A Conversation with Steven Johnson

by Jesse James Garrett
August 8, 2006

Bestselling author Steven Johnson’s keynote kicks off Day One of User Experience Week 2006, August 14-17 in Washington, D.C. In this conversation with Adaptive Path’s Jesse James Garrett, they talk about his writing career, the past, present and future of interactive media, and the role of experience designers in shaping our culture.

Jesse James Garrett: It’s been nine years since the publication of Interface Culture. In the world of technology, we tend to think of that as a long time. How much do you think has actually changed since then?

Steven Johnson: It seems like a longer stretch of time for me, actually, because I came up with the basic argument for Interface Culture in the summer of 1994, when I was still in grad school and before Stefanie Syman and I started FEED. I wrote a proposal for the book then, and shopped it around to a few publishing houses. But of course, this was before the Web had become a mainstream phenomenon, so I think people were somewhat baffled by the idea. I was a little baffled myself: The original proposal kept referring to this emerging new “medium” called Mosaic.

I think an immense amount has changed, and much of that change has been encouraging to see. The one constant online — which of course supports one of Interface Culture’s main arguments — is that linked text is still central to the medium and its interface innovations. When I was writing the book in 1996-97 there was a sense in the industry that the textual basis of the Web was just this accident of limited bandwidth and processing power, and as soon as those limits disappeared, the Web would become a multimedia experience, and hypertext would go the way of the command line. Obviously, there’s much more audio and video online today, but the medium — and its new interfaces — are still mostly about doing things with words.

JJG: Speaking of hypertext, in Interface Culture you spend some time analyzing the hypertext style of Suck, the daily essay site that for many people epitomized the irreverent voice of online writing in the ’90s. You credit much of Suck’s effectiveness to the site’s use of hypertext to suggest hidden connections rather than making those connections explicit. This is almost exactly the same argument you make years later in Everything Bad Is Good For You regarding television narrative in shows like The Sopranos. Do you think the ascending complexity of other media can be attributed to the influence of the Web in our culture, or are both symptoms of some deeper trend?

SJ: I think the ascending complexity that I called the Sleeper Curve is in part attributable to the influence of the Web, or at least to the rise of interactive media in general. All other things being equal, a person who is used to making active choices while consuming media is going to be more receptive to complex media, even if the medium is a passive one, like television. Also, the Web supports complexity because there are so many para-sites (as I called them in IC) annotating and explaining every little detail of The Sopranos or The Simpsons. I think Lost is a show that could only have been made in the Internet age, and from the way the creators have embraced the fan sites, it’s pretty clear that they feel the same way.

JJG: What interface approaches are you excited about these days? If you were writing Interface Culture today, what applications would you point to as examples of where the field of interface design could go?

SJ: Like most of the people reading this, I suspect, I’ve been incredibly energized by all the grassroots Web 2.0 applications that have exploded over the past few years, most of them descendants of Firefly in one way or another. (Someone — and come to think of it, it’s probably me — should go back and track all the core ingredients of today’s Web that were visible at Firefly circa 1996.) So my list is the usual suspects: Blogger, Six Apart, Flickr, Delicious, Technorati, 37signals, Flock, all the GMAP mashups, and so on.

JJG: It’s interesting that you mention Firefly, because I’ve often thought that the Web today might be a really different place — and that we might have reached this stage in its evolution much more quickly — if Firefly had remained an independent company rather than being swallowed and buried by Microsoft. Are there other examples that come to mind of early Web innovations that expired before fulfilling their potential?

SJ: It really is interesting about Firefly, isn’t it? I’m curious how you imagine that non-Microsoft scenario playing out. I was and am very good friends with some of the key people behind Firefly, and lived through a lot of that period side-by-side with them, since I was doing FEED at the same time. But I’ve never asked them if they think about alternate scenarios, and what they would have meant. It was just a little too early, I think: They were very Web 2.0 in the whole user experience — it was all architecture of participation/wisdom of crowds stuff. But at the same time, they had a more proprietary relationship to the data they were collecting. If they’d opened the system up — followed the del.icio.us model — it would have gone some other way, for sure. I’m just not sure what way it would have been.

