Myth of the Universal Digital Library

A lot of Web geeks believe that one day everything ever created by humans will be available online. Call it the myth of the universal library. Here's how the myth goes: Because there is unlimited real estate in cyberspace and because media can be digitized, we can fill cyberspace with all human knowledge and give everyone access to it. Without further ado, I present to you three arguments for eliminating the myth of the universal library.

1. Cyberspace does not exist. The term cyberspace was invented in the late 1970s by a science fiction writer named William Gibson, who used it to describe a "consensual hallucination" experienced by people who were neurologically linked to computer networks. Even within Gibson's novels, the author is careful to explain that the illuminated buildings, glowing roads, and avatars that his heroes meet in cyberspace are simply convenient representations of abstract data structures.

My point is that computer networks are not space and they are not real estate. They are data storage and manipulation devices connected together by wires and radio waves. They cost money and require massive amounts of power. They take up real-world space. And they break. In other words: No computer network is infinite. Storing all of human knowledge on a computer network would be expensive and intensely difficult to maintain. There is no infinite cyberspace -- only finite computer networks subject to wear and tear.

2. Your human knowledge sucks. I was recently in a very interesting conversation with several smart librarians, all of whom are keen to use computers for preserving and disseminating information. Somebody pointed out that a good example of publicly accessible universal knowledge is the French Gaumont Pathé Archives, which makes hundreds of hours of searchable historic newsreel footage available online for free. The problem, as film archivist Rick Prelinger pointed out, is that the Gaumont Pathé project, like many of its kind, has had to pick and choose which films it can afford to archive. So the group focused heavily on politics and left the fashion and pop culture reels undigitized and therefore less accessible. The guy who'd brought up the archive thought this was just fine.

"No, it's not," Prelinger replied. "If you want to know what everyday people cared about historically, fashion is going to tell you a lot more than newsreels about famous politicians."

The point is, people don't agree on what "all of human knowledge" means. Is it great art and political history? Or is it xeroxed zines and fashion history? Who decides what gets digitized and what gets tossed into the ashtray of the unsearchable, the unnetworked? Do commercials go into our mythical universal library? What about hate speech and instruction manuals for hair dryers? Are those documents not also part of human knowledge? We will never reach an agreement on what all of human knowledge really is, and therefore we will never be able to preserve all of it.

3. Digitizing everything is impossible. Consumers can buy terabyte-size disk storage. The glorious Internet Archive buys petabyte storage devices by the bushel. You can fit your entire music collection in your pocket, and your book collection too. But even if we agreed on what all of human knowledge really is -- which we never will -- you couldn't digitize all of it. Turning books into e-books takes time, as does turning film and television into digital video files. And what about rare scrolls, artworks, and machines? How do you put them online? Some medieval manuscripts and textiles are so delicate they can't be exposed to light. Making something digital isn't like waving a wand over it -- poof, you're digital! No matter how hard we work and no matter how much money we throw at this problem, there is simply no way to turn all physical media into digital formats.