May 5, 2006 · Tim Wu
Jane Jacobs, the great theorist of all things urban, died recently. It had been my dream to go find her in Toronto but that will never happen. She’s obviously influential to urban planners, but I’ve found her writing tremendously helpful for thinking also about network design.
If you aren’t familiar with her work, Jacobs was an enemy of bad central planning. She believed in cities that grew up in a willy-nilly, unpredictable way, allowing new buildings to gradually replace old, or be converted to new purposes. She believed the causes of urban blight were dullness, and hated housing projects, mega-blocks and other doomed efforts to make people live just so.
What Jacobs favored is letting neighborhoods be. She thought city planners ought create small roads and small blocks that worked on a human scale, and then stand back let the inhabitants decide how best to use their neighborhoods. Here thinking wasn’t quite economics or sociology, liberal or conservative, but rather a powerful attack on our constant tendancy to overestimate our own abilities to plan how people should live their lives.
The comparisons to network design should be obvious. Network designers, like say the writers of ATM, who have too specific an idea of what they want their users to do create abominable networks that imprison their users and become obsolute quickly. The more general purpose and useful the network, the more it does for society and individuals, and the better it evolves from one use to another.
Consider the comparison: a SoHo building can begin life as a factory, become an artist’s loft, then a boutique, then a condo, and so on. Some of the networks and even applications have led constantly evolving lives. The internet supported usenet, gopher, veronica, the web, ICQ, IM and so on, in a steady kind of evolution that was unpredictable in advance. The WWW itself has shuffled through static sites, through “home pages” of the Geocities era, through the rise of the search engine, through the blog, and through 2.0-style sites. Someone, maybe Danah Boyd, should write “The Death and Life of Great American Applications.”
Jacobs understood that the point of urban planning was not planning for a moment, but trying to cultivate healthy, evolving cities that make people happy to live in. Much of the same can be said about information architectures – the best planned networks don’t overplan, but somehow manage to create a kind of life of their own.
You can learn this in The Death and Life of Great American Cities, or any of Jacobs’ other books.