stevenberlinjohnson.com

Air Purifiers For Data Smog

My latest Discover column takes a look at a question a lot of us are grappling with -- how do deal with information overload -- and speculates a little on whether, in effect, we can use software interfaces to solve a problem that software interfaces created in the first place:

People already prioritize by thresholds of concentration. That's why you may say to an assistant: “Please, don't bother me with calls –– unless it's my spouse.” Computers should be better at this kind of filtering, but they're not programmed to anticipate how your attention shifts from one minute to the next. Your e-mail client doesn't know that you're trying to focus on another, more pressing problem. But it would be easy enough to create protocols that define different modes of concentration. Many laptops have location settings that allow you to switch from office mode to home mode and thus change a whole host of settings. Why not offer a comparable option for defining different mental states?

In “focus” mode, a computer could automatically switch to a full-screen view of the current document and notify all other applications that you're to be bothered only in an emergency. You could define the criteria for breaking the cone of silence by creating a list of important people authorized to interrupt. Or your computer could compile the list in the background, by watching how quickly and reliably you respond to different people over time. That vice president from accounting always gets a rapid response, so the software automatically puts him on the white list. But all those unanswered e-mails from your mother-in-law? She doesn't make the cut when you're focused.

This sort of idea is clearly in the ether right now. The always fascinating Clive Thompson had a long piece a couple of weeks ago in the Times Magazine that tackled comparable issues, and Merlin Mann from the most excellent 43Folders site apparently proposed something quite similar to what I was imagining in the column a while back.

Why The Web Is Like A Rain Forest

The latest issue of Discover on the stands is their 25th anniversary issue, and so for my column this month my wonderful editor Dave Grogan asked me to do a kind of "state of the internet" piece. I decided to take a look at the premise of Web 2.0 that has been tossed around a lot over the last few years, without getting too bogged down in the details (i.e., what's the difference between Web 2.0 and the semantic web?) For most readers here, I suspect that the example scenario I give will be old news, but the more interesting stuff comes at the end. I think this is a useful way to talk about what's happening online right now, whatever you want to call it.

The difference between this Web 2.0 model and the previous one is directly equivalent to the difference between a rain forest and a desert. One of the primary reasons we value tropical rain forests is because they waste so little of the energy supplied by the sun while running massive nutrient cycles. Most of the solar energy that saturates desert environments gets lost, assimilated by the few plants that can survive in such a hostile climate. Those plants pass on enough energy to sustain a limited number of insects, which in turn supply food for the occasional reptile or bird, all of which ultimately feed the bacteria. But most of the energy is lost.

A rain forest, on the other hand, is such an efficient system for using energy because there are so many organisms exploiting every tiny niche of the nutrient cycle. We value the diversity of the ecosystem not just as a quaint case of biological multiculturalism but because the system itself does a brilliant job of capturing the energy that flows through it. Efficiency is one of the reasons that clearing rain forests is shortsighted: The nutrient cycles in rain forest ecosystems are so tight that the soil is usually very poor for farming. All the available energy has been captured on the way down to the earth.

Think of information as the energy of the Web's ecosystem. Those Web 1.0 pages with their crude hyperlinks are like the sun's rays falling on a desert. A few stragglers are lucky enough to stumble across them, and thus some of that information might get reused if one then decides to e-mail the URL to a friend or to quote from it on another page. But most of the information goes to waste. In the Web 2.0 model, we have thousands of services scrutinizing each new piece of information online, grabbing interesting bits, remixing them in new ways, and passing them along to other services. Each new addition to the mix can be exploited in countless new ways, both by human bloggers and by the software programs that track changes in the overall state of the Web. Information in this new model is analyzed, repackaged, digested, and passed on down to the next link in the chain. It flows.

Finding Lost

In my list of weird coincidences, this has got to rank pretty high: earlier this year, I'm flying to London to do the publicity tour over there for Everything Bad. I'm sitting next to a British guy, and and an hour or so into the flight, we start chatting. He asks if I'm flying over for business or pleasure, and I say business, and he asks what I do, and I say I'm promoting a new book.