As for the lost arts of the early Web, I wrote a piece a while back about the John Snow archive that’s been collated by a wonderful UCLA professor named Ralph Frerichs. I’ve spent a lot of time there because Snow is the hero of my new book: It’s filled with old maps, original documents, quotes, pictures, bibliographies, links. It’s a great resource, and it’s the kind of thing that academics first started doing in the early days, but then the whole genre kind of died out. I suppose this is the kind of thing that Wikipedia could eventually evolve into, but right now the experience is completely different. It’s the difference between an encyclopedia and Walter Benjamin’s Arcades Project…

Connecting virtual space to real space is clearly the most exciting area right now, though in a funny way I think we’re living through another phase of being too literal with our spatial metaphors — mapping tools are amazing, and incredibly useful for some applications, but they’re not always the best way to represent physical places. I’m actually working on a project that deals directly with some of these issues, and not to get too dramatic about it, but I’m hoping to be able to show some screens from it in D.C. this August.

JJG: One implicit aim of Interface Culture seems to have been to encourage people to view software as a cultural artifact, both influencing and influenced by cultural trends. To what extent do you think this view has taken hold?

SJ: I think it has completely taken hold, though people don’t always use the word interface when they express the view. Google is an interface for exploring the Web. Blogger (and Typepad, WordPress, et cetera) are interfaces for publishing to the Web. They are both runaway successes because their interfaces let you do something that previous interfaces weren’t very good at doing. I think it’s fair to say that Google and blogs are widely seen as influencing and being influenced by cultural trends.

The part where it gets a little trickier with Interface Culture is where I talk about the interface as a new art form. Certainly, there are more instances of genuine interface art out there today than there were when I was writing the book. (I think of the great stuff that Jonah Peretti and the Eyebeam folks do, or Rhizome over the years.) But I don’t think it’s quite right to say that interface design became an expressive artistic medium, in the way I described it. But hey, nine years isn’t that long. There’s still a lot of evolution ahead of us.

JJG: Do you still consider interfaces a means by which people understand the world? Do you see designers applying that perspective in the interfaces they create?

SJ: Clearly interfaces are tools for understanding the world. So many of the most interesting debates in the “new media” space revolve precisely around the question of how specific interfaces will shape the user’s view of the world. And those debates play back into the design decisions that shape the next generation of software.

A great example of this is the power law/long tail discussion that’s been percolating over the past few years (and is likely to hit the best-seller lists in the next few weeks, when Chris Anderson’s book comes out). Think back to the debate that swarmed around Clay Shirky’s “Weblogs, Power Laws, And Inequality.” That was effectively a debate about whether the combined interfaces of blogs, blogrolls, Google, and Technorati (not to mention the original linking interface of HTML) was creating a skewed perception of the entire community of blogging voices online, in that some A-listers had a thousand times the traffic of other, potentially equally worthy, blogs. (The “long tail” debate, not surprisingly, is a mirror image of this: emphasizing the army of Davids, and not the A-list Goliaths.)

But this wasn’t purely a theoretical debate. It involved some academics, but it also involved the people designing the interfaces in question. And it shaped a whole generation of tools aimed at amplifying those voices out on the tail. I just went back and tracked down my own blogged response to Clay and his critics / admirers, and what I wrote then is probably a better answer to the question than anything I could write now:

“The most interesting thing to me about Clay’s essay — and the subsequent response — is that the active participants in the power-law system are having a conversation about the distribution and what it means, and whether they want their little ecosystem to look like that.

Most systems that display this kind of behavior 1) don’t have component parts with that level of self-awareness, and 2) don’t have the opportunity to change the dynamics of the system if they choose. We hear a lot about architecture being destiny in the digital world, but the fact is that architecture has never been more flexible, and there have never been so many connected, smart people interested in flexing its joints for good causes. A few years ago, when I was writing in Emergence about the limitations of the one-way linking built into the Web, there were very few practical applications out there that attempted to remedy this flaw. Now the Web is teeming with them (Trackbacks, various Google hacks, Blogdex.) To a certain extent, the increased feedback of two-way linking may have amplified the scale-free phenomena that Clay describes. But the key point is that the one-way architecture isn’t necessarily our destiny anymore, partially because some very smart people started to think that two-way links would be better for the system as a whole, and they set out to add them to mix.”

JJG: In Interface Culture you hold high hopes for the possibilities of 3-D interfaces for information access and visualization. This hasn’t really materialized. Is the transition to 3-D interfaces just taking longer than you thought, or did something happen in the interim that you didn’t expect?

SJ: Do you think it reads like that? That’s interesting. I guess the way I would describe it is that I spend more time writing about 3-D information visualization interfaces than I would were I writing the book today, but almost all the examples I look at — Magic Cap, Bob, et cetera — are pretty much slammed. This is the whole section of the book that starts with the line: “The failings of the present day come from taking that [desktop] metaphor too literally.” Were I writing Interface Culture today, I probably wouldn’t have bothered talking about them at all. (At least as information visualization tools; games and online communities are another matter altogether.) So I suppose the attention was a form of “high hopes.” But to me, the chapters of IC where I’m really bullish on the future are the text, links, and agents chapters, which are not about 3-D at all.

