Howard Dean - A Negative Scenario

Now that all the regulars have weighed in, let me throw a monkey wrench into the this emergent system. Thanks to Paul Boutin and Tom Morris for helping me focus my thoughts on this. Tom Morris is the Web master for Scientists for Dean. Obviously, a Dean supporter. But also a biologist with an interest in computer modeling. When I emailed him to as about scaling and security issues, he responded as an engineer: Hi Gary, thanks for the note. First, I just need to clarify one point with you. By, 'scaling,' do you mean...

1. The inherent behavior of a system (of concurrent systems) changes non-linearly as the system grows or shrinks.
2. The inherent behavior of a system of concurrent systems becomes unpredictable as the overall system grows or shrinks.
3. As some subsystems of a system of concurrent systems grow or shrink disproportionately to other subsystems in the system, the behavior of the overall system becomes unpredictable.
4. As a system of concurrent systems experiences a stream of diverse, dynamic circumstances, subsystems spontaneously sort into an identifiable hierarchy.
5. Nothing like that at all, Tom. Sheesh! Here is what I really mean...

Can you steer me into your particular area of concern? :-) I thought this mode of questioning was appropriate, and I tried to respond in spirit. In formulating my response, I managed to articulate why I thought admirers of Dean's Internet strategy might be disappointed in the coming months: Hi Tom, I like your clarification system! The areas I am interested in are: Area 1, Area 2&3 (which I would group together), and Area 4 to the extent that this is a feature of the growth of the Dean campaign. System 1 = The Internet component of the Dean campaign.
System 1a = Blog/email/Web component
System 1b =Meetup component
System 2 = Mainstream Media communication component
System 3 = Polling component
System 4 = Fundraising component
System 5 = Institutional component (Seeking/getting endorsements and support from well established collective actors - unions, DNC, etc.) Area 1 refers to the changes to be expected in the Internet component of the campaign as it grows. Imagine 1000 blogs instead of about 100 - 10,000 comments per day instead of 2000. Then double these numbers again. Does this product a qualitative change? Or does it simply get larger? Area 2&3 refers to the changes to be expected in the Internet component of the campaign (System 1) as the overall campaign grows. Growth of the campaign will also shift the relative importance of each component in relation to the others. This is a system whose components interact complexly. Growth in System 1 produces growth in System 2 and 4 in the early phase of the campaign; but spikes in System 2 have created spikes in System 1 since the beginning, also. Part of the logic of the system at first was that the energy contained in erratic spikes in System 2 could be used to power more steady and sustainable upward movement in System 1. But the power of a sustained burst in System 2 (which is almost certainly on the way), will change this relationship, driving System 4 and 5 and perhaps, if System 2 grows enough, making System 1 seem like a redundant component. The possible changes in System 1 produced by the growth of System 1 (more and more comments on more and more blogs, a fragmentation of attention, less personal investment in any one discussion space leading to increased vulnerability to trolls) combined with disproportionate growth of System 2, could lead to a completely new campaign dynamic than we have seen so far. In other words, the Internet phase is over. This is just a scenario, mind you. But this afternoon Paul Boutin called me and suggested some links: Andrew Orlowski's Story on Googlewashing Michael Wolff's Comparison of Dean's Internet to McGovern's Direct Mail And of course the Pew Internet Report: few people read blogs. Wolff's piece gleefully ignores the obvious differences between Internet organizing and direct mail, and he makes much of Dean's non-electability, which is clearly a weapon in the hands of interested parties, but his distinction between the logic of an early phase of an insurgent campaign and the logic of a national election is consonant with scenario I offered to Tom. This has not gotten much attention yet in Deanspace.

