[Translations: Japanese]

A book is a self-contained story, argument, or body of knowledge that takes more than an hour to read. A book is complete in the sense that it contains its own beginning, middle, and end.

In the past a book was defined as anything printed between two covers. A list of telephone numbers was called a book, even though it had no logical beginning, middle, or end. A pile of blank pages bound with a spine was called a sketchbook. It was unabashedly empty, but it did have two covers, and was thus called a book.

Today the paper pages of a book are disappearing. What is left in their place is the conceptual structure of a book — a bunch of text united by a theme into an experience that takes a while to complete.

Since the traditional shell of the book is vanishing, it’s fair to wonder whether its organization is merely a fossil. Does the intangible container of a book offer any advantages over the many other forms of text available now?

One can spend hours reading well-written stories, reports, and musing on the web and never encounter anything bookish. One gets fragments, threads, glimpses. And that is the web’s great attraction: miscellaneous pieces loosely joined.

There ARE books on the web. Lots of them. I posted one of the first full books that was in print on the web in 1994. But because you pass no border to reach these pages, bookish material tends to dissolve into a undifferentiated tangle of words. Without containment, a reader’s attention tends to flow outward, wandering from the central narrative or argument. The velocity of shifting focus creates a centrifugal force which spins readers away from the pages of the book.

A separate reading device seems to help. So far we have a tablet, pad, and handheld. The handheld device is most surprising. Experts had long held that no one would want to read a book on a tiny few-inch wide glowing screen, but they were wrong. By miles. Many people happily read books on their smart phone screens. In fact we don’t know yet how small a book-reading screen can go. There is an experimental type of reading called Rapid Serial Visual Presentation that uses a screen only one word wide. Your eye remains stationary, fixed on one word, which replaces itself with the next word in the text, and then the one after that. So your eye reads a sequence of words “behind” one another rather than in a long line next to one another. The screen does not need to be very large.

Other new screens are making a home for books. Reflective e-ink is currently overthrowing the old publishing world. This technology is a sheet of white paper reflecting the ambient light around it, covered with dark text that can change. To the average eye, the text on this special “paper” (actually a plastic sheet) looks as sharp and readable as traditional ink on paper. The first generation of this black and white e-ink has made the Kindle a runaway bestseller.

In this substantiation of e-ink, the “book” is a tablet, a plank, which holds a single page. The single page is “turned” by clicking a button on the plank, so that one page dissolves into another page. A key feature of e-books on e-paper is that the font size can be individually adjusted. You want bigger type? Just dial it up and the entire book reflows into your desired form.

An e-ink page can be paperback-size, or it could be larger; the Kindle already comes in two sizes. As usage settles in, we are likely to see an e-book come with a recommendation: “This books is best viewed in a sized 3 tablet.” You’ll probably have more than one size 3 reader. Your favorite might be covered in soft well-worn leather, molded to your hand. The recommended reader for an arty magazine like Wired might be quite large. Perhaps it only rests on your coffee table.

But there is no reason an ebook has to be a tablet. Eventually e-ink paper will be manufactured in inexpensive flexible sheets. A hundred or so sheets can be bound into a sheaf, given a spine and wrapped with two handsome covers. Now the ebook looks very much like a book of old. One can physically turn its pages, navigate the book in 3D, and go back to an earlier place in the book by guessing where the spot was in the stack. To change the book, just tap its spine. Now the same pages show a different tome. Since using a 3D book is so sensual, it might be worth purchasing a very fine one with the most satin, thinnest sheets.

Personally, I like large pages. I want an ebook reader that unfolds, origami-like, into a sheet at least as big as a newspaper today. Maybe with as many pages. I don’t mind taking a few minutes to fold it back into a pocket size packet when I am done. I love being able to scan multiple long columns and jump between headlines on one plane. The MIT Media Lab and other research labs are experimenting with prototypes of books that are projected via lasers from a pocket device onto a nearby flat surface. The screen, or the page, is whatever is handy.

