. You can’t do that by devoting all your resources to electronic information, which serves primarily the elite, and you certainly can’t do it by substituting entertainment videos for books.

No, I’m not saying libraries should stay away from videocassettes, any more than they should avoid mystery novels or romance novels. Public libraries should be sources of entertainment as well as enlightenment—but in balance.

Again, I have a lot more to say here; you’ll find some comments about false pasts and doomed futures in the handout.

Libraries and Users Are Not Monolithic

All libraries are not the same; all users are not the same. Different kinds of libraries have different missions. Public libraries are not primarily scholarly research institutions—but research does take place there. University libraries do not primarily serve entertainment needs—but every good university library has a fiction collection.

Many libraries must take special care to preserve the record of the culture. No other institution does this as well. But that’s not the mission of every library, although it is one of the missions of the field.

The Greatest Dangers: Elitism and Irrelevance

Networked information is now predominantly a service to the elite: a large elite, to be sure, but an elite nonetheless. Telling the average public library patron that they can check various Internet nodes to find a good WAIS server to locate the information they really need is basically telling them to go to hell. Right now, there’s nothing in NREN development that makes me believe that elitism will cease to be a factor in online electronic access—or that the elite seriously want that elitism to end. Indeed, I’d be surprised if we don’t see most new Federal money go to provide even better service for those who are already well-served, perhaps with a few crumbs for broader access.

Access to the Internet is non-trivial even for those who have the money or connections; it demands a level of computer literacy (and interest) far beyond what the average person has or needs for her or his daily life. Do I see that changing? Not at any great rate.

Perhaps the most bizarre aspect of the new elitism, the push for virtual libraries, is that it’s apparently a case of telling the users what they need, even though it’s not what they want. As noted in the RLG report on Preferred Futures for Libraries, "when push comes to shove, faculty want materials on campus. They don’t want to be dependent on other distant libraries for needed materials. Many of them also, because of the structure of their disciplines, still depend on at-the-shelf browsing. Efforts by librarians to de-emphasize ownership are interpreted as a failure to understand both the political environment and legitimate differences in research methodologies among disciplines." Somehow, however, we apparently know better. Or do we? As Steven Kirby (University of Georgia) says in a March 3, 1992 PACS-L message: "Faculty and advanced graduate students at research universities are the most likely beneficiaries of the virtual library. This is the group that most often makes use of obscure, possibly expensive, and seldom used materials. And if the prime beneficiaries have reservations about the benefits of the virtual library, then who are we building it for?" Perhaps, as some have noted, because it may be possible to do so—whether it’s desirable or not.

The virtual library that doesn’t meet the needs and preferences of its clientele is an irrelevant library, one that will become truly virtual, as in unfunded and nonexistent. That’s not a future that does much for me.

The Safest Projection: Change Will Continue

Printed books, magazines and newspapers will be vitally important for a very long time to come; certainly for the rest of my life, and probably for some centuries beyond that. The mix of new media and existing media will change, and will keep changing—just as it has for the past, oh, few centuries.

Change: that’s a safe projection. By its very nature, the process of turning data into information, information into knowledge and knowledge into understanding and wisdom can never be a mature or static process. But let’s not go overboard about change. Here’s a good futurist statement: "Change is the hallmark and only constant of society, making the known and accepted obsolete." (ASIS 1992 Annual Meeting Program) Change, yes; making the known and accepted obsolete—that really depends. Fundamentally, the second half of that statement is nonsense.

"The one thing we can be sure of is that the future will be at least as subtle and complicated as the present." That’s not original, but it’s true.

The handout includes some comments on a paradigm I believe will serve librarians and libraries for the future: Libraries and librarians should serve their users and preserve the culture. That’s pretty simple, and not at all futuristic, but I believe it’s appropriate. That means that a typical physics branch library on a University campus must be shifting resources toward journals and toward electronic resources; it also means that a public branch library may not do any such thing.

A Dream in Closing

I’d like to close with two little scenarios about the future. The first, which some of you may have seen already, is a bit downbeat. The second—well, I’ll let you decide.

The first appeared in the August 1991 Database Magazine as a guest editorial. It’s called "A Hypertext Story" and I’d like to quote most of it. I’m quoting myself in this case. (Read story portion of Hypertext Story).

Now, here’s a second scenario that distills my misgivings about many of the futures that people project.

In this future, electronic access to data and even information does indeed become omnipresent. The elite, the researchers and professors, have profiles and knowbots, personalized information services such that they’re instantly aware of anything that affects their work. And, of course, they can feed material back into the stream almost instantly, becoming part of that ever-growing world of linked information.

I see that future, with researchers spending more and more of their time reading more and more messages, spending what’s left of their time trying to refine their profile to cut down on the overload without cutting out the stuff they really need to have, spending their grants on knowbots and filters, perhaps even resorting to some form of virtual reality to try to cram more and more information intake into every waking hour...

Some won’t be satisfied, and will make a discovery. They will begin to get original work, important synthesis and true research done by a curious stratagem. They will disconnect. Pull the plug on the network; change their computer back into a personal computer, a tool to support thought. Go out into non-virtual reality, spend some time with nature, chat with real people, sit back and read a book, perhaps even jot notes down on paper.

It could be the dawn of a new humanism, with philosophers, writers and the like being the only creative forces left. It could be.

Text unchanged since 1992. Uploaded to Web May 23, 1999; layout modified July 18, 1999.