By 2010, more than 50 percent of books sold worldwide will be printed on demand at the point of sale in the form of library-quality paperbacks.

http://www.longbets.org/bet/6

There's an excellent new book called The Myth of the Paperless office, written by a couple of cognitive scientists, which delves into the reasons why paper will survive in the long term. They argue it has little to do with screen resolution, contrast, portability, battery life or any other problems that better technology can supposedly fix (although they were mostly studying office documents, not novels).
Technologists should read this book....it's about how people behave in the real world.
Both e-books and on-demand publishing are great ideas, but when has being a great idea ever been a guarantee of success?

Is there a third scenario here - that both sides lose the bet because both print on demand and e-books fail to challenge books as we know them?

I personally agree that this is a wonderful idea which would naturally manifest itself anyway in the electronic age. Doesn't the guarantee of success of on-demand publishing teeter on what Lenny Riggio is presented by his team? Will 100k a machine, two or three machines at each retail location, be too much of an investment to be able to successfully turn a profit? The reader will require at least a book cover to flip through. We need to see the hipe, liner notes, etc. How exactly will these books be presented for the perusal of the prospective buyer. Do you think that everyone will walk through the door, belly up to the counter and announce the title and author? The retail buyer is not reading the Times book review every week. They need to flip through sections, see what Oprah likes and kill some time away from their spouse. Also, as this 100k machine is spitting the book out, is that the time that you order the double skinny mocha? If Lenny thinks he might make the 2 bucks on the coffee, well you might just sell those machines.

That's a nice approach to take, analyzing the plausibility of the experience, the peripheral benefits of the old way, plus the sheer familiarity of the old way. Something new has to offer serious and multiple benefits to overcome the highly evolved familiar pattern.

Good point....exactly the sort of critical analysis of innovation that was sadly lacking during the technology bubble.

(Maybe one of the failures of forecasting in general is to focus too much on grand vision and not on little details like these.....I know it sounds like nit-picking but surely there is a balance?)

I agree with you in that people want a "physical" object to inspect before buying, and also the act of looking for a book is a ritual of society in and of itself. BUT, two problems. First. 100k, for a printer capable of printing a book (even in full color) is OUTRAGOUS. I worked in a graphics shop once, and our printers (while capable of highspeed printing, and color, and binding) cost about 35K. So the price is a little overblown. Not too mention the money saved on retail space, by not having rows and rows of books.

Someone else made the point that people are interested in seeing and reading the cover, the linenotes, and other such periferals. Those things could easily be provided, either in a digital format, or in an actual "store" copy of the book.

My opinion on this is that books of the paper variety will die out slowly, and will never completly go out of style. But, they will be replaced (at least majoritly) by some digital form of themselfs.

HTML was orginaly made as a way to "revolutionize" information delivery. It did this, but not at all the way the creator intended. The vision almost never lives up to reality...

Within the last several years a clause has appeared in most publishing contracts entitled, "Grant of Electronic Rights." Many authors have already granted these rights to the houses without discussion of reasonable exploitation and fair compensation. If the publisher is currently investing heavily in these new technologies, the author will benefit "in house." If the publisher has not, these potentially huge profits will be farmed out to another house or imprint who has been gearing up. This will be the publisher's future back list.
Maybe it won't be Lenny and his boys selling the on-demand books. Perhaps one or two houses will get on the stick and create a new type of publishing house, one that exclusively buys or leases electronic rights from other houses (I wouldn't be surprised if Time Warner was already analysing this possibilty)and who then opens a chain called "On Demand Books" where you can browse the web site for books, liner notes etc., buy that cup of coffee and wait until your wife goes to bed before going home.

Hey Jason, you know what I just thought of? Why can't the shopper also request a CD to be burned or a magazine printed on demand.
"On Demand Media and Music" That's what we'll call it! I can picture walking into a storefront, seeing computer stations where people are browsing books online, young kids with headsets in listening booths and the smell of fresh brewed coffee that doesn't put hair on your chest.

