November, 1999
Penn National Commission

"Cybercommunity"

Led by Larry Lessig (Harvard Law School) with Julian Dibbel, Ken Deutsch, and Marc Ewing.

Celebrating the Conversation: Public Discourse in Action Tape 3, Session 2

Mr. LARRY LESSICK: OK, we're going to start. Welcome to our afternoon session, which is on the subject of cyberspace communities. I'm Larry Lessick and, with most of you for the last three years, have been a member of this commission. And anxiously been waiting and eager to bring this topic to your attention.

We have in the room--I wanted to make sure everybody realizes--a class here. Mitchell Marcus and Dave Farber's class, which is doing the impact of information technology on society, is with us to participate in this conversation.

But we're going to have a conversation of the following form. I've got three people here, I'll introduce in a minute, who are going to tell us a relatively short version of their cut on the issue we're going to talk about, and then open it up for a conversation. And here's the way I want to introduce this conversation. I know, from watching the conversation we've had for the last three years, most of you are deeply skeptical that there's any reason to use the word community and cyberspace in the same sentence. And most of you believe that this is, you know, a lot of hype, bunch of kids--we got 20 more up in the back here--believe in. But, really, when you talk about what's essential, and the idea of community, it has nothing to do with cyberspace.

And I've been anxious about how to bring this topic to you because I'm not yet sure how I break you of that thought because, in fact, there is such a thing as communities exis--existing in cyberspace in all the relevant senses that you would identify as relevant for a community. There are places where people spend an extraordinary amount of time devoted to communal goods, to bringing about things that are of value to a collection of people, where they feel identified and feel like they have a role and responsibility in shaping that space. There are extraordinary collections of people who work in common on problems that they consider their problems and problems of society in general. There are these places you would recognize as communities.

But all of cyberspace is not that. Cyberspace is not automatically a community. And the theme that we want to bring out in this conversation is the extremely delicate relationship that exists between particular architectures or designs of cyberspace and the communities that they bring out, because some places in cyberspace will have no recognizable features that you would call a community. And, in fact, that's probably what you think of most of cyberspace: e-mail or the Web or chat places. These are not places we're going to try to convince you are community spaces. But there are other places that really do function as communities.

Dave Farber runs a discu--a--a--a mailing list that constitutes a group of people that really do function, in an important sense, as a community. And each of our three speakers today have a different type of community they want to talk about. So let me introduce them to you.

The first is author Julian Dibbell, who's written what is, in this world, a foundational text, very short, which makes it easy to be a foundational text, about cyberspace, this article which you have called A Rape in Cyberspace. And Julian's going to talk about a particular kind of community that exists in something called a MUD, and he will do a really good job of explaining what that is and how it works.

Marc Ewing is the chief technology officer of a company called Red Hat. Now Red Hat is a company which facilitates the--the distribution and advancement of an operating system called Linux. And Linux is an operating system that is the product of something called the open source software movement. And the open source software movement is one of the most dramatic instances of collective construction that we have ever seen. It is an extraordinary example of a kind of community mind that has been enabled to do something extraordinarily productive. And he's going to tell you a little bit about what made that possible, what made it work.

And Ken Deutsch might be a little bit of the dark side of the story we're going to tell here. Kenny is the first full-time political--Internet political cam--campaigning consultant. Since 1993, his job has been: How do we take the technology and architecture of cyberspace and use it for politics? And he's going to tell you a little bit about how the technology plugs into what we think of as more traditional efforts to mobilize communities or to facilitate interaction politics.

OK. So here's the way I want to work it. I want to mix--they're going to each talk for a relatively short time, I'm hoping no more than 15 minutes, and I'm going to make sure that they've said what I think it's important that they say. So I'll pretend to ask questions if they haven't really covered what I think you have to see. And then after the cycle of it, then we open it up for questions, and then--and then towards the end, we'd also like to invite people from the--students to participate in questions as well. So let me start with Julian.

Mr. JULIAN DIBBELL (Author): OK, yeah, as--as Larry said, I'm going to try to do a--a proof-of-concept or existence proof here of the phenomenon of community online. For those of you who are only familiar with--with e-mail or chat rooms or Web sites or other--the kinds of transient sites, locations on the Internet that have become to--have--that have come to predominate, the community I've written about, not just in this brief foundational text, but in a--in a book that emerged from it is a place called Lambda MOO.