At the same time, a screen that we watch can watch us. The tiny eyes built into your tablet, the camera that faces you, can read your face. Prototype face tracking software can already recognize your mood, and whether you are paying attention, and more importantly where on the screen you are paying attention. It can map whether you are confused by a passage, or delighted, or bored. That means that the text could adapt to how it is perceived. Perhaps it expands into more detail, or shrinks during speed reading, or changes vocabulary when you struggle, or reacts in a hundred possible ways. There are numerous experiments playing with adaptive text. One will give you different summaries of characters and plot depending on how far you’ve read.

Such flexibility recalls the long expected, but never realized, dream of forking stories. Books that have multiple endings, or alternative storylines. Previous attempts at hyper literature have met dismal failure among readers. Readers seemed uninterested in deciding the plot; they wanted the author to decide. But in recent years complex stories with alternative pathways have been wildly successful in videogames. (And by the way there’s a lot of reading need in many games.) Some of the techniques pioneered in taming the complexity of user-driven stories in games could migrate to books.

Particularly books with moving images. We don’t have a word for these yet. Books with lots of still pictures we call picture books or coffee-table books or art books. But there’s no reason images in digital books must remain static. And no reason to think that are movies. On a screen we can marry text and kinetic images, one informing the other. Text inside of moving images as well as images inside of text. A few interactive diagrams produced by the New York Times and Washington Post have come closest to this marriage of word and movement.

This hybrid of movies and books will require a whole set of tools we don’t have right now. Presently it is difficult to browse moving images, or to parse a movie, or to annotate a frame in a movie. Ideally we’d like to manipulate kinetic images with the same facility, ease and power that we manipulate text — indexing it, referencing, cut and pasting, summarizing, quoting, linking, and paraphrasing the content. As we gain these tools (and skills) we’ll make a class of highly visual books, ideal for training and education, which we can study, rewind, and study again. They will be books we can watch or TV we can read.

When a table can double as the display of a book, and a book can be something we watch, we have to return to the question of what constitutes a book. And what happens to it once it is born digital?

The immediate effect of books born digital is that they can flow onto any screen, anytime. A book will appear when summoned. The need to purchase or stockpile a book before you read it is gone. A books is less an artifact and more a stream that flows into your view.

The current custodians of ebooks — Amazon, Google and the publishers — have agreed to cripple the liquidity of ebooks by preventing readers from cut-and-pasting text easily, or to copy large sections of a book, or to otherwise seriously manipulate the text. But eventually the text of ebooks will be liberated, and the true nature of books will blossom. We will find out that books never really wanted to be telephone directories, or hardware catalogs, or gargantuan lists. These are jobs that websites are much superior at — all that updating and searching — tasks that paper is not suited for. What books have always wanted was to be annotated, marked up, underlined, dog-eared, summarized, cross-referenced, hyperlinked, shared, and talked-to. Being digital allows them to do all that and more.

We can see the very first glimpses of book’s new found freedom in the latest Kindles. As I read a book I can (with some trouble) highlight a passage I would like to remember. I can extract those highlights and re-read my selection of the most important or memorable parts. More importantly, with my permission, my highlights can be shared with other readers, and I can read theirs. We can even filter the most popular highlights of all readers, and in this manner begin to read a book in a new way. I can also read the highlights of a particular friend, scholar or critic. This gives a larger audience access to the precious marginalia of another author’s close reading of a book (with their permission) a boon that previously only rare-book collectors witnessed.

Reading becomes more social. We can share not just the titles of books we are reading, but our reactions and notes as we read them. Today, we can highlight a passage. Tomorrow we will be able to link passages. We can add a link from a phrase in the book we are reading to a contrasting phrase in another book we’ve read; from a word in a passage to an obscure dictionary, from a scene in a book to a similar scene in a movie. (All these tricks will require tools for finding relevant passages.) We might subscribe to the marginalia feed from someone we respect, so we get not only their reading list, but their marginalia — highlights, notes, questions, musings.

The kind of intelligent book club discussion, as now happens on GoodReads, might follow the book itself and become more deeply embedded into the book via hyperlinks. So when a person cites a particular passage, a two way link connects the comment to the passage, and the passage to the comment. Even a minor good work could accumulate a wiki-like set of critical comments tightly bound to the actual text.