Now you've got me going. In this store the buyer would log onto the store's web sight which contains all of the media available for on-demand publishing. One would be able to view the cover, liner notes, and critiques and also be able to read a few pages. This electronic publishing house selling directly to the end user will be able to charge the lessee of the rights for the production of the web page that will eventually sell the book to the propective buyer in the store. Preference on the store's search engine can be bartered as well. Hey! Who's paying me for all this stuff?

re: Books being replaced by other media.
I don't see it happening by 2010. Even "computer professionals" today, people with no tech-phobia, tend to have large stacks of manuals and reference material piled around them when the same information is also available in an electronic form. Getting the majority of the general public to both accept and have access to some form of "personal reader" is still, I believe, a long way out.

re: Retail plausibility.
Amazon.com already offers "browsing" a book by cover (front and back) liner notes, excerpts, and reviews all on-line. The retail on-demand bookstore could pull in more people by (as Zac suggested) having perusal copies of popular titles in the store. One of the advantages to the retailer would be reducing the space needed to run the store, while increasing what they have to offer. They would never be "out" of a book available on-demand. Plus they could offer "large print" and a choice of multiple languages. With an on-line tie-in, the retailer could allow people to shop and purchase on-line, get an e-mail when their book is ready, then walk down to the corner store and pick up their freshly printed purchases and a cup of java.

re: Big Business plausibility.
I think a lot of the major players feel uneasy at the thought of distributing books/music/etc. in an entirely electronic (read "easy to copy") fashion. The on-demand bookstore would allow them to have control and accountability over what gets printed and how many times. It would also alleviate their cost and overhead to publish.

I don't think I'd bet against Jason on this one.

Would Barnes and Noble want on-demand publishing? They've got bookselling sewn up. A book is purchased, the purchase is then logged into their marketing software and another book is earmarked to be put on the truck for delivery to replace the one that was sold. He's making money. Lenny doesn't need on-demand publishing. Amazon, on the other hand, could print and mail the same day and not worry about overstocking. The publishing companies already contract with Amazon to buy back any unsold covers, so would this be adventageous to them?

50% printed on demand in shop, 50% e-book ...

I'd wonder whether by 2010 there will just be 50% less books sold then than now, I'd think such an outcome is about as likely as the two being bet upon. [Maybe it's just as well for my pocket that Long Bets doesn't seem to take three way bets]

The post made about "why not have CD's burnt on demand" has already been asked by Blockbuster. This is their idea:

The user swipes their credit card at a terminal. (If the user has been to Blockbuster before, the system recognises them and suggests purchases.) Using a touchscreen, the users find the CD they want and the process starts. While the disk is burning, a printer produces the cover and the attendant assembles it and hands it to you. The terminal also suggests "Do you want to listen to this similar group while you wait?" to prompt extra sales.

This was a few years ago, so I guess they are ironing the bugs out of it.

As an avid reader, I will never stop buying books in paper form as long as they are available. That said, print-on-demand has huge possibilities.

I would definitely pay more for the convenience of a new copy of an out-of-print or hard-to-get book that I can have printed right then and there, rather than special ordering from the store or a web site and having to wait for a possibly inferior used copy. This is a great way for publishers to sell formerly out-of-print materials, with little cost to themselves. They make the book available electronically (with a one-time cost for scanning or other form of upload) and continue to make money for as long as people buy the book. Authors continue to get royalties as long as people want to read the book.

In addition to convenience, it would force authors to write better books that people would continue to buy once a fad was over. This is similar to music downloads. If you can buy albums one song at a time, it forces musicians to write better songs, so you'll want the whole album.

As for the browsing problem, since stores don't need to devote as much room to shelf space, they could devote more room to information about books and "reader services". They could provide internet access to book reviews and reader web sites. They could host more discussion groups and have more author visits. They would have to hire more employees who are informed about books!

Finally, this could be wonderful for students, or anyone who wants to customize. You could buy a book printed only on one side of the paper, so you have plenty of room to make notes (an pay extra, of course, for the extra paper). As noted earlier, you could choose large print or foreign language. You could choose your paper quality and binding. Color illustrations or not. The possibilities are almost endless.

Every time I have thought about this technology since it was introduced, my eyes have glazed over thinking about the possibilites, and why it will take off once publishers and bookstores figure out how to make a profit on it.

I agree completely with Carl here, first there are too many technical hurdles against print-on-demand for it to gain such widespread use in such a short time. On the other hand there is no way electronic readers can ever replace paper books adequately. There is simply too much information and ease of use than can ever be replaced by a flat screen interface. You have earmarks and creases and the ability to flip back and forth in different page increments instantaneously. You can look at multiple pages simultaneously. You can mark 4 or 5 pages with your fingers and flip through them without any thought. The problem with e-readers is there is always going to have to be an interface, and unless it is directly wired into your brain, it's only going to give you a small fraction of the information "throughput" that a paper book will give.