And it's a MOO, M-O-O, and MOO is a kind of MUD, M-U-D, which stands for multi-user dungeon. And to give the very simple technical description of a MUD, it's a chat room with furniture, but that is--that is too simple of way of describing it. It is a text-based chat system. You log on and you--it is presented to you as an architectural space. So you log on and you see a description of, for instance, a lecture hall with whatever textual descriptions and elaborations the author wants to put in. And you will see a list of names also; so and so is here, Dufast is here, Joe Fleischman is here, Neil Smeltzer is here. And you'll see--and if you click on their names or if you type the commands to access those names, you'll see the descriptions that they have given themselves in text. So already, you have a rich sort of contextualization, you know, before you begin to interact with these people. And, you know, rich is relative here, but--and that's--that's not the end of it.

The fact is, in a place like Lambda MOO, there is an intensity of involvement that you just do not see in a chat room or in any kind of online site with pretense--any pretenses to community that I've ever seen. People will spend up to 30, 40, 50, and in ex--extreme and disturbing cases, 70 hours a week hanging out in this place. They will spend not only intense bursts of time over a short period, but there are people there who have been there by the time I was there, which was in 1993, '94, people who had been there for years and who are still there now, I guess. You know, people who have been coming back to this place regularly for five, six, seven years now.

Why do they do this? Well, one of the main things that involves people in Lambda MOO is that they are able to help create the architecture of the space. This is not a given space. It is given when you first arrive, but there is the possibility to add on to it. You can build your own room. You can build public rooms. You can program strange kinds of objects. The programming environment is entirely open to manipulation by the--the players, as they're called. So you can get involved on that level, you know, building cool toys that people want to use, building systems for new kinds of discourse, in fact. This is above all a discursive community and I'll get to that in a bit. You know, different kinds of ways of interacting with other people online can be built by anybody who comes along and wants to build them. And, of course, you can build new rooms and build--add on to the back yard of Lambda MOO, which now stretches on for miles all the way to the Pacific Ocean and has underground passageways and, you know, alternate realities built into it; tiny snow globes that if you shake them, you go in through the looking glass, and you know, appear in an alternate reality. So the environment itself is rich and gets richer all the time with all of the involvement.

But there's also even more--you know, there's even a greater gravity in the social interactions. People--people develop friendships and because there is a return, people keep coming back. These friendships deepen. They're not just, you know, the one-night chat room stands or, you know, the arguments that you get in with somebody on a--a mailing list. They're the full range of conversation that goes on in a friendship or an enmity, as the case may be, because not only is this a discursive community, it is a very political community as I will also get to in a bit. But there are--and there are also lots of ways to play with the kinds of social interactions that you have online because you can shift your gender, as it is presented in this environment, because you can adopt different personae. You can have kinds of interactions with people that are shaped by those--by those interventions on your part. So there is already, as there is in any--in--in a lot of MUDs, this broad range of behavioral possibilities that have to do with shaping the environment and have to do with the interactions that you have with other people on the level of friendship.

And then at Lambda MOO, what happened is there is a whole other arena in which to get involved and to keep you coming back and that is the political. And the way that that came about is the subject of the story that's in your packets, and then I'll just review here a little bit. The story is called, A Rape in Cyberspace, and it's about an incident that happened in April of 1993 at Lambda MOO in the living room of Lambda MOO.

Lambda MOO is described as a sprawling mansion in the hills of Palo Alto, and it's--it's gotten very sprawling over the years. But because architecture is really the governing metaphor here, the sense of community is very tied to architecture. And where the sense of community was greatest, at least at this period of time in the history of Lambda MOO, was the living room. It makes sense. The living room is where everybody hangs out. It's also where you first--when you first arrive in Lambda MOO, when you log on, you pop into the coat closet and you burst out of the coat closet and you're in the living room. And everybody's there and so there's usually a big party going on.

So it's--it's sort of they symbolic core of the community, which contextualizes what--what happened next, which was that a bunch of people, old timers from Lambda MOO, were hanging out in the living room and a character who called himself Mr. Bungle and described himself as an obese, dirty, slovenly clown of various evil intents, proved that he did, in fact, have evil intents by taking a devise that another programmer had written called a voodoo doll. And a voodoo doll is, in most context in Lambda MOO, it's sort of like a--a party trick, you know. It's something you use to make your friend appear as if they're doing something stupid, you know, because part of the convention of interaction in Lambda MOO is that every time you interact with somebody--you cannot--you see, you can't--you don't just speak with people as in a chat room, you can also do what is called emoting and that's--you--you perform actions. This is--and for those of you who have studied speech act theory, this is--this is wonderful, you know, rich territory to get into. You've got--it's--Lambda MOO is like nothing but performative utterance all over the place. And so you can interact with people either by saying, you know, `I say this,' you make a direct statement. So--so my lecture here, for instance, translated into the interface of the MOO would be--you will be seeing, `Julian Dibbell says, "My lecture here translates into the language of the MOO."' And then you would also see, `Julian Dibbell, you know, fiddles with his hands in a attempt to add expression to his statements.' So there's--there's--there's that two--those two levels of things. But--but what--what unites them both is that my name, my character name is always prepended to the action or the statement. And that's part of the conventions of the MOO is that whenever you see a statement that's--that begins with somebody's name, you know that that is something they have done.

