Draft of my talk on the end of information at the Berkman Center. [NOV 11: Here's the video of the webcast done on Nov 9. Ethan Zuckerman's extensive and amazing live blogging of the talk is here..]
I have been working for weeks on a talk I’m giving at a Tuesday lunch at the Berkman Center, where “work on” means erasing more than I’ve written. I’ve done more complete rewrites than I can count, mainly because I can’t figure out what the point of the talk is. I started out knowing what the point was, but as I actually wrote it, I knew less and less. So, here’s a rough outline of the current sorry state of the talk.
I. Information has been the dominant metaphor
This is the easy part. From cradle to grave, we’ve reconceived of ourselves and our world as information. But, except for the technical definition, we don’t know what it is (and most of how we’ve reconceived of ourselves has nothing to do with the technical def, and most of us don’t know the technical def anyway).
II. A discontinuous history
“Info” has two ordinary senses that precede its take-over by Claude Shannon in 1948: It’s something you’re about to learn, and it’s the content of tables. Shannon then introduced his technical definition, which only a tiny percentage of the population understands. Nevertheless, info became the dominant paradigm. So, what enabled it to take over our culture? Two notes: 1. I am explicitly not going to talk about its utility or its politics of control and mastery, both of which are obviously crucial to the answer. 2. I am going to contrast the Info Age with the Link Age (or whatever we’re going to call the new epoch).
Enabler 1: Information scales
Info scales sufficiently to enable large corporations to manage themselves. But its scaling strategy is to exclude everything that doesn’t fit its rows and columns. E.g., the personnel database contains only a tiny bit about what employees know about one another. In the Age of Links, we include everything. Links create a world of abundance. The irony is that while the Info Age’s strategy was to exclude bad and useless info, in the Age of Links we’re better able to manage the abundance of crap than the abundance of good stuff.
Enabler 2: Info is a resource
It’s a resource in that it’s useful to us. We can retrieve stuff from it, using the criteria of precision and recall: Did our query get only the right stuff and all the right stuff? In the Link Age of Abundance, however, getting all the right stuff is a disaster. (Which is why we invented two new criteria: relevance and interestingness.)
Furthermore, info is a resource from which we fetch nuggets of value. The Web, though, is a place that we enter and navigate. The irony is that in the Age of Info, we thought about entering an info space as becoming Jeff Bridges in Tron. Or, we thought that if we entered the info space because it engulfed us, it would be a cold world of men with clipboards, as in movies such as Desk Set. In the Link Age, the place we enter is fully social, and is becoming completely integrated with the real world space.
Enabler 3: Bits apply to everything
We sometimes talk about atoms vs. bits because anything can be turned into a bit. Bits are thus coextensive with the universe. But, bits can represent anything in the world because they are so fundamentally unlike the world. Every other measurement measures some property of the world (height, weight, shoe size, whatever), but bits measure pure difference. The world bits model always shows itself in particular ways, in particular properties. Bits are thus profoundly unnatural; they exist only because we take them as bits. They are thus very much unlike atoms.
Further, bits reduce everything to the simplest of differences: yes/no, 1/0. Links, on the other hand, are put in place to find and tease out differences that are complex enough to require language and to be worth pointing out.
Enabler 4: Information explains communication
Although Shannon expressly was not trying to explain human communication, his diagram matches our basic view of communication as the movement of code through a conduit. (Paul Edwards is good on this, as on many other issues.) Plus, Shannon’s popularizer, Warren Weaver, expressly said the theory applies to people speaking, pipers piping, dancers dancing, and just about every other form of communication. Still, we have to ask why think of communication as the process of moving symbols through conduits when so much else is required, and so much more is implied, by even the simplest of human conversations. Part of the answer is, I think, our Cartesian metaphysics that thinks that we experience representations of the world, and thus can only communicate by shipping messages to others that affect their representations of the world. The world itself has dropped out of this equation: We only have heads and conduits between them.
This basic picture of communication of content moving through a medium to a receiver treats communication as an obstacle to be overcome, for noise keeps banging on the conduit. This is how the world looks if you come out of an experience where communication was difficult, as was the case for the early info scientists, some of whom had worked on how to improve communications on a noisy battlefield. (Paul Edwards again: The Closed World is excellent.) But hyperlinks are neither content nor medium; more exactly, they’re both. Like a path, a hyperlink assumes an existing world, a shared ground. (Links are a very special sort of path, though, because they are generative of their world.)
Enabler 5: Information lets us understand the world
Models let us find what is essential and common among all that which they model. But they deny the abundance of the world and the fact that the world doesn’t behave the way we want. The contingent does show up in the Info Age view of the world. It shows up as noise. In the Link Age, succeed by making the world noisy: creating a path among ideas that differ. (This is not noise in Info Theory’s sense.) Of course, we rightfully worry that amidst this differential linkage we will only seek that which is familiar and reassuring. The success of the Link Age depends upon it remaining as noisy and full of difference as possible, the opposite of how the Info Age measured success.
So, as I write this out, I can see some sections that don’t really add up. For example, Enabler 3’s discussion is pretty incoherent. But that’s why I’m writing this out now.
I have one day left to get something presentable out of this, since I am out all day on Monday. And I’m jetlagged and pretty exhausted now. Ack.