MISCELLANEOUS FROM A TO Z

From page 97 of: “Everything is miscellaneous : the power of the new digital disorder” by David Weinberger. Holt Paperbacks : Henry Holt and Company, LLC, 2007.

“On paper, it sounds like a terrible idea. Build an encyclopedia by letting anyone create or edit an article, even anonymously. Yet four years after its launch at the beginning of 2001, Wikipedia had more people reading its pages than the New York Times' Web site did. By the middle of 2006, Wikipedia boasted over a million articles in its English edition, with more than a hundred editions in other languages.

The traditional sources of authoritative knowledge have begun to pay close attention to the new kid on the block, and not only to its content. Traditionally, the articles in a work that size would be carefully arranged. But Wikipedia's organization is as bottom-up as its content.

The Encyclopaedia Britannica does not have the luxury of being as thoroughly miscellaneous as Wikipedia. If we're looking for the Britannica's article on elephants, we count on being able to open the volume with the E stamped on its spine and page through alphabetized entries until we get to the one we want. If we're feeling adventurous, we can check out the carefully planned cross references at the end of the article. Or we can go to Mortimer Adler's Propaedia to find a family of Adler-approved concepts related to elephants. Either way, we are able to find information in the Britannica precisely because it isn't miscellaneous.

At Wikipedia, there are no volumes—not even digital representations of volumes—to thumb through. There is an alphabetical listing of the topics, but it's poorly done—Mortimer Jerome Adler is listed under the M's—probably because the listing is rarely used. There are tens of thousands of entries for each letter, on average. That's a lot of riffling, whereas with eight keystrokes and a press of the Enter button, you could have searched for elephant and found the article about pachyderms instantly. At the top of the elephant article, there's a link to a page that lists all the other articles in Wikipedia you might have meant to find when you typed elephant into the search box: a film by Gus Van Sant, an album by the White Stripes, a World War II German antitank vehicle, a brand of beer, or the 105th chapter of the Koran. Wikipedia reminds us that even a word as simple as elephant has a touch of the miscellaneous about it.

Even if you use Wikipedia's alphabetical index, the pages are not really in alphabetical order. In fact, a Wikipedia article isn't a single object. Although an article's Web page looks unified to the reader, as with many pages on the Web, its text, graphics, and formatting rules are each stored separately and are pulled together only when a user requests a page by clicking on a link. If you search for elephant at the Wikipedia site, it's probably the computer named Vincent (after Vincent of Beauvais, a Dominican priest who compiled an encyclopedia with 3,718 chapters in the thirteenth century) that comes up with the list of articles that use the word. If you click on the link to the main article, this sends a request to another computer, which checks to see if that article was recently requested by someone else; if so, a copy of that page is kept ready to go and a third computer—perhaps the one named Will Durant, after the historian of philosophy—simply sends the page you're looking for. If not, Wikipedia sets about constructing the page for you. It randomly looks at one of the half dozen computers (including one named after Mortimer Adler) that store the complete text of the current articles in Wikipedia. Wikipedia then looks on Bacon (named after the philosopher Sir Francis Bacon) or one of the other computers that store the graphics, and passes both the text and the graphics to one of the dozens of computers that do nothing but assemble contents into Web pages based on templates. The finished page is then passed to your computer, where you see a text-andgraphics page about elephants.

Another level down, Wikipedia, like all computer applications, is even more miscellaneous. The computer may decide to store any single element of an article—say, the text or a photo of an elephant—in discontinuous sectors of a hard drive in order to fit the most data onto the drive and to optimize the time it takes to retrieve all those bits. That's why when I asked Brion Vibber, the chief technical officer of the Wikipedia organization, where the text information for the elephant article is actually stored, he replied, in the chat room we

were in:

<brion> god only knows.

<brion> On the disk somewheres

A shame-faced admission of an appalling ignorance? Not at all. The gap between how we access information and how the computer accesses it is at the heart of the revolution in knowledge. Because computers store information in ways that have nothing to do with how we want it presented to us, we are freed from having to organize the original information the way we eventually want to get at it. The bits and pieces of Wikipedia are, in effect, an enormous reserve of miscellaneous information that can be assembled in precisely the ways we need at precisely the moment we need it. That's true all the way through Wikipedia, from the microscopic bits stored on the hard drives to the finished articles we read. At the top level of this hodgepodge of bits, images, text, articles, and ideas, something remarkable happens. The million articles in English are not arranged alphabetically. They are not put into a Dewey-like categorization scheme. There is no controlled vocabulary. There is no usable overview. Yet this enormous miscellany gets organized richly and in tremendous detail. How it happens would have driven Mortimer Adler over the brink: Wikipedia articles are packed with hyperlinks created by anyone who takes the time to add one. No qualifications are required, and no expertise is needed beyond knowing that to link the word elephant in an article to its entry in Wikipedia, you type "[[elephant]]". In some entries, almost every second word is linked to another article. Together these links constitute a web of knowledge, communally constructed, ever shifting, and frequently extraordinarily useful. Wikipedia's hyperlinked web, like the Web itself, does not look like a tree. It is a far, far more complex structure. But its shape, freed from the two dimensions of paper, better represents the wild diversity of human interests and insight.”