I'll have to agree with Carl here; it is unlikely that technophobes will relinquish the paperback regardless of other adequate solutions - especially not enough to account for 50% of the paperback sales.

This is not to suggest that those sales could be done in just-in-time publishing devices. The oooo and aaaaa effect of having multiple copies of books could be handled just like Burger King handles its sales. The possibility of combining Max Headroom-ish avatars could actually expand the potential of peronalizing and expanding author and book related events.

Imagine, if you will, an AI writing under a psuedonym with a CGI avatar selling his books and interacting with customers on demand.

"I'll have to agree with Carl here; it is unlikely that technophobes will relinquish the paperback regardless of other adequate solutions"

'Adequate' is the key word there. As an extensive reader both of paper books and of various e-texts online I definately feel that a paper book is the better medium.

However I am also dubious about the likely 'build quality' of 'in the shop printing'.

I think a lot of people are confusing whether on-demand will be possible versus whether people will embrace it. 50% of books printed per year through on-demand is a bit much i think. I think the majority of people who are true booklovers, people who have libraries in their home stacked with books, are the type of people who enjoy going into bookstores and wandering around looking for something interesting. That will never change. However if you are looking for something specific and were sure you were going to buy it, then it would make no difference to you whether you got it on-demand or in a bookstore like Barnes & Noble. At that point price and ease of getting the book would rule your purchase. As for the comments about printing on-demand. I work for a design firm and I spoke with one of our printers. Depending on the type of book, I suppose you could create a machine for 100K that would make books on demand but the time to create a book would vary depending on the kind of book, photos or illustrations, color or black and white or duotone etc. The on-demand process works best for a plain all-text black and white books. I cant see creating an art book on it where color and tone would be an issue. Too many factors are involved there. In the end, it comes down to the readers and how they prefer to get their books.

I suspect that, worldwide, most books sold in a given year are not new books. If that is true and continues to be true then Epstein has no chance to win even if 100% of new books sales are print on demand.

Of course, collecting the data for all book sales, as compared to retail book sales, presents its own challenge. Is there an implicit understanding that the bet refers to new books sales only?

Here are what I consider the major strikes against print-on-demand:

1) It's labor-intensive compared to regular bookselling. Someone has to stand there and watch the printer going

2) It's expensive technology

3) It is no less likely to break down than a heavily-used computer printer is - probably more likely, since it does more things

4) storing the paper for the machine takes up a lot of space

5) There is a noticable waiting time for the book, compared to pulling it off the shelf

6) People, even avid readers, rarely go into a bookstore knowing exactly what they want to read - and much of the money made in bookstores is from someone who walked in looking for the latest bestseller and walk back out with more than that one book. If you are only selling books people walk in looking for, you're losing business, and if you need to hype the books on-site in something akin to how they are hyped now, ie, placed right near other books, you'll take up just as much room

7) How does the author reliably get paid? To print the book on-site, the bookstore must have it availible in printable form on-site one way or another. What's to keep bookstores from failing to report sales and keeping all the money?

9) Expense. Economies of scale do not apply as strongly when you are printing books one-by-one. An out-of-print book is fairly easy to find on bookfinder.com, and often for much less than its cover price. A print-on-demand book will be more expensive than a used copy or even a new one, so unless the current demand outstrips the new/used supply, there will be no way for print-on-demand to compete.

Bottom line is, from the customer's perspective, the book distributionm system isn't broken, so why spend money to fix it?

The technology required to create CDs on demand at a record store or have daily newspapers delivered via fax machine has been available since the 1980's, but neither of these services has been implemented on a wide scale.

Why? Largely due to the economics of scale. It costs the average consumer many times more to print dozens of pages from their fax machine than the cost of a newspaper. It would cost a record store more to maintain an industrial CD-burner than to stock mass-produced disks from the record companies. By the same token, a bookstore would pay more to maintain a digital press and bindery than it would cost to obtain the book from a publisher.

I just wish that I owned development rights for on-line e-book downloading,as well as North American rights to an e-book stores franchise system consisting of a central computer,20 or 30 peripheral screens,unlimited availability of downloading any book, and RENTAL e-books.
Remember rental video machines?
Needless to say this would just be a start.
To me e-printed books would be too expensive, bad for the ecology, and not particularly advantageous. I love to read books,not necessarily collect them. Why not make it easier, cheaper,more environmentally friendly? We could also eliminate dog-ears and physical book marks! I could also get rich at the same time! Sorry folks, e-books will dominate the traditional market before anything else.