JJG: Since the rise and fall of VRML, using 3-D for non-game applications has pretty much been considered a dead-end — at least until Second Life came along. Do you think the success of Second Life will lead designers to reconsider the possibilities for 3-D environments beyond gaming? Or is Second Life’s success attributable to some other aspect of its design?

SJ: That’s a great question, and I’m not sure I have an answer, except that I feel like something very interesting is happening over there. I suspect part of it stems from the way Second Lifers subtly integrate gaming elements, without making them explicit. Edward Castronova and Julian Dibbell figured this out before any of us: What’s addictive initially about these environments is scarcity — resources, land, objects, talents. Introduce scarcity, and game play naturally develops, as people compete to gather as many resources as they can, or invent new ways of sharing. (The real estate markets in Second Life are fascinating.) The genius of Second Life is to keep those game elements in the background; it’s a social system first, and a game second.

JJG: A lot of people have been asking me lately what I think about Nintendo’s upcoming Wii console — specifically the gestural controller designed to appeal to people who don’t currently play games. Nintendo’s strategy seems to run counter to the argument you make in Everything Bad Is Good For You that audiences are seeking ever more complex experiences. What do you think of Wii’s prospects?

SJ: It seems pretty smart to me. There’s such a huge gap in the gaming world: There are people who play a ton of games, and there are people who barely spend any time with them. It seems logical that that distribution should even out a little.

JJG: Nintendo says a key part of their strategy for Wii is the design of the controller. They’ve found that typical console controllers are intimidating to a non-gamer audience, so they deliberately made this one something you would hold like a remote control — in fact, they actually call it a “remote” rather than a controller. In other words, their hypothesis is that the interface is the obstacle to the growth of the gaming market. Meanwhile, interaction designers are constantly bemoaning the limitations of the hardware interfaces we design for. We’re dying to be able to support more complex interactions than point-and-click can get you. So you have two opposing trends: Nintendo saying the hardware is too complex, and designers saying it’s too simple. Who’s right?

SJ: I tend to fall in the camp that the hardware should be as simple as possible, and the complexity should all be on the screen. I’m not some kind of one-button-mouse purist or anything, but I certainly don’t want my mouse looking like my Xbox 360 controller. I loved those graphics showing Apple’s Front Row remote next to the Windows Media Center remotes, where the Apple remote has literally six buttons and the other remotes have forty. Granted, the Front Row software doesn’t do nearly as much as Media Center — you’re going to need numbers on the Front Row remote if they ever do a DVR — but still the idea is a profound one: people are much more likely to figure out what they need to do if their options are up on a big screen, and not on a tiny device in their hands.

JJG: One major part of the mass culture that is omitted from Everything Bad is music. Do you think music is immune to the drive toward complexity you see happening in other media? If not, how is that complexity manifesting?

SJ: I didn’t include music because enjoying or understanding music isn’t quite the same sort of problem-solving experience that you find in decoding a narrative or making sense of a video game. The question of why music is so powerful to the human brain is one of the great, enduring mysteries, but clearly it’s much more intimately connected to the brain’s emotional centers than the kind of abstract reasoning skills I was writing about in Everything Bad. So I decided to carve it out.

JJG: You probably could have devoted an entire chapter of Everything Bad solely to the television show “Lost,” which embodies many of the narrative principles you describe with its multi-layered storytelling that rewards ever-deeper scrutiny. The show’s producers have also been pushing boundaries by extending the show’s narrative into other media, with various fictional Web sites, a novel, and now an entire alternate-reality game, “The Lost Experience.” Is “Lost” establishing a new model for cross-media entertainment that is likely to persist, or is it an oddity that we shouldn’t expect to see replicated?

SJ: Oh, I think it’s certainly a sign of things to come. Interestingly, some of the techniques that you describe — the extensions into other media, integration with Web community and discussion, game spinoffs — are all techniques that have also been explored by the reality shows of late, which are of course the great example of formal complexity and innovation matched with largely trivial content.

JJG: The writer you seem to be compared to most often is Malcolm Gladwell. Do you see your work as covering similar territory as Gladwell’s, or do you think you’re pursuing different goals?

SJ: I’m a big fan of Malcolm’s, and so being compared to him is always flattering. I think because he’s had so much commercial success, and because he has such an easy, accessible voice on the page, people sometimes underestimate how hard it is to do what he does. I’m so glad he’s blogging now, because it’s great for him to have the vehicle — when the books or articles come out — to respond to the critics, and expand the ideas a little.