And the voodoo doll is a trick that subverts this and somehow enables another person to make a statement that's prepended with your name, and there's a little tag at the end that says--and you hear so and so laugh--laughing evilly in the distance to let you know who actually played this trick on you. But it's actually--while in most contexts, it's used as kind of a joking game that you play on your friends, because of the nature of identity in the MOO, the way that identity is defined by this convention, it really kind of gets at you. It can kind of get--you feel as if someone has gotten under your skin and into your body.

And what this Mr. Bungle did with this voodoo doll was basically to enact what the community decided to call a rape. He had various characters in the living room performing violent sexual acts on themselves and on each other, and he kept at it. And they screamed at him. They told him to get out of here. Cut it out. He kept at it. They transported him, there are various ways to pick people up and throw them across the room and out of the--out of the building. They transported him out of the living room. He kept it up long distance. Somehow he had some way of making the voodoo doll work by a remote control. And eventually someone in ch--someone of a higher order--the wizards who are in charge of maintaining the--the database. The wizard came along and--and locked him up in a special cage that stopped him from doing this. And there was a great outrage in the community about this.

And very often when I have discussions about this story, as--as I have, It's been anthologized and--you know, so it becomes a thing that classes read and, you know, I'm often invited to--to talk to the classes. This is usually where the discussion stops because we get into: Well, was it really a rape and, you know, why didn't they just turn off their screens and, you know, why did they have to get outraged by it? And I think that it's--it's--it is a very interesting discussion, and there are very interesting things to say both critiquing the idea that they characterize this as a rape and responded to it at that level of gravity in some ways. And, on the other hand, very interesting cases to be made for why that language was appropriate; you know, why--how identity works in this setting and--and the extent to which that really was a violation.

But it makes me a little sad when the discussion stops there, because to me, what--what was even more interesting was what happened afterwards, which was that this discussion, this outrage, which started on one of the local mailing lists--not only is there direct interaction with people in Lambda MOO, but there are in-house mailing lists. So when you log on, there's--you'll see not only a description of the room, you'll also see, well, you've got mail. And you can check out, you know, what the discussion is on the mail--on the mailing lists. And in this forum, there was just an expression of outrage--you know, Mr. Bungle did this last night. I'm pissed off about it, so forth--that gradually evolved into a--you know, we put up with this sexual harassment online all the time, and especially in this setting, it's constant. What are we going to do about this? How are we going to respond to Mr. Bungle? How are we going to respond to people like Mr. Bungle?

And as the conversation evolved, it became clear that this was evolving into a kind of constitutional question because--footnote: What had happened in the evolution of Lambda MOO, as happens with the evolution of a lot of MUDs, is it starts out that the wizards set up this program for the use of the public. the wizards maintain the database, maintain it at the level of code. But also, as the population gets larger and conflict emerges within the population, `Oh, so and so stole my, you know, magic car that, you know, can drive you to the moon,' and they run and complain to the wizard about that. The wizards become involved in mediating these conflicts. And as things grew more and more complex, as Lambda MOO grew larger--and at the time I arrived, it was--there were about 4,000 people coming back to this place on a regular basis. It shot up eventually about 8,000, and then eventually there were population control measures implemented that brought it down to about a steady 5,000 and that's what it's at now.

So this is a vast community by--certainly by online standards and certainly for a handful of wizards to manage autocratically. And at a certain point, the wizard said, `Enough of this. We wash our hands of it. We will maintain the di--database, but you, the users, have to figure out how to manage yourselves socially.' And this was where the MOO was in its evolution when Mr. Bungle came along and when this debate emerged. And so they--there was no mechanism in place for them to handle this guy. And so it was this wide open question: How do we define ourselves as a community capable of responding to people like this? First of all, how do we deal with this guy? Do we kick him off the system? Do we chastise him? What do we do? But--but how do we implement the decision-making process that leads to whatever we do about him?

