Prime Palaver #10

Eric Flint

May 12, 2002


     For this edition of "Prime Palaver," I'm posting a long letter I got recently. I think the points covered are well worth everyone's attention.

     Nothing in it surprises me at all. I can tell you that -- like every professional fiction writer I know -- I do all my serious editing of my own writing by first printing up a paper hardcopy and reading that. I write directly on a computer and I do a certain type of editing directly on the screen. Basically, the relatively mindless searching for typos and obvious errors, or sentence-at-a-time polishing. But when it comes time to read a big chunk of a new novel I'm working on so that I can really THINK about it, and consider how I might want to rewrite or reshape it in important ways, I always print a hardcopy. (And, as I said, I've asked other authors and they all do the same thing.)

     The reason is because, for whatever complex psychological reasons, there just seems to be something about holding a book in my hands that concentrates my brain. Whereas scanning through text on a screen starts essentially dissolving my concentration. After a while, I'm not reading so much as "scanning." Great for quickly picking up pieces of information or for spotting obvious problems, but almost useless for any "higher functions."

     One thing I've also found is that I invariably wind up reading something on a screen much faster than I do on paper. I'm not sure if that's eyestrain. That's the obvious explanation, but I think what's involved cuts much deeper. I will easily spend 8-10 hours a day staring into a screen, while I'm writing, and not feel any eyestrain at all.

     I'm not making any general pronouncements about this, because I'm sure there are people who have a different experience. But I strongly suspect that my own preference is the most common one, and I am highly skeptical that the two standard explanations are really adequate.

     Those two explanations are the following:

     First, that as technology improves and electronic books become as readable as paper books, this "hardcopy preference" factor will fade away.

     Second, that there is a generational issue involved, and that as more and more old farts like me who grew up on paper books die off the newer generation "accustomed to reading on a screen" will set the preferences.

     I think both explanations, while they contain a kernel of truth, are grossly overstated. For the following reasons:

     The "eyestrain" factor is simply exaggerated. I spend most of my working life, as do many people, staring at a screen. Very rarely do I experience any eyestrain or difficulty in reading electronic text. A bit now and then, certainly. So what?  I also occasionally strain my eyes reading paper books.  In fact, the single worst instance of eyestrain I've ever suffered in my life, which required me to stop reading entirely for two weeks, happened when I was a young man long before the advent of computers. That happened because I was cramming too hard for final exams in college -- reading paper texts.

     What this argument ignores is that the real difference between a paper book and an e-book is a qualitative difference in the packaging. They are, quite literally, bound in two different ways. An e-book, no matter how sophisticated the technology, only allows you to see one page at a time. Whereas a paper book, in an odd sort of way, actually allows you to see all the pages simultaneously. Granted, you only read one page at a time. But, because of the binding, you can literally grasp the whole thing in your hands and -- this is my experience, at least -- makes it easier for your mind to "wrap around" the entire text. Not to mention the fact that you can flip pages a lot faster than you can scroll through a text on a screen.

     The end result, I think, is that the two formats serve two different purposes. And I strongly suspect, although I certainly can't prove it, that eventually psychological researchers will discover that the human brain is essentially "hard-wired" in such a way that it reads the two different formats differently.

     If I'm right, in short, paper books are going to be around for a LONG time to come -- because the key is not the "paper" as such but the very different ways that the sheets are bound and packaged, compared to electronic texts. I really think what's going to emerge is a "dual" reading culture, in which people switch back and forth between the two formats for specific purposes. Each has its advantages and disadvantages.

     As far as the generational issue is concerned, in itself, I mainly think it's a lot of hogwash. I didn't start using computers until I was in my mid-forties, and I didn't have any trouble at all making the adaptations necessary. (Well... not much.) And I know lots of middle-aged people who adapted even quicker and more readily than I did. In the last factory I worked in prior to becoming a full-time writer, I knew at least eight machinists -- two of them older than me, and only one of them a "youngster" -- who'd become enthusiastic computer-users. Three of them liked to build their own, and one of them -- a man in his late fifties -- had developed his own graphics program. All of these men were blue-collar workers, mind you, not scholars or academics or intellectuals of any kind.

     And, on the flip side of the coin, as the article quoted in part below shows, researchers have found that modern college students -- i.e., the very same "computer generation" which is supposed to be accustomed to reading on a screen -- have far worse retention when they read that way instead of from paper editions.