Him: Oh really? What's the book?

Me: It's this funny little thing called Everything Bad Is Good For You.

Him: [smiling] Ah yes, I've read it.

Me: No way! It hasn't even come out in Britain yet!

Him: Actually, the paper where I work bought the first serial rights.

Me: How amazing. So you worked at the London Times? What do you do there?

Him: [somewhat sheepishly] I'm, um, the editor-in-chief.

Crazy, huh? I said later that not only was he the first person I'd randomly sat next to on a plane who'd actually read one of my books -- he was also the first person I'd randomly sat next to who had bought the first serial rights to one of my books.

I say all this because during the flight I was also watching episodes of the ABC show Lost on my laptop, and I brought it to my seatmate's attention because there are often shots of the original plane crash interspersed through the episodes, and I didn't want him looking over and thinking I was some kind of a freak who liked to watch footage of plane crashes while actually flying in a plane. So we got started talking about Lost, which had not made it over to the UK at that time. And then three months later, I get an email from an editor at the Times saying that the show is now airing over there, and that his boss recalls me saying something interesting about the show and did I want to write a piece about it.

I thought it sounded like a great idea, because in many ways Lost embodied a lot of the themes I was exploring in the book, but I didn't get around to seeing it until the manuscript was done. (As is so often the case, I was prodded into watching it by my longtime pop culture confidant, Alex Ross, who was a huge help when I was writing the book.) So I wrote a nice little column for the Times, which for the life of me I cannot find on their website. But about half of it will be familiar to most of you who have followed the Everything Bad discussion. Given that ABC is about to start airing season two of Lost next week, I thought I'd quote a little from the Lost-specific parts of the piece:

By genre, Lost is a disaster narrative -- closest in spirit to the airplane disaster movies over the seventies, which were as a genre so awful that they spawned an entire sub-genre of parodies. But Lost's creator J. J. Abrams -- who co-wrote and directed the show's breaktaking two-hour opening episode -- announced in the very first seconds of the show that this was no Airport 77 remake. If the networks had made Lost thirty years ago, it would have followed a fixed narrative flight path: introduce all the passengers and the pilots and the feuding stewardesses; learn each of their “backstories;” and then have the engines fail. Abrams did away with that entire prologue: Lost begins seconds after the crash, and so from the very beginning of the show, the twenty-odd survivors that we focus on are complete mysteries to the audience. We know nothing about them, and so the narrative pleasure comes from watching these interlinked histories being slowly revealed over the course of the season, in flashbacks and reminiscences.

Thirty years ago, of course, no American show would have dared to put twenty recurring characters into a network drama. (Even the socially complex prime-time soaps like Dallas tended to max out at around ten primary characters, while the sitcom's sweet spot seemed to be at around six: just enough for a nuclear family and the wacky neighbor next door.) But no show back then would have dreamt of submitting the audience to so much deliberately murky narrative information. Only the notoriously opaque Twin Peaks -- a minor network hit in the early nineties -- compares to Lost's entanglements. And indeed, Lost came very close to disappearing off the map itself: executives at ABC and its parent company Walt Disney had so little faith in the project that they fired the network chairman who had originally proposed the idea of a plane crash epic to Abrams. Only the lavish budget already spent on the pilot -- $12M, several times larger than television norm -- persuaded the network to give the show a chance.

Mystery, of course, is a staple of much serial drama. (Dickens, after all, compulsively ended his installments with a tantalizing cliffhanger.) But when American television has withheld information for the purposes of suspense, it has historically focused on a single unanswered question, “who shot JR?” being the canonical example. As uncanny as it was, Twin Peaks itself would have never attracted a mainstream audience without a central, catchphrase mystery at its core: who killed Laura Palmer?