And there were all kinds of ideas. Well, we need to have a parliamentarian system. You know, we need to have--have an assembly. And we need judges and jails. And we--or we need, you know, come on, you know, we're--we're fine as an anarchy. This will work itself out. Or there was this sort of technolibertarian position very common of like, well, we can fix all these problems by just recoding the place. And there was, you know, the royalist position. Well, let's bring back the wizards because, you know, they were born to--to this position, and they--they--they're the ones who need to be strong and rule.

But, you know, as you all know, deliberative processes don't often lead anywhere. It sort of went around in circles, and ultimately, a wizard sort of had to do one of these leadership interventions and say, `OK. First of all'--one wizard said, `OK. First of all, Mr. Bungle is--is kicked off the system.' And then the arch wizard, Haakon, a guy named Pavel Curtis, who works at XEROX Park and--and had set this place up, said, `OK. Yes. And we need to implement a system--a pol--a political system that can handle things like these.' And what he instituted was a petition system whereby if anybody wants a s--change to the code implemented that is of a social nature--in other words, somebody wants to set up a system for arbitration or somebody wants to set up population control measures that will be built into the code, they can float a petition. The entire community can vote on it, and the wizards will implement whatever is decided.

And so just to round this out, I mean, I want to point out that this--this very--well, let me just say that then what I wrote about was the six months or so that followed the implementation of this political system, which, you know, was as rife with conflict and intrigue and the kinds of political narrative that we were, you know, talking about this morning in a very exciting bare-knuckle stuff, really rivaled in my study of other communities t--for--you know, looking for comparisons only by, you know, Athens in, you know, the high time of the--the agora in--in Athenean democracy. So I just want to say, `Look, there it is. There's your existence proof. This kind of thing that can exist online.'

But I want to add, as--as Larry said, this is not something you just bump into by logging on. This is a rare instance of a rare kind of rich community. And I just want to signal, you know, before closing, why I think it--it was able to achieve that kind of richness. And it has to do with the architecture of the place. In other words, the reason people are so invested in this place is because it is totally programmable by anybody. It is totally open to any player who wants to come in and add programming to it. There is a level at which, you know, the wizards are in control. But then finally, when a political system was implemented, the arch wizard had the wisdom to say, `Well, this political system has to work on the same model as the actual architecture of the code. It has to be open to anybody and, you know, we will just serve as--as a kind of technological extension. We will be the implementors of the code. We are not determiners of the direction or of the discourse. We are just going to implement.' So I'll leave it with those couple of notes.

Mr. LESSICK: OK. Now I think you--I think you did a great job in laying out the issues. Are there particular questions that people are stuck on, very briefly, before we contrast a different kind of community? Yeah.

Unidentified Panelist: Do the people who are in this process know who Mr. Bungle is?

Mr. DIBBELL: That's a good question. And there--there was a lot of debate about whether--you know, how relevant that was. Because the wizards were able to say, `Well we know'--because when you log on, you're anonymous to the rest of the community, but you have to re--you have to give your real e-mail address to the wizards. So the wizards knew what his e-mail address was, and they introduced into the discussion, `Well, do you want to know what his e-mail address is?' because then you could pursue him into real life. And, you know, either go to his systems administrator or, you know, make--go for--attempt a legal intervention.

But they decided, no, that his transgression--you know, it would be too weird to try to explain, `Well, you know, this guy, you know, in my toy--little toy world came and, you know, violated my toy character and, you know, he needs to be punished.' And so, you know, for that reason and because just the sense of poetic justice, I think they decided, no, the punishment has to be within this realm. And, in fact, one of the--the wizards did leak to one of the aggrieved parties the--the e-mail, and she did not pursue that. But--and--and it would have complicated had they even gone there. Because it turns out Mr. Bungle was actually a bunch of guys who shared the account and, you know, liability was sort of hard to pin down in real life.

Unidentified Panelist: Just a quick, quick follow-up. So--so when you put Mr. Bungle in a cage, are you putting the fictive persona Mr. Bungle in the cage or are you caging the e-mail address or the Internet address?

Mr. DIBBELL: You're caging--you're caging his character. You're caging the--the vehicle with which he moves around in the world. Unidentified Panelist: So he could come back as somebody else.

Mr. DIBBELL: In fact, he did come back as somebody else, you know, and there was--there was that kind of leakiness to the whole system, but there was also a sense of, well, justice here is also operating on a kind of symbolic level the way everything is here. So justice has been done. He's been toted, is the--is the phrase they use.