We have a lot of shared interests, both in topics — Emergence and The Tipping Point are variations on a theme, as are Blink and Mind Wide Open — and in approach. I suppose the most significant difference thus far is that I’ve been more focused on writing books, and he’s been more focused on doing magazine pieces. But that might change, for both of us, I imagine.

JJG: Back when you were running FEED, you seemed to be most interested in cultural criticism, but since then your work has taken a sharp turn into science journalism. What prompted this transition, and what do you see as the connection between these areas of interest?

SJ: The first step was that I looked up at my bookshelf one day and realized that the last 15 books I’d read had been science books. So I thought: if this is what I want to read, maybe it’s what I should write. And then I read E.O. Wilson’s Consilience and thought: I love every bit of this except for the part where he talks about culture. I thought: it would be nice to have someone who came out of a culture crit background who was genuinely building bridges to the science, and not deconstructing it. And really, ever since then, that’s what all my books have been trying to do, in their different ways — to write about culture in ways that are genuinely open to the insights of science, where they’re appropriate.

JJG: Your upcoming book The Ghost Map tells the story of a cholera outbreak in London in 1854. What drew you to this story?

SJ: I’d known about the story for many years, and I’d been looking for a way to write a book that would have a sustained historical narrative that I could use as a kind of scaffolding to support some of the ideas and themes I’m interested in. My other books have all been more-or-less pure idea books, with short narratives occasionally introduced almost as color. The story of John Snow and Henry Whitehead solving the mystery of this cholera epidemic — and, in Snow’s case, solving the mystery of cholera itself — struck me as an amazing intersection of many of the themes I’ve been following over the years: urban life, the connections between science and culture, information design, thinking across scales of experience, amateur expertise, and so on. And as you can tell from reading Interface Culture, my grad school years were spent studying the nineteenth-century Victorian novel, and so I knew the period quite well.

JJG: Edward Tufte has suggested that information design played a vital role in containing the outbreak. What do you think?

SJ: Tufte played a huge role in popularizing the story — to this day, most of the people I meet who are familiar with it read about it in Tufte first. He actually wrote about it twice, originally in The Visual Display of Quantitative Information, and then a longer section in Visual Explanations. His original assessment was factually wrong on a number of fronts — it greatly overstated the role of the map in solving the mystery of where the cholera was coming from, and the map itself that Tufte included was a heavily modified replica created for a 1912 textbook on public health. In the later book, he got the story right, though I think he’s a little too bullish on the map’s originality as a work of information design.

Basically, the argument that I make at the end of The Ghost Map is that the map was not a tool for solving or containing the outbreak, but rather a tool for explaining to others how the outbreak had come about — effectively, convincing them that the cholera was in the water, not the air, which was Snow’s great insight. But what made the map so original as a document is that it drew upon the local knowledge of the neighborhood itself: both Snow and his collaborator Henry Whitehead were residents of Soho, and had to do a great deal of shoe-leather detective work to figure out who had died, and who had consumed water from the contaminated pump. So the map is not just a triumph of information design; it’s also triumph of a certain mode of engaged, urban amateur expertise: local experts representing their own neighborhoods. There’s an immediate connection here to the whole Neo-Geo wave of Google map experimentation going on, which is one of the themes I discuss at the end of the book.

JJG: Is The Ghost Map an extension of your earlier work, or a departure?

SJ: It’s a departure in the sense it does tell a sustained story, and has — I hope — a page-turner quality to it. But thematically, it is very much an extension — mostly of Interface Culture and Emergence. I describe it sometimes as a kind of fractal way of writing history: it’s the story of one terrible week in 1854, but it’s an attempt at the same time to tell that story simultaneously on multiple scales of experience: from the bacteria itself, to the individual human lives of Snow and Whitehead and their nemeses, to the broader cultural trends of the time, all the way up to the super-organism of London itself. When I was a few chapters into writing it, I was describing it to my editor, Sean McDonald, and I was saying that it reminded me of Emergence, if Emergence had been a disease thriller. And Sean nodded, and said: “Right. It’s like Emergence if the slime mold had started killing people in Chapter Three.”

JJG: Thanks Steven, and we look forward to hearing your keynote at UX Week in D.C. on August 14!

Jesse James Garrett is President and a founder of Adaptive Path. He is the author of the widely-referenced book The Elements of User Experience. Jesse’s other essays include The Nine Pillars of Successful Web Teams and Six Design Lessons From the Apple Store.