     Bah, humbug. I think paper books are going to be around for a lot longer than people assume.


     And that's enough from me. Here's the letter:




     Your introduction to the library, and your Palaver #6, are the best pieces I've read by an author/publisher on the advantages of free distribution of literature from which authors would like to make money. (Scholarly journal articles are nearly unique in that their authors aren't paid anything for them and consent to have it this way.) You mention the experience at MIT Press. If the example of publishers of scholarly books can help you make your own case, then an even better example is the National Academies Press, . It publishes all of its 2,500 books both in free online editions and in priced printed editions and, like you, has found that the free editions stimulate sales of the priced editions.

     Here's a brief statement of NAP's rationale:

     Here's a longer statement of NAP's rationale, from the September 14, 2001, Chronicle of Higher Education. Sorry if this is more than you want to hear, but I'm cutting and pasting because the article is not available for free online:


Academic Press Gives Away Its Secret of Success


     It's been a bad year financially for nonprofit publishers, according to most reports. High returns from inventory by booksellers closing their doors or trimming their stock, combined with sagging sales of what are considered discretionary products in a slowing economy, have forced many nonprofit publishers to rethink their plans and budgets. Even some of the largest and most well-known university presses are whispering about deficits.

     So it's almost embarrassing when I tell colleagues that the National Academy Press is on track for a record year in book sales. And it dumbfounds them when I mention that we make every page we publish in print available online -- free.

     Ever since new technologies began to hint at the possibility of reading books digitally, publishers have been haunted by the prospect that e-books would make print versions obsolete. The publishers have been trying encryption schemes, lockout mechanisms, and restriction systems to prevent unauthorized access to online material, with limited commercial success. For nonprofit presses, which operate close to the margin, the electronic future has looked like a minefield.

     Our experience may calm a few jitters. And it may suggest some ways that nonprofit presses can expand their influence in the electronic age, with relatively small investments and limited risk.

     Our press is the publisher for the National Academy of Sciences, the National Academy of Engineering, the Institute of Medicine, and the National Research Council. We publish more than 200 book-length works per year, and are required by our charter to perform a dual task: to disseminate as widely as possible the works of the academies and to be self-sustaining through book sales and fees for services we perform for internal and external customers.

     Those two mandates may seem contradictory, but we have found that, at least for a publisher of scientific and technical analyses and policy reports, doing the first encourages the second: Making our material easily and freely available helps us sell books. Our Web site ( makes more than 2,100 books -- comprising 400,000 book pages -- fully searchable, browseable, and even printable by the page, all free. The material is made available in easily navigable page images, and we are in the process of providing even more easily readable and quickly downloadable page-by-page HTML text. Expanded research tools are in the process of being developed.

     Our site is very busy -- from January through mid-August of this year, more than 3.2 million people had viewed more than 28 million Web pages, including 15 million book pages. While those are great numbers in terms of wide dissemination, the more remarkable thing is that, over the same period, we have sold more than 40,000 books through the same site -- something approximating 25 percent of our overall book sales, and already surpassing the number we sold during all of last year. Moreover, our other sales -- via bookstores, an 800 number, fax, and mail -- have apparently not been cannibalized, staying pretty much in line with industry sales.

     It would seem axiomatic that giving away pages means that fewer people will buy the books, but that confuses the content with the product. Sugar, butter, flour, eggs, and vanilla are the contents of a pound cake, but quite obviously more than those contents is required to create something pleasing to the palate. It's clear to us that the material we publish -- the final printed book -- has a value quite distinct from the content itself, and a utility independent of any particular page. The handy, readable, formatted, bound volume is still the way most people want to read a book-length work.

     Comparing books to food is dicey, of course, but the appetites -- whether intellectual or gustatory -- have similarities. For some kinds of hunger, quickly digested information -- the fast food of the Internet -- serves a number of useful purposes. Doing research on facts, addresses, news, and the like has never been easier. However, in the olden days, before the Web, few of us actually purchased books to learn that kind of information anyway. We went to the library, we consulted an almanac or an encyclopedia, we asked friends, we called the operator, we subscribed to newspapers or magazines.

     We bought books we wanted to savor, not data to munch. We bought books we wanted to own, books we wanted to sink into. That's still the case.