Mr. LESSICK: Neal.

NEAL: I'm a resident of Palo Alto and I'll tell you this is not cyber. This is Menlo Park politics.

Mr. DIBBELL: Who's Mr. Bungle then?

NEAL: There's a lot of them.

Mr. DIBBELL: No, in fact, I should point out that--that the people--the physic--the physical locations of--of the--the people in Lambda MOO stretched from, you know, from England to Australia with every place in between, with occasional interesting conflicts arising from that, but, in fact, since it sort of hovered within the Anglo-American world, you know, the cultural problems were not that drastic.

Ms. DEB RONALD: Do we know how many MUDs there are on the Internet now? Is there a way of tracking that or...

Mr. DIBBELL: There are people who track it. There are lists online. I mean, it...

Ms. RONALD: Six hundred?

Mr. DIBBELL: ...it fluctuates between--it's--it's never more than 1,000, and most of them, also--I mean, not only is this kind of community rare online, it's rare within the world of MUDs. Most MUDs are--are really strictly fantasy role playing games, and so they're much more sort of scripted in a way. And it's not to say that interesting social formations don't happen on them, but...

Mr. LESSICK: Right. So the story you have in your--in your packet is this account of the rape of cyberspace and then--in cyberspace and then the construction of the democracy. His book, "My Tiny Life"--That's the title, right?--is really a wonderful account of a much richer set of life that gets constituted here; all the community, as well as individual. Yeah, Michael.

MICHAEL: I guess I missed the--what--this is--this is kind of a naive question in light of the talk. But what's the difference exactly between this community and a fa--and what you just called a fantasy role pa--playing world?

Mr. DIBBELL: Well, it's just th--that there is an...

MICHAEL: This is a fantasy world...

Mr. DIBBELL: ...element of fantasy and role play involved, but it's--it's the--the--the knobs on that part of it are turned way down and the--and the knobs on, you know, actual social interaction not mediated so much by the--I mean, a fantasy role playing game has--has rules, to a certain extent, you know, and you can advance through certain levels by engaging in certain scripted, you know, battles with dragons and--and--and ax-wielding dwarfs and finding treasure and all this. A place like Lambda MOO is descended from that kind of world, but it's--it's--it's--it branches off from that evolutionary tree.

At the moment where people started, you know, after the battles with the dragons, started hanging out and talking about their problems and, `Oh, I don't like the way the dungeon master's running this place. Let's try to change that.' You know, there were kinds of social formations going on around the edges of these games that people eventually decided, well, let's just make that the main point of--of the game. And also, this--the building of--of--of the actual architecture is--is not something you're not permitted to do in a role playing game. You have to fight your way up through the hierarchy until you become a wizard yourself, and then the arch wizard may grant it to you to--to build your own realm, whereas Lambda MOO is--is much more democratically constructed that way.

Mr. LESSICK: OK. We're going to shift. Unless this is really quick because we're...

Unidentified Panelist: Very quick question. Do you or anyone else have enough information about the people's real lives as to how they have 30 or 50 hours a week to spend doing this?

Mr. DIBBELL: Yeah. A lot of it is--I mean, when I say 30 or 50 hours, they're logged in 30 or 50 hours. That doesn't mean they're interacting all of that time. They're--but they're open to that space. It's a window on their computer. They may be actually working in the Excel window or whatever, you know. And obviously certain kinds of jobs are more amenable to this or non-jobs, as the case may be.

Mr. LESSICK: There's a great book by Sherry Turkle called "Life on the Screen," which is an effort to understand these people and their relationship to this. There's a great passage where one guy's talk--describes living four characters at the same time. So he has four different screens open and he says, `Real life is just like a fifth screen and it's not usually the most interesting one.'

OK. We're going to talk--shift now to Marc Ewing. So I described a little bit about Linux. I can tell you this: There have been no leak to documents for Microsoft that describes the d--justices--the Justice Department's antitrust case as a threat to Microsoft. But there is a leaked document from the Justice Depart--from Microsoft that describes what Marc Ewing's part of the world is doing as the single greatest threat to Microsoft's dominance in the--in the operating system market, and that's the open source software movement, and Linux in particular. And that's come from a community. That's not come from Marc--Marc's company. It's come from thousands of people across the world who have built this voluntarily. No--you know, very little--there's not compensation that builds this, so that's the introduction here.