(1) The framers of the bet certainly seem to distinguish a "book" (meaning a printed, paper-based medium) from an "ebook". The bet specifies "books". Ergo, the number of ebooks sold does not impinge upon the conditions of the bet.

If only eleven printed books are sold in 2010, but six of them are POD products, then the affirmative wins the bet. The negative (Cerf) seems to have presented arguments based on the idea that ebooks will displace books, but, per the apparent terms of the bet, that doesn't seem terribly relevant.

Naturally, that doesn't mean that POD will displace other forms of book sales, but it seems worth noting.


(2) Epstein writes, in his argument, "new POD technology - like a prototype developed by March Technologies - will have fully integrated components and a cost of about $100,000 per unit."

Three or four years ago, when I first read about POD paperback sales, the article I read (almost certainly at either CNN or the NY Times online) dicussed a model made by a Japanese company -- I think it was Matsushita -- was said to be able to produce an "average" mass-market paperback in under five minutes, including binding and covers, and that the unit cost was about $10,000. The article also stated that POD books should cost about the same as regular books.

I realize Mr. Epstein may be thinking of a higher-output unit than one that only produces about a dozen paperbacks an hour, but the cost per unit he quotes still seems too high.

I think the real market for POD will be where the customer orders remotely, by phone or online, and either has the book(s) shipped or else goes to a bookstore (which would have traditionally produced books, too, most likely) to pick up the specially ordered books.

I don't think most POD, in the near future, will typically be a "while you wait" system, but more like one-hour photo development.


(3) Ebooks will take off when ebook media (computers) become elegant (ie, highly portable, easy and pleasant to use) and inexpensive. I think it's likely that the first really widely successful ebook computer will look something like a real book, and, for reading purposes, will pretty much 'work' like a real book, too. Except that you probably won't physically turn the pages.

When the technology is made comfortable and nearly transparent to the user, you'll see mass-market appeal.

It sounds like the bet hinges on the success of "print on demand". If this technology fails to dominate the market, then the Internet Society gets the dough, even if "print on demand" is far more successful than electronic books!

Good bet, Vint! ;-)

"I love to read books,not necessarily collect them. Why not make it easier, cheaper,more environmentally friendly?"

Isn't that also known as a "library?"

I don't think that e-books will replace printed media anytime soon. There is a certain tangible quality to paper that makes printed material difficult to replace, if only because people are resistant to change. Virtually all offices today are equipped with email and intranets, but how many have you seen that were honest-to-goodness "paperless?"

>> "I love to read books,not necessarily collect them. Why
>> not make it easier, cheaper,more environmentally
>> friendly?"
>
> Isn't that also known as a "library?"

Library use does seem ("seem") to be way down. It certainly is among people I know, compared to as little as a decade ago. I'm not really sure why.

My personal experience suggests that in the last two decades libraries, on average, have become more impersonal. But that's a highly subjective impression.

In any case, I now know more people who regularly sell their old books and buy 'new' ones at used bookstores than who use libraries. For some people, at least, swapping books has replaced borrowing them. Being allowed to keep the ones you like best can definitely be a perk.

> I don't think that e-books will replace printed media
> anytime soon. There is a certain tangible quality to
> paper that makes printed material difficult to replace,
> if only because people are resistant to change.

These are good arguments, but they've also been used before. About the advent of the automobile, for instance, and the telephone, and electric lights.

> Virtually all offices today are equipped with email and
> intranets, but how many have you seen that were honest-
> to-goodness "paperless?"

None, myself, and I haven't even seen many that seem to have reduced their paper output much. But at the same time that computers have made paperless communication easier, they've also made paper communication easier, thanks to computer printers. If the alternative to email was a typing pool, then email would have come a lot farther in replacing paper memos.

Beyond that, electronic communications are often applied to applications where the medium is not yet as 'mature' as paper is. An email memo is swift, easy, and neat, but if you want everyone to bring a copy to the meeting, they either have to print it out or bring a portable computer of some kind.

Right now, paper often has the edge over mobile computing. In five or ten years, when "e-paper" will most likely have hit the market in a big way, that may no longer be the case.

It's conceivable that within ten years a mobile computer no larger than a paperback, with an equivalent ease-of-use (overall) and with better 'screen area' and 'resolution' will be cheap and widely available. Such a beast would undoubtedly be able to store much more information than a paperback, too, and possibly, in its most typical configuration, be capable of high-speed wireless data transfer.