     Book-length material tends to posit an attitude, a position, or a conclusion; it may hypothesize, assert, or persuade; it may entertain or enlighten; it may surprise or delight. It has, in short, its own context. Extract a page or a chapter, and it's no longer the same product. That's part of the reason that Web technologies, whether they offer page-by-page representations or chapter-by-chapter material in Web-ready form, can rarely compete effectively with book-length works in print.

     People are happy to find and browse through online material, but nobody -- and I mean nobody -- seems to be interested in devoting lengthy periods to reading for meaning online. Our server logs indicate that most people skim a book -- they choose a few pages, perform a few searches, print a few low-resolution pages. Apart from the act of printing, that is just library or bookstore-browsing behavior, not a threat to our livelihood.

     There is mounting evidence that people will read for facts online and, while they'll read small chunks of material -- articles -- for perspective, few will read anything that runs for more than 30 pages onscreen. And when they do, it's unsatisfactory. Researchers at Ohio State University reported on a study last year indicating that even for college students who are making an effort to absorb as much as possible, material read on a screen is harder to understand than the same material read on paper. Last year, Forrester Research released a report showing that dropout rates for online courses can be as high as 80 percent. Why? In part, the Internet-research company found, because retention is 30 percent lower for material read online than for material read in print. A few months later, Forrester forecast slow growth for both e-books and e-book readers. Why? Because the company found that not only do people generally dislike reading text-heavy documents on a computer screen, but they also retain less of what they read.

     The Web's promise is vast and still mostly unrealized, because the dot-com gold rush diverted energy from what the Web is best at: connecting people with ideas. From our perspective, the Web is already the best dissemination engine ever, which has the side benefit of providing vast new markets and audiences for our work. Scientists or program assistants or policy analysts in Goteborg or Kampala or Tulsa can find a policy recommendation or an expert conclusion in our publications -- from a book that they probably wouldn't have found before the advent of the Web. A student in Lubbock can explore Science and Stewardship in the Antarctic, and a teacher in Kiev can browse Preventing Reading Difficulties in Young Children. If any of them want to, they can purchase the book at hand. Enough do so to support our program.

     Does all this mean that every book publisher should put its books online at no charge? Alas, few for-profit book publishers are willing to invest money in giving content away. Their business models have profit maximization as the main goal, within which framework good people have to do good work. Opening content up, without locks or timers or payment, is just too outside the paradigm to be considered.

     Most nonprofit book publishers I talk with would like to be able to do something similar to what we are doing, and a few are doing so. The Brookings Institution Press is making more than 100 recent books available for browsing via its Web site (; to date, more than one million visitors have browsed those titles, and online sales of the books have more than doubled. The MIT Press, the University of Illinois Press, the Columbia University Press, and other innovative publishers have initiatives that include free access to some book-length material. To my knowledge, no book by any publisher has ever sold less than expected because it was available free online.

     Only a few nonprofit book publishers have actually undertaken the risk, however, because most have very limited financial flexibility. They aren't blessed, as we are, with a parent institution willing to support a grand experiment, and any loss in today's straitened circumstances would take a big bite out of limited resources.

     The "crisis of the monograph," much discussed over the past decade, is at heart a crisis of limited resources. When the editing, production, and marketing costs of a book exceed income from sales, a press loses money. But a large proportion of a publication's cost is its marketing and promotion; if it were easier for books to find their own audience by being more freely accessible, presses might be able to afford to publish the scholarly monographs that are beginning to be too costly to produce. Free online access to the books might help us out of the crisis of the monograph.

     It therefore would behoove universities and the other parent organizations that sponsor, support, or otherwise give room to nonprofit publishing houses to consider a small investment that could have a big payoff. With an injection of $100,000 or $200,000 for initial staff and digitization costs -- and, perhaps more significant, a clear statement of institutional support for experimentation in scholarly publishing -- a lot more university presses could make a lot more of their publications available online in ways that would enhance scholarship and knowledge worldwide. It could even enhance their financial status. Successful initiatives like the National Academy Press's seem to show that the risks are not as great as once was feared, and that nonprofit publishing may flourish best when it is most open.

[NOTE: Michael Jensen is director of publishing technologies at the National Academy Press.]

  Eric Flint