Howard Dean - Five Airports, One Twice

My trip to talk to Dean took my from SFO (my home town airport) to Denver International, then from Denver to a small private airfield called Centennial where the Governor's plane was sitting, then the Las Vegas on the campaign plane (I did the interview on the way), then from the small Las Vegas airport where we landed to the main Las Vegas airport, then back to SFO. Three flights, five airports (one twice) - it was a taste of the campaign life. I rode a limo to the Las Vegas airport with a pilot who had been ferrying them around for a few weeks. He was exhausted and eager to go home to his wife in Salt Lake City. I woke up this morning thinking - these guys do this for a full year and the day it's over one of them has to start work as the President? The interview is not off my tape recorder yet, but we had an interesting conversation. I'm not going to post a lot of it here - at least not until I have my own thinking more clear and finished the story for Wired - but I will say that I did not get the candidate to commit himself to overturning the DMCA. Sorry, everybody. Ditto on the Sonny Bono Copyright Term Extension Act. Sorry, again. Last night and this morning I've been trading emails with a frequent poster on the Dean blog known as the L1 Ranger. The questions I've been asking him (and others): how does this type of decentralized campaign protect itself from intentional disruption; also, what are the scaling problems? Yesterday, I wondered if I was looking at the climax of the 'utopian' phase of the Dean campaign, the moment of turning from underdog into front-runner. The blogs are happy now. The trolls are barely making their discordant voices heard. There have been no defections of disappointed ex-supporters, no sectarian split-offs. Lots of online communities have gone through this phase. And then there's another phase.

Dean Campaign Manifesto Discussion

The imaginary retroactive manifesto for the Dean campaign sparked an intelligent and wide ranging discussion that is growing quickly. I also got a few responses via email I want to highlight here. Steven Johnson, with whom I've had great conversations at about two year intervals since 1995, put his finger on a question about the Dean campaign that has been on my mind for several days. What is the role of the leader? In his book, Emergence, Steven talks about the anti-WTO protests in Seattle, writing that "there can be power and intelligence in a swarm, and if you'r trying to do battle against a distributed network like global capitalism, you're better off becoming a distributed network yourself." In his email, Steven questions how much the success of the Dean campaign has to do with Howard Dean. "First, there's no denying that the world needs leaders: not just because we need them sometimes to make executive decisions for us, but also because we have a hard time identifying with a swarm, for reasons both biological and cultural. The most powerful and progressive political system to date -- democracy -- is a beautiful mix of bottom-up and top-down, at least on paper. We choose our leaders from below, but then they get to decide things from above. "The problem is in practice, the bottom-up component to democracy tends to be driven by top-down forces: major corporate donations, mass advertising, etc. So the Dean campaign is absolutely right in its attempts to get a better balance here between the two forces. "Now, in the book I talked about how one of the remarkable things about the anti-WTO protests is that they didn't seem to have leaders, and that the iconography was different: swarms in the streets, not speakers in front of the Lincoln memorial, etc. I would argue that -- however appealing Dean is personally as a candidate, and I happen to think he's very appealing -- the primary story driving the explosion of interest in the campaign was its organizational structure: here's this little-known small-state candidate, and somehow they're raising all this money using the Net, and holding all these self-organizing meetups all over the country. (Oh, and incidentally, he's a good speaker.) That to me runs parallel to the Seattle protests: it's the swarm that drives the story, not the Queen Ant." Steven says this very well, and he is not the only person I've heard it from. Even some of the people close to campaign acknowledge that it was the process, not the policies or the personality of Dean that drew them to the campaign. David Weinberger, author of Small Pieces Loosely Joined, sent me an email taking up and arguing with Steven's original point, in yesterday's post, that the campaign could be taking advantage of collaborative tools to develops its policies, not merely its tactics, from the grassroots. "For me, Steve Johnson's criticism is central. Why would you need a leader if you're going to let the message emerge? But I suspect the problem with that approach is that you need a leader (at least so far) to enable a group to form. (Groups do form around issues - abolitionists, e.g. - but there the trickling up of ideas makes less sense since the idea is already in place ex hypothesis.) I suppose you could be a leader without ideas, and Ahnuld is pretty much an example of that, but outside of celebrity, what would be a qualification for leadership and what would cause the swarm to swarm around you? "I want to agree with Steve's criticism because I want to be as thoroughly grassroots populist as I can (just because it feels so good), but I'm having trouble with it. Sure, every campaign always wants fresh ideas to come up from the grassroots. But that's not what SJ's talking about. He wants position papers written on a wiki. I don't. I want comments on papers written by the campaign. Maybe I'm just too old-fashioned. "I do think SJ's missing part of the dynamic: ideas from the grassroots don't have to go back up to Central HQ to be adopted; the Dean campaign instead gives you the tools to instantiate your ideas without involving HQ. That's even more grassrooty and emergent than his wiki idea. "Anyway, I think SJ opens a fruitful kettle of fish (wait...can I go for the triple mixed metaphor?) that ought to be run up the flagpole (yes!)." I'll be offline most of tomorrow as I go to interview Dean for Wired. I welcome your suggestions for questions for him - I'll be reading comments tonight and early before I leave.