If so, we may well see serious inroads made against paper communication. There's something nice and tangible about owning a CD, but look at all the folks who've flocked to MP3 players.

'Print on demand at store' and the present system.

There are (in my experience) basically two types of book purchase methods at a book shop.

1. Browse through the section in question, looking for titles + authors (even covers ;-) that catch your attention. While you'd think a catalogue (electronic or otherwise) would be as good that isn't my experience.

2. Search for a specific author / title / obscure subject. I have many second hand books where I have 1 or 2 from a trilogy. However the 'gaps' can't be easily filled because often the remaing books are out of print / only available in America / Japan.

In the second case a few machines at the book chain's HQ could provide a way to print one-off orders. As a start, this should be much more economic than setting up a printer per shop, while only delaying book acquisition by a few days.

Of course to work it would have to be routine to have e-text available to book chains after books are published.

The 'no book out of print' model of on-demand book printing is a likely one, especially for the intermediate future.

The simplest way to set it up is like so:

-- A small number of companies (or possibly just one) provide the service and secure the rights to out-of-print books (as well as copies of the text).

-- Consumers can place orders on-line, by phone, or at participating 'partner' bookstores.

I feel pretty certain that a chain like Barnes & Noble would jump at the chance to partner with this kind of print-on-demand company, at least until they could supplant it themselves.

-- The printed books are shipped directly to the consumer or to the store where the order was placed, at the consumer's discretion. Presumably, you could also have a gift book shipped to the recipient.


Books still in print would be published and distributed normally. The real issue that arises is that a great many books (more than most people would ever expect) are not profitable for publishing companies.

If print-on-demand became popular, it's likely that many publishers would cut way back on how many titles they printed. Especially since profit margins for publishers are vanishing these days.

This, in turn, would likely boost support for non-traditional distribution systems -- ebooks and print-on-demand. I have no doubt that both will become important in the next two decades, but I'm not sure what the balance will be.

Good e-paper is already a reality. It just hasn't made the jump to market applications yet (except for some very specialized ones). It could be truly huge in as little as five years.

"If so, we may well see serious inroads made against paper communication. There's something nice and tangible about owning a CD, but look at all the folks who've flocked to MP3 players."

True, but when you're talking about the actual experience of listening to the music it becomes irrelevant what medium the data is stored in, so long as the sound quality is good. The same does not apply toward the paper vs. monitor experience, at least not with today's display technology. One day, perhaps, but probably not until someone solves the screen glare and resolution problems.

> True, but when you're talking about the actual experience
> of listening to the music it becomes irrelevant what
> medium the data is stored in, so long as the sound
> quality is good. The same does not apply toward the paper
> vs. monitor experience, at least not with today's display
> technology. One day, perhaps, but probably not until
> someone solves the screen glare and resolution problems.

From what I've read about existing e-paper technology at MIT (and its corporate spin-off), e-paper *currently* has better display properties than typical mass-market printing (ie, a typical paperback book).

Better contrast, better resolution.

That's for black-and-white. Color e-paper is somewhat further behind. And, currently, e-paper costs a lot more than standard printing. But you only need one page to display an entire book, so production costs for the medium probably don't have to come down that far to become competitive.

Often, it's not the technical end of things that falls behind, but the design end. Even if fantastic and relatively affordable e-paper were available to manufacturers, it would be all too easy to turn out a bad product. And to fail to market it properly.

Sony didn't think the Walkman would sell. And a lot of big companies thought that Magic Book portable computers would be huge.

Most books are purchased with the worthy intention of being read, but in fact are purchase with the idea that knowledge can be owned. As such people will ultimately purchased books from the low cost provider which will not be print on demand, but rather the mass merchants such as Wallmart or Costco. Additionally, the perception exists that a hardbound book is of greater substance than a paperback. As you cast your eyes upon the liabrary shelves of your friend they will not want you to see a row of obviously unread paperbacks.

Most books are purchased with the worthy intention of being read, but in fact are purchase with the idea that knowledge can be owned.

I'd disgree with that. They ware _sold_ with the idea that knowledge can be owned - or at least that it can be copyrighted. ;-)

The great majority of purchasers couldn't care less whether other people also have access to the same knowledge. If copying a book was as easy as copying a computer file book piracy would be even more common that it already is.

As such people will ultimately purchased books from the low cost provider which will not be print on demand, but rather the mass merchants such as Wallmart or Costco.