The genius of Lost is that its mysteries are fractal: at every scale -- from the macro to the micro -- the series delivers a consistent payload of confusion. There are the biographical riddles: why was the beautiful Kate accompanied by a federal marshal on the flight? There are geographic riddles (“why have the rescue teams missed the island, and why does it appear to have a history of attracting castaways?”) and historical ones (“why has that SOS signal been playing for so many years?”) And then there are existential riddles: are these people even alive at all? Perhaps there were no survivors, and these characters are just ghosts haunting an island of lost souls. Or does Abrams have up his sleeve an elaborate homage to The Island Of Dr. Moreau?

These are only a handful of the unanswered questions that arise in the first six episodes. I did a back-of-the-envelope calculation and came up with roughly thirty genuinely mysterious plot elements that season one offers up. I won't give anything away, but each of the main characters turns out to be, well, more complicated than they initially appeared, and the island itself turns out to be the most complicated of all. Narratives by definition work by withholding information about future events; you tune in to find out what will happen next. But with Lost, the mystery lies in the present tense: half the time, you have no idea what's happening right now.

The Urban Long Tail

Like many of you, I suspect, I'm eagerly looking forward to Chris Anderson's upcoming book, The Long Tail -- partially because his pre-book blog has been so interesting, but also because Chris is one of the sharpest and least doctrinaire thinkers that I know. I've been following the Long Tail debates from afar, in much the same way that I followed the original power law debates that Clay Shirky started. But I've put together a new column for Discover that looks at the long tail question in the context of cities. This is an angle that I haven't seen explored to date; up to now, in fact, I think the long tail premise has a tacit anti-urban bias to it, since it used to require big city scale to find obscure long tail books or albums that are now readily available to anyone with an Internet connection. But some long tail services can't be Fedexed or downloaded -- people, for instance. I started thinking about all of this thanks to the ingenious Dodgeball service, created by two former students of mine. (I had nothing to do with the idea for the company itself, which is too bad for me, since it's just been bought by Google.) At any rate, you can read the whole article, but the main argument is this:

Dodgeball suggests an intriguing twist on long tail theory. As the technology increasingly allows us to satisfy more eclectic needs, any time those needs require a physical presence –– whether it's sipping your cold soup or meeting your crush in a bar –– the logic of the long tail will favor urban environments over less densely populated ones. If you're downloading the latest album from an obscure Scandinavian doo-wop group, geography doesn't matter: It's just as easy to get the bits delivered to you in the middle of Wyoming as it is in the middle of Manhattan. But if you're trying to meet up with other fans of Scandinavian doo-wop, you'll have more luck in Manhattan.

Open Letter To Hillary

Today's L.A. Times is running an open letter to Hillary Clinton that I wrote, responding to the whole Grand Theft Auto panic. Those of you who have been following the blog or who have read the book will know most of the argument, but there are some new little twists in this one. Here's the opening:

Dear Senator Clinton:

I'm writing to commend you for calling for a $90-million study on the effects of video games on children, and in particular the courageous stand you have taken in recent weeks against the notorious "Grand Theft Auto" series.

I'd like to draw your attention to another game whose nonstop violence and hostility has captured the attention of millions of kids –– a game that instills aggressive thoughts in the minds of its players, some of whom have gone on to commit real-world acts of violence and sexual assault after playing.

I'm talking, of course, about high school football.

A Cut Above

I have a fun little essay at the beginning of this month's Wired that basically argues that the current controversy over steroids in sports will be nothing compared to the ethical questions posed in the near future by elective "enhancement" surgery. (Thanks to my brother for suggesting the idea, and my brother-in-law for some of the medical background in the piece.) Here's a snippet:

In fact, there's a chance you've had one of these next-generation procedures: laser eye surgery. Great [baseball] hitters anticipate the type of pitch being thrown - fastball or curveball? - by detecting the rotation of the seams of the baseball, which means that good eyesight is as valuable to them as strength or agility. One study of more than a dozen players who had opted for laser surgery found that "players coming off eye surgery are likely to see substantial improvements in batting average and power."