The Dean Campaign

I'm still working on my story about the Internet side of the Dean campaign for Wired. It is giving me information anxiety. There is too much going on. It is impossible to get a bird's eye view. This problem is inherent in the nature of the campaign. In September, according to the campaign headquarters, there were more than 1500 events. Reporters called HQ and asked which ones would be good to cover, and the staffers couldn't really tell them. They weren't processing all of these events and evaluating them according to some algorithm that would sort them hierarchically. You could easily find an event in your neighborhood, using a Dean tool called "Get Local." But you couldn't take a look at a list of all the events, indexed by relevance. These higher order functions have not yet evolved. I'm used to stories having an innate structure that allows you to start anywhere in the vicinity of the key actors and find your way in. You just follow the trails. When I'm talking to other reporters who are confused as to where to start their research, I always say: start anywhere. (It is too late to explain why this works in more detail, but anybody who has done research will probably know what I mean.) But the Dean campaign is different. Yes, it is centered on a single man - the candidate. But its activities are widely dispersed, control is decentralized, and many of the "happenings," for lack of a better word, seem to have equal weight. Even the aura of celebrity, which usually moves with the candidate and one or two others, such as a famous spouse or a telegenic campaign manager, is spread out in this campaign. As Mathew Gross told me tonight: "People are going to take time off work to see Zephyr Teachout on her tour! Zephyr Teachout, if you don't know, is the director of Internet organizing at the Dean campaign. She is taking a restored Airstream around the country in the kind of tour that candidates generally make. And Mathew is probably right about the crowds. Among certain types of Dean supporters, she's very famous. So, I've naturally decided to do what the campaign itself has done - that is, I'm making the network work for me. Below is a draft of an idea inspired by Rem Koolhaas's remarkable pseudo-history of Manhattan called Delirious New York. In his book, Koolhaas pretends that Manhattan was designed according to a theory of the modern city. The imagined manifesto gives Koolhaas a way to sketch a portrait of Manhattan as it actually exists, to take it seriously as manifestation of human creativity. His book is a just-so story, a fabricated history that explicates real forces. Here, I've offered a Retroactive Manifesto of the Dean Campaign. These are the rules that might have been posted on the wall of campaign manager Joe Trippi's office, if there were such a list of rules. I am looking for examples and counter-examples - confirmation and correction. Are these really the principles that underlay the architecture of the campaign? Are there concrete examples you can suggest? Is something here plainly wrong? Hack away. (Each of these rules is taken from the work of the writer whose name is parentheses. ) A RETROACTIVE MANIFESTO FOR THE DEAN CAMPAIGN ALLOW THE ENDS TO CONNECT (David Weinberger) This is the principle that resulted in the massive Meet-Up momentum, with 128,878 sign-ups as of tonight. What goes on at all the meet-ups? The Dean campaign has only the vaguest idea. "They are allowing the ends to connect without any centralized control from the campaign," says David. "The goal is not necessarily to have messages flowing up and down. Democracy is supposed to be about people talking with each other about what matters to them and then organizing to get the things they want. If all you have a is a TV set and a ballot box, that's a shadow of democracy." DON'T BUILD THE SYSTEM - GROW IT (Kevin Kelly) The Internet component of the Dean campaign has no blueprint, if a blueprint is taken to mean a set of plans that specifies the final structure. Instead it offers a constantly evolving layers of tools the provide a medium for network growth, along with a stream of encouragement (cycling in a feedback loop), that serves as a nutrient. As Kelly wrote in 1998: "The network economy favors assembling large organizations from many smaller ones that keep their autonomy within the large. Networks, too, need to be grown, rather than installed. They need to accumulate over time. To grow a large network, one needs to start with a small network that works, then add more sophisticated nodes and levels to it. Every successful large system was once a successful small system." SWARM AND SELF-ORGANIZE (Steven Johnson) "The needs of most progressive movements are uniquely suited to adaptive, self-organizing systems: both have a keen ear for collective wisdom; both are friendly to change. For any movmeent that aims to be truly global in scope, making it almost impossible to rely on centralize dpower, adaptive self-organization may well be the only road available." (From Emergence, 2001) When I talked to Steven on the phone about the Dean