The Dream of the Archive

"I have always imagined that Paradise will be a kind of Library" -- this quote from the Argentinian writer Jorge Luis Borges once adorned a bookmark I got from the Brown Bookstore. And the man knew whereof he spoke, having spent much of his life in charge of the National Library of Argentina. There, Borges was free to indulge his ideas, and to imagine the kinds of strange, playful worlds that his fictions became known for.

In his quintessential story, "The Library of Babel," Borges masks a seemingly simple question: What if there were a library that contained every possible book? There are limits: each book has but 410 pages; each page 40 lines, each line 80 characters, randomly generated. Of course, 99.99999% of the books will be utter nonsense ... but of those that are not, every book one can desire, imagine, or write would already exist: the true story of your life, the unfinished tales of the Canterbury Tales, the true catalog of the library (and a nearly infinite number of false catalogs), you name it. Of course, most of the titles, as this site which generates them illustrates, will have less memorable titles, such as the puzzling "IJJRE TIORBQBRXYG RXBASM HNF GAJF QNHODP" or its no less baffling shelfmate "FFX BIPB MPMXLCO YQTYX XRO AOQ."

In Borges's world, the librarians themselves don't know what they're really seeking, or whether they will find it. In the end, many of them despair, committing suicide by jumping into one of the Library's stairwells. And yet here, with the stairs nearly infinite, they fall through space but never reach the bottom, slowly becoming mummified, and turned thence to a fine dust. A couple of years ago, I sent in a submission to NPR's "Three Minute Fiction" contest" -- the prompt was to begin a story with the sentence "She closed the book, placed it on the table, and finally, decided to walk through the door." I thought at once of Borges's library, and this story was the result; that it didn't win the contest I can attribute only to blatant favoritism.)

Fast forward to the present: The World Wide Web presently contains roughly 47 billion "pages," each of which contains, by one estimate, an average of about 6.5 printed pages of material -- which gives us a rough estimate of 305.5 billion pages. This is, of course, only a minuscule atom compared to Borges' library (which can be calculated at 25 to the power of 1,312,000), but it's nevertheless well past the capacity of any human reader to peruse. At the average rate of 300 words, roughly one page, per minute, it would take a single reader 1,394 years to read them all, assuming the reader never sleeps. And, as with Borges's collection, the number of variants of any single text is enormously high, and with it the potential quantity of inaccurate or completely false information. Of course this is only the "static" web -- a single dynamic site such as Twitter reports roughly 200 billion Tweets per year!

So we are the new librarians. And, if we want to avoid their despair, we must learn to sort and sift through the enormities of what's available if we want to discover useful, accurate, relevant information.