The distinction here is between people who want to buy _a_ book and those who want a _particular_ book. Mass merchants are fine for 'beach books' - the kind of casual purchase of someone who has only a casual interest in reading. Traditional book shops cater for a wider range - trading off carrying less popular books vs. attracting the more discerning customers. Print On Demand in theory offers a range beyond that that can be supported by present practices. However my personal expectation is that Print On Demand would be a service available to _order_ books printed in some central location(s) and delivered to the bookshop / direct to home within a few days.

Additionally, the perception exists that a hardbound book is of greater substance than a paperback. As you cast your eyes upon the liabrary shelves of your friend they will not want you to see a row of obviously unread paperbacks.

Certainly this is a factor. In fact places sell 'wallpaper' books for this purpose. However, again, this depends on the people involved. Certainly I expect my friends to have books they have read paperback or otherwise - and my friends _know_ that my walls are covered in well read paperbacks.

However my personal expectation is that Print On Demand would be a service available to _order_ books printed in some central location(s) and delivered to the bookshop / direct to home within a few days.

It's entirely possible, especially if an online model becomes dominant. If you can go to, god help us, Amazon.com, order any book every published, and have it delivered to your home in three days, very few people will care if it was printed on demand or shipped from a warehouse.

Another possible model is the video-rental model. You'd go into a bookstore that had sample copies of books on its shelves. You'd select books you wanted and pay for them at the front desk, and your books would be ready in a short time -- ten minutes, say, for a single book, or an hour, for several.

Bookstores already combine with coffee shops, so I think that many consumers would be willing to wait, while others would go run other errands or come back the next day to pick up their books, as one might do with one-hour photo development. And you'd be saved the hassle of going from bookstore to bookstore looking for a book.

Of course, a blend of the two marketing models might be even more likely. Or perhaps good quality ebooks will take precedence.

I've heard that POD printers (capable of producing MMP-style paperbacks) could fall as low as $5000, if mass-produced. Even so, I would expect few homes to have one.

In 2010 there will still be plenty of books, and plenty of stores to sell them. This bet centers on the debate about production efficiency vs. inventory and product selection. It would be a logistical and operating nightmare for book retailers to get involved with a project such as this. Even if they decide to do it, it will fail, because the benefits do not justify the costs. There are countless examples of on-demand technology that just does not translate well into real-world retailing experience. Books are highly tactile and low tech. Retail POD goes against those two retail advantages.

No way this happens.

Ben Bova, editor of "OMNI" magazine, wrote a science fiction novel called "Cyberbooks" in which the story centers around precisely the point of Bet 6 - at what point will the multiple inefficiencies of publishing as we know it lead us to use electronics to distribute books?

I have to confess that I would settle for being able to buy and download the novel of my choice to my laptop.

In fact, my idea of heaven (at least one part of it) would be to wake up and find that some kind soul had scanned every book, magazine, movie and music album I own to a large-capacity (800 gigs ought to do it) hard drive, which I would then be able to slide into my laptop and bring along with me.

Imagine the freedom! Moving would no longer involve moving half a ton of paper, vinyl and bookcases from point A to point B. The major obstacle (to my wife and me, anyway) to living on a boat or motorhome would no longer exist.

Anyway, getting back to Bova's novel, the poor inventor of the "Cyberbook" is thwarted at every turn by all of the people whose oxen would be gored by a wholesale shift to electronic publishing and distribution - the assorted people WITHIN the publishing industry whose jobs depend on paper books being moved around on trucks, the companies who depend on publishing being as inefficient as it now is, the Mafia (ditto)... eventually the hero decides to slide around the roadblocks by calling his invention a "toy" and selling it that way, then subverting the publishing market with a gadget that is now already a commercial success.

I only mention "Cyberbooks" because there are several real-life companies developing "digital ink" - something that looks, handles, etc like a paper book, but is actually a (say) 300-page text monitor. Moving just one of these around would be a lot better than buying many paperbacks at the local bookshop's printer/binder machine, then having to move all of your locally-printed books around.

I think it'll take longer than 2010 for book production to devolve to (say) Barnes and Noble or Borders. And by that time, several generations of us will have grown up to regard reading as something you do from your personal computer, not necessarily or even primarily with a book. The capital investment it would take to put a printing/ binding machine in every city - at the local Borders/ Barnes and Noble - will probably be used to make alternatives to paper books - probably several competing copy-proof methods of publishing books, magazines, textbooks, videos and music which protect copyrights while allowing efficient distribution, sale and storage of all that content.