campaign he was admiring but critical. Here's part of what he said: "The Dean campaign is using lost of tools - they have a swarm like, ant-like structure. But there's something missing. All of these tools are about organizing people and organizing their money. Finding people and putting them together to support a candidate. What they haven't experimented with is how this could be used to generate ideas, to create emergent political values rather than just to organize support. Can you do grassroots organization of the message? Can you think of your ants not as thousands of people knocking on doors and stuffing envelopes but generating messages, and then creating a system to see which ones rise to the top? A campaign platform is the perfect place for Wiki style document creation. A set of filters could be used to see which contributions rise to the top. Say, the 100 posters to the blog with the most karma drop the first draft of the document and then people can post changes to it. I don't know if you'd just want to go ahead and publish something like that, but it is at least a great brainstorming method, better than just sitting down with your high priced political consultants and figuring out what's going to sell in Iowa." UNTETHER (Howard Rheingold) By permitting the action to drift away from the desktops at headquarters, the Dean campaign has taken advantage peer-production methods and open source models. Groups of supporters and volunteer workers are building tools that raise the value of the network. "Collective action does not just take place behind a desk," says Howard. "It gets to you where you are in real time." YOU'RE NOT A LEADER - YOU'RE A PLACE (Joi Ito) For people who support the Dean campaign, Dean himself is only part of the story - and sometimes a rather minor part. The larger part is the community itself. I asked Joi Ito today to tell me what he thought was the difference between an emergent and a political opportunist. Both respond to the emerging direction of the crowd. Both become the public face of a movement. What is the difference? Joi said: In an online community, "you can't really jump on or rally support with force. You're not a "leader." You're a place. You're like a park or a garden. If it's comfortable and cool, people are attracted. Deanspace is really about that. He represents a place for people to hang out. For instance, the NAN (Net advisory Net) of Dean, some of us don't even agree necessarily with his politics but we are his advisors because it's a great experiment and we're meeting cool people doing cool things. It's not really about Dean. It's about us. The good thing is that Dean listens. The key to leadership here is listening. The good thing about emergent systems is that you can hear what they are saying even though they involve millions of moving parts." MAKE THE NETWORK STUPID (David Isenberg) The Dean campaign is a network rather than an army - and that's its strength. But it's also a stupid network, and that's its other strength. "Stupid" is used in the technical sense defined by David S. Isenberg in his classic telephony paper, "The Rise of the Stupid Network." In this paper Isenberg advanced the principle that under conditions of uncertainty a network should not be optimized for some limited set of uses presumed to be definitive. Instead, the network should be as simple as possible, with advanced functionality (and intelligence) moved out to the ends of the network - to the users. "Whatever we discover to be the new Stupid Network value proposition, my working hypothesis is that it will be based on intelligent end user devices, intelligent customers, employees whose intelligence is valued as a corporate asset, and companies that can learn." (The Rise of the Stupid Network, 1997.) I got Isenberg on the phone today and talked to him about the Dean campaign as an implementation of a stupid network. Here's a little of what he said: "I'm struck by how different that is from the Karl Rove point of view, where reporters are directed to cover the four or five stories they've selected - go to the aircraft carrier, set up the cameras right here so Bush's face looks like another bust on Mount Rushmore, or whatever. For the first time in the information age we have tools appropriate for a real grass roots, bottom up campaign." "In the old telephone company, central planning was needed before the network could grow. You had to manage the scaling from the top down. This worked as long as growth was predictable. But the Internet was not predicted. It grew from the bottom, from interpersonal agreements among sysadmins at the edges, from a collection of networks, including small ISPs that were basically modem farms in somebody's garage. Having a network without a strong center allows massive scalability without central planning." "If you have a Karl Rove, you know exactly where events will happen, who has to be there. But if you are a Howard Dean, and you are willing to let things happen from the bottom up, you can scale without doing all that planning." STILL COMING (I HOPE): CLAY SHIRKY AND OTHERS