Indeed dense hyperlinking among books would make every book a networked event. Right now the best a book can do is to link to another book’s title. If another work is mentioned in passing or in its bibliography, an ebook can actively link to the whole book. Much better would be a link to a specific passage in another work, a technical feat not possible yet. But when we can link deeply into documents at the resolution of a sentence, and have those links go two ways, we’ll have networked books. This, by the way, was Ted Nelson’s original vision of the docuverse. (He also envisioned a micropayment and credit system built on top for a full literary economy.)

You can get a sense of what this might be like by visiting Wikipedia. Think of Wikipedia as one very large book — a single encyclopedia — which of course it is. Most of its 27 million pages are crammed with words underlined in blue, indicating those words are hyperlinked to concepts elsewhere in the encyclopedia. Wikipedia is the first networked book. In the goodness of time as all books become fully digital, every one of them will accumulate the equivalent of blue underlined passages as each literary reference is networked within that book and all other books. This deep rich hyperlinking will weave all networked books into one large meta-book, the universal library. Over the next century, scholars and fans, aided by computational algorithms, will knit together the books of the world into a single networked literature. A reader will be able to generate a social graph of an idea, or a timeline of a concept, or a networked map of influence for any notion in the library. We’ll come to understand that no work, no idea, stands alone, but that all good, true and beautiful things are networks, ecosystems of intertwingled parts, related entities and similar works.

Wikipedia is a book that is not only socially read, but socially written, famously so. It is still unclear how many books will be collectively written. Clearly many scientific and technical works will be constructed via decentralized collaboration because of the deeply collaborative nature of science. But the central core of most books will probably continue to be author by a lone author. However the auxiliary networked references, discussions, critiques, bibliography, and hyperlinks surrounding a book will probably be a collaboration. Books without this network will feel naked.

The complete universal library, all books in all languages, will soon be available on any screen. There will be many ways to access a book, but for most people most of the time, any particular book will essentially be free. (You’ll pay a monthly fee for “all you can read.”) Access is easy, but finding a book, or getting it attention will be hard, so the importance of the book’s network will grow, because the network is what brings in readers.

One quirk of networked books is that they are never done, or rather that they become streams of words rather than monuments. Wikipedia is a stream of edits, as anyone who has tried to make a citation to it realizes. Books too are becoming flows, as precursors of the work are written online, earlier versions published, corrections made, updates added, revised versions approved. A book is networked in time as well as space.

But why bother calling these things books? A networked book, by definition, has no center, and is all edges. Might the unit of the universal library be the sentence, or paragraph, or chapter article instead of a book? It might. But there is a power in the long form. A self-contained story, unified narrative and closed argument has a strange attraction for us. There is a natural resonance that draws a network around it. We’ll debundle books into their constituent bits and pieces and knit those into the web, but the higher level organization of the book will be the focus for attention — that remaining scarcity in our economy. A book is an attention unit. A fact is interesting, an idea is important, but only a story, a good argument, a well-crafted narrative is amazing, never to be forgotten. As Muriel Rukeyser said, “The universe is made of stories, not atoms.”

At the moment we are in a scramble to find the right container for digital books. Released from their paper shells, books seem to need more than the open vastness of the web. They like the viral compactness of a PDF, but not its rigid appearance. The iPad is sensual and intimate (like the content of books) but currently heavy in the hand. The Kindle has the advantages of focusing attention, which they like. The latter two containers charge for their convenience and interface, which feeds authors. Books can appear on any screen, and will be read anywhere it is possible to read them, but I think they will gravitate toward favorable forms that optimize reading.

In the long run (next 10-20 years) we won’t pay for individual books any more than we’ll pay for individual songs or movies. All will be streamed in paid subscription services; you’ll just “borrow” what you want. That defuses the current anxiety to produce a container for ebooks that can be owned. Ebooks won’t be owned. They’ll be accessed. The real challenge ahead is finding a display device that will focus the attention a book needs. An invention that encourages you onward to the next paragraph before the next distraction. I guess that this will be a combination of software prompts, highly evolved reader interfaces, and hardware optimized for reading. And books written with these devices in mind.