A particularly intelligent way of doing this would be to allow "sharing" - distribution of copies of music or video in files that permit only one to three plays before the person with the copy must "register" it by paying the publisher to play it any more.

You could let people share printed works the same way - but only let them read a few pages - enough to get them "hooked" - before making them pay for the work in question. Alternatively, public libraries could carry copies of new books and provide URLs where people who liked one book in particular very much could buy a copy for themselves.

That way, distribution is handled for free over the Internet, and all the artist/author and publisher have to do is collect the money.

The book-publishing industry is currently struggling mightily and has already, recently, undergone a business revolution, with many large, old publishing companies being bought up and consolidated by larger companies that were not traditionally involved in publishing book, or even with publishing at all.

Meanwhile, paper costs continue to rise, profit margins continue to shrink, and competition is fiercer than ever. If there was ever an industry ready for a paradigm shift, it's book publishing.

Moreover, POD books present almost no difference to the consumer. The books would, in some cases, be purchased slightly differently, but the books themselves would be indistinguishable from current paperbacks. The biggest difference would be that availability would go way up. I don't see consumer resistance being a serious problem, anymore than it was for one-hour photo development.

But consumer resistance to electronic books has been enormous, dooming several generations that had major industry push behind them. Accordingly, I don't see on-screen books supplanting 'paper' books any time in the next thirty years, if ever.

I put 'paper' in quotes, though, because I think it very possible that 'smart' books will become a big deal in the near future, very possibly by 2010. So-called e-paper improves all the time -- Phillips just unveiled a new version capable of full-motion color video -- in a mixed display, no less -- and the technology uses very little power. If production costs can be brought down (which every report I've seen says will happen), we could all have books whose print changes on command.

After all, you only need black and white for a normal book, and a 'smart book' would have the same look and feel as a regular book. It would just have fewer pages, and the text and images displayed on the pages could change like those on a TV or computer monitor. The spine of the book would hold the batteries or fuel cell, the memory storage, a simple processor, and the I/O system.

I'll personally be surprised if such devices aren't available for under $100 by 2010. Very possibly under $50. It'll be the next MP3 player.

"Meanwhile, paper costs continue to rise, profit margins continue to shrink, and competition is fiercer than ever. If there was ever an industry ready for a paradigm shift, it's book publishing."

Agreed. I think that electronic viewing of text (whether on our PDAs, which is how I read public domain works such as the work of Jules Verne or Nathaniel Hawthorne, our laptops, or on Ebooks of one sort or another) IS the new paradigm. The exact form of the new paradigm will depend on many imponderables.

"Moreover, POD books present almost no difference to the consumer. The books would, in some cases, be purchased slightly differently, but the books themselves would be indistinguishable from current paperbacks. The biggest difference would be that availability would go way up. I don't see consumer resistance being a serious problem, anymore than it was for one-hour photo development. "

Actually, as long as Jeff Bezos can mail you paperbacks in a day or so as cheap or more cheaply than a POD copy, amazon.com will dominate the book distribution market, the exceptions being those of us who either distrust the Internet as a medium of commerce or go to Borders for social purposes.

What will drive a stake right through the heart of the paper book market is the subtraction of the cost of paper from the price of a book. I'd want the book on a CD-ROM, just in case the mass storage device on my computer dies, but still it should be possible to distribute books that way more cheaply than printing them.

Supposedly, it will soon be considerably cheaper for Amazon to print books on demand and then send them to consumers. There wouldn't be an appreciable delay from the consumer's point of view. Amazon would no longer have to deal with warehousing or suppliers, directly or indirectly. They'd just need an electronic library of texts they'd secured the right to distribute, pretty much the same as any publisher.

For a big chain like Barnes & Noble, it's supposedly already cheaper to use POD technology -- but that would mean abandoning rather considerable investment in the existing infrastructure. Big chains are slow to change paradigms (B&N was rather slow to move onto the web); small bookstores can't afford POD yet.

Personally, I expect the technology price will continue to drop until the big chains (and Amazon) pick it up to avoid competition from the smaller chains. And, remember, a POD bookstore never has to tell you a book is out of print or out of stock.

A CD would be far cheaper than a paper book, but it's not a book-friendly format -- not especially reader-friendly. But you hardly need a CD: even a good-sized (200k word) novel can be stored in less than a megabyte (in Rich Text Format, say), properly formatted. A cover illustration, if included, wouldn't need more than another megabyte.