Howard Dean and Emergent Media

In the middle of research for a Wired story about Howard Dean and emergent media. The more I learn about the campaign, the more fascinating I find it. A recent front page story in the Wall Street Journal (reposted here) described the campaign's Internet success as "a story of desperation, risk and luck." But I wonder if this characterization is accurate. Sure, risk played a role, and luck, and perhaps desperation also. But there is something missing from this list. If risk, desperation, and luck were main fuel, any candidate could have had equal success - and still could. Perhaps this is true - but perhaps not. I called Joe Rospars today. He is a twenty-two year old who works full time on the Dean blog. A year ago he was living in Stockholm and posting to his blog, Not Geniuses. On a visit back to the states he caught up with the Dean campaign and was inspired to volunteer. Now, he sits at a desk outside campaign manager Joe Trippi's office and - among his other tasks - reads thousands of blog comments a day. I asked him the question I've been asking other people. Why Dean?

"People imagine we are a traditional campaign with an Internet component grafted on," he said. "Like we figured out a new cookie to serve at dinner."

This was the beginning of a long, interesting conversation. The essence of what I am hearing is that the means are the message - that people like Rospers are attracted to the Dean campaign because of the process of the campaign. Is this a trivial reason, like being attracted to a campaign because it has snazzy graphics? I'm not sure that it is. In an interview yesterday, David Weinberger put it this way:

"Democracy is supposed to be about people talking with one another about what matters to them and then organizing to get the things they want. If all you have a is a TV set and a ballot box, that's a shadow of democracy. He's unleashed a loyalty that's not so tightly bound to his positions as it is to the architecture of democracy."

Amazon's Search Engine - Alexandria II

Amazon made some news today. They've just launched a gigantic digital library containing more than 120,000 titles, with the goal of quickly incorporating a lot more. The interesting thing is that these are not ebooks. Yes, all the text is contained in a database. Yes, you can search it. Your search brings back a list of links with short excerpts, and the links take you to pictures of the book's pages. The pages are beautifully scanned and very readible. You can browse forward and back a couple of pages. But then... you need to buy the book. I've been following the development of the library for a couple of months - and wondering when it would launch. They surprised me by going live this morning - and they surprised my editors at Wired, too. My piece has been posted on Wired News, here. I'm having fun trying out the tool over at Amazon right now. [Alexandria II is what I first heard it called at Amazon. The official name is "Search Inside the Book."]

Social Networking and Howard Dean

For a story about Howard Dean tentatively scheduled for Wired's January issue, I've been collecting references to theories of the Web, social networking, emergent media and [insert favorite buzzword here]. These references are easy to find, of course - just follow the blog. Lacking an automated tool to compare blogrolls, I will be doing my own non-automated analysis of the most influential theorists and theories, trying to make sure that everybody's favorite is included. Of course, I'm open to suggestions...