But poor vision can hardly be considered an injury. At least a pitcher with a torn ligament can say that he required surgery to repair an injury to his arm. Some batters naturally possess more muscle tissue than others; artificially manipulating that endowment with steroids violates the ethics of the sport. Other batters naturally possess better vision than others, but for some reason artificially augmenting that endowment is perfectly acceptable. If the sport objects to taking a pill or applying a cream that temporarily changes your body's chemistry, surely it should be an even graver offense to reshape your cornea or reengineer tendon and bone structure.

Tool For Thought

This week's edition of the Times Book Review features an essay that I wrote about the research system I've used for the past few years: a tool for exploring the couple thousand notes and quotations that I've assembled over the past decade -- along with the text of finished essays and books. I suspect there will be a number of you curious about the technical details, so I've put together a little overview here, along with some specific observations. For starters, though, go read the essay and then come back once you've got an overview.

New York City: Now With Free Technical Support

The new issue of Wired has an opening essay from me in their Start section about NYC's amazing 311 service, which I think ranks alongside London's congestion charging program as one of the two most significant innovations in urban information technology over the past decade. (Yes, I know there are other implementations of both ideas in other cities, but these are two the most ambitious in scale.) Here's a clip from the article:

Is tonight's Central Park concert canceled due to rain? Call 311. Looking for the nearest day care center? Call 311. Is alternate-side parking in effect? Call 311. "So if someone calls in to report a homeless encampment, that call gets routed to the police department's quality-of-life division," Menchini says. "If someone wants to help the homeless, the call gets routed to the Department of Homeless Services."

[With 311] the government learns as much as the callers do. That's the radical idea at the heart of the service: Every question or problem carries its own kind of data. Menchini's system tracks all that information; just as the heralded CompStat system mapped problem crime areas with new precision, 311 automatically records the location of each incoming service request in a huge database that feeds info throughout New York City's government. Think of 311 as a kind of massively distributed extension of the city's perceptual systems, harnessing millions of ordinary eyes on the street to detect emerging problems or report unmet needs - like those worries about unrefrigerated insulin. (Bloomberg himself is notorious for calling in to report potholes.)

Already, 311 data is changing the government's priorities. In the first year of operation, noise was the number one complaint; the Bloomberg administration subsequently launched a major quality-of-life initiative combating city noise. Today, geomapping software displays streets with chronic pothole troubles and blocks battling graffiti - all integrated into custom dashboards on city officials' laptops.

Google For President

Two quick Google items: first I have a short piece in Slate about the strange reluctance of the campaigns to use Google Adwords as part of their advertising strategy. Second: Google Desktop Search is out! (Sadly, not on OS X.) I will not say another word about the implications of this, because John Battelle has said everything I would have said, plus about fifteen other things that wouldn't have occurred to me.

The Political Brain

Today's Sunday Times Magazine is running an essay of mine on the neuroscience of political affiliation -- as embryonic a line of research as you're likely to find, but an intriguing one. I first started mulling over the ideas this spring, when the Times covered an early study commissioned by two Democrat consultants that performed fMRI scans on people as they viewed campaign ads. The typical response to these studies at the time was that there's something creepy about political hacks using brain scans to make more effective ads. But to me -- having just written Mind Wide Open -- it seemed much more interesting, and much less creepy, if you looked at these results more as a political science experiment than an exercise in neuromarketing. Instead of scanning brains to devise more persuasive means of securing votes for Candidate X, scan brains to find an answer to that most mysterious of questions: how do political values form in the first place?

You can see how I ended up answering the question in the piece itself, but one thing is probably worth reiterating here: whatever conclusions we end up extracting for this line of research won't come exclusively from the neuroscientists. You'd need sociologists and political scientists and philosophers -- not to mention those political strategists -- to make sense of the results, to put them in context, and to propose new avenues of research. The neuroscientists would be mostly there to explain the results in terms of brain anatomy and function, leaving it to the social scientists to interpret the results on the level of human experience. In other words, the scans don't give you answers. They give you new kinds of questions.