If you had an electronic book, you could download novels (smaller than MP3 Top 40 songs), or buy them through the mail or in stores, on flash memory sticks like digital cameras use. Or you could bring your e-book to the store and have the books you bought uploaded almost instantly. And, as I said before, I expect that e-books will soon (five years or less) cost about the same as current MP3 players.

As a graphic designer with extensive experience in the printing industry, I'd like to point out that paper is almost always the LEAST of the expenses involved in short-run (or single run) printing jobs. Most of the costs involved are for the set-up and maintenence of the printing equipment itself. The exception to this rule is ink cartridges for small private printers (like your home inkjet unit), which are usually priced so that the ink costs more per ounce than gold.

Availablitiy of print-on-demand technology is not the main issue here. We've had the technology to have our daily newspapers faxed directly to our homes for at least two decades, but most subscribers still have a hard copy delivered to their doorstep every morning. Why? Because it costs one heck of a lot less to run one printing press that churns out a million copies of a newspaper in a few hours than to operate a million fax machines that each print out one copy a day. What could cost several dollars to print "on demand" costs only pennies to print at a central location and have shipped.

Perhaps one day improved printing efficiency will cause a paradigm shift for book sales... but I certainly don't see it happening in the next six years. Economics, infrastructure, and the way we buy books are going to be slower in changing.

I don't think cost has much of anything to do with why people have newspapers delivered to their homes. Reading newspapers online can deliver the same content at less cost (if you already have an ISP), but plenty of people still have newspaper subscriptions.

I don't know of anyone who would want to have a fax machine spit out an entire newspaper every day -- even if it were free. The resulting 'newspaper' would be in an less pleasing and less convenient form, and form is very important to content. Plus, the majority of people I know don't have a fax machine in their homes in the first place.

The advantages of POD go far beyond cost considerations, and I don't think cost will be what will make it catch on.

"The advantages of POD go far beyond cost considerations, and I don't think cost will be what will make it catch on. "

If the POD books are more expensive than those purchased through traditional book distribution networks, then what are the advantages of POD to the consumer? I only see one- the consumer is guaranteed that the book they are searching for will not be out of stock. However, I don't believe that most consumers experience this problem frequently enough to accept higher prices and longer waits at the bookstore. (It would take much longer for a bookstore employee to print, bind and cut a paperback than to simply ring up the copy that the customer has in hand.)

I would never suggest that the problems associated with cost, quality and convenience from POD systems are completely insurmountable, or that such a system will not become commonplace some time in the next century. However, I do not believe it is reasonable to expect that the majority of booksellers worldwide will not only make such technology viable, but completely restructure their distribution chains, in less than six years.

I think that the advantage of never having a book out of print is what will make POD catch on, simply because big booksellers use availability as another way of attracting customers.

Japanese manufacturers say that within a few years a $5000 printer will be able to produce a mass-market paperback book in about ten minutes. All the bookstore employee will have to do is select the book to be printed and press a button; the machine does the binding, etc. To a large bookstore, the cost is small and the customer service and marketing considerable.

After POD becomes part of the bookselling business, large chains will quickly adjust. They'll only need a fraction of the warehousing they currently have, and they won't need to return 'destroyed' unsold books to publishers (which will stop accepting them, anyway).

Fast-moving product, like bestsellers, will still be stocked and shelved, but not as extensively, because JIT (Just In Time) production will be easy, since large stores will have multiple, higher-volume POD printers. Think of the video rental store, where a single copy of most movies is on the shelves but highly popular films are represented several times. The difference is that if the book you want is already taken (ie, the shelf copy is there but the stock has been sold out), you have to wait maybe ten minutes for a replacement to be produced.

Or you could have it mailed to you. Naturally, POD works even better for mail-order places like Amazon. There, the best business model would probably remain to have bestseller printed en masse, as they are today, and lower-volume books produced on demand.

Will the market shift so much, within six years, that the bet is a winner? Probably not, but I don't think the bet is too far off.

I believe that paper books will become somewhat of a 'vintage' thing. Soon, you will probably be able to purchase literature online and read via a PDF format. Those who buy books will buy them out of nostalgia and having a high-tech digital monstrosity crank out your book in front of you will deter the few consumers that you’d have.

Too early. Libraries will most definately change completely into book-printing shops, but not as soon as predicted. Ten years further sounds more reasonable.

Since so little progress has been made by 2007, it has become very unlikely that this prediction can materialize on time. I buy hundreds of books a year. I own one (1) book that was printed on demand.