Mr. MARC EWING (Chief Technology Officer, Red Hat): Right. So I'd like to talk about this open source software development community. It's another online community. And as you described this, I--I notice a lot of interesting parallels, but--and I'm sure you all picked them up as well. And I think in--in Larry's earlier remarks, he--he sort of alluded to the potential importance of open source software and the products that it's creating. And just to sort of back it up before I--back up that statement before I jump into some more explanations so maybe you actually believe me.

There are a couple of products, some of which you may be familiar with or have seen in--in the papers. One called Linux, which is a new operating system that even Microsoft's produced at its trial as a competitor, some--somewhat insincerely probably. But it--it could get to that point. There are a few other products which are interesting, which are also open source software products; one called Apache, which I believe the recent Netcraft survey numbers have proven--have shown that 56 percent of the Internet Web sites today are running on Apache. So--so literally, when you go out and--and run around the Web, more than--than every other site that you visit is actually running on open source software. An even more startling number: There's a--a software package called sendmail which is a--called a mail transfer agent or it's basically like a server program. The--it's the--it's the software that delivers your mail from one machine to another around the Internet, your e-mail. And an astounding 75 percent of all the e-mail on the Internet is delivered via sendmail. So--so you're already using open source software in a huge way, though probably don't even know it.

So what it--I should describe what open source actually is. There are--open source is not a particular type of software. It's not necessarily a development community, but it--what it started from is a licensing model. And the license, unlike traditional proprietary software models that you--you might get from--from Microsoft where you get what's called a shrink wrap license that allows you to sort of borrow the software as long as they want you to use it. An open source software license allows you to use the software in any way you see fit, allows you to distribute it to anybody you want at no cost, allows you to modify it. And--and part of the open source software license also stipulates that you get the source code, if you want to, to your application. And the source code is sort of the human readable instructions which you can use to modify the operation of your program. So you can actually get that source code and modify it. And you can distribute your modifications as well.

There's one key--there's one final element which says that--and there aren't--there's actually sort of a spectrum of open source software licenses, but there's--there's one key one which is called the GPL which is sort of the founding father of--of open source software licenses. And it has a term in it which says that you can't--you can't restrict the recipient of the software. If you give the software to them under the GPL, if you can--and--or if you pass on software that you received under the GPL, you can't pass it along at anything less than the GPL's permissiveness that--that you got it under.

Mr. LESSICK: So it has to be as open as...

Mr. EWING: So is has to be as open so it perpetuates. And you can't eventually get all your software to be "closed," quote, unquote. So that's--in a nutshell, that's what open source software is. What it allows you to do then as a software consumer is to get software from either off the Net or from a friend or from a vendor who might be distributing open source software, tailor it to your needs, if you need to, if the ne--a slight modification or a bug fix or an improvement that you want to make, you can make it and you can distribute it around your company, to your friends. You can distribute it to--to your business partners literally at no cost to you. There's no--no cost that you would have to direct back to the--the initial vendor or creator of the software, so that describes the license of the software.

And what has sort of built up around--around this license and around the Internet is an open source software development community. And it's literally a community of--of hundreds of thousands of--of developers and participators in the development of this. They may not actually be writing code, but they may be involved in testing documentation or in--you know, in artwork, graphic arts. And this community is literally worldwide, and one of the--the great--one of the great products of it has been this Linux operating system, and it came about in the early '90s, just started by a guy named Linus Torvalds while he was at school in Helsinki as part of an operating systems course he was taking. He started tinkering with a very tiny little kernel of an operating system that just--it did very little except for throw A's and B's at the screen. But then he posted it to the Net under one of these permissive, sort of open sour--open source, the term had not been invented at the time--but a very permissive license, sort of a do whatever you want, and if you got anything to add to it, go ahead and do it and send it back. He posted it to that and said, `Anybody want a, you know--if you have any interest in this, it's probably not going to very exciting but--but you might want to check it out.'

And slowly, but surely--act--and actually, not that slowly. I mean, the course of six years, it's grown to the size it has to be a sort of very capable operating system that competes effectively with other UNIX operating systems and--and other server operating systems. But it did--it was sort of a g--gradual growth. You know, at first, a few hundred people downloaded it and--and made some improvements, sent them back to Linus, you know, here's a patch, here's some more code that you can add to your--your Linux product to--to make it better or work on a wider range of hardware or to fix a particular bug. And o--over time, this grew to be a--to thousands of people have contributed to the Linux operating system.

Now that's one particular way. There--there--there's a number of ways in which open source software can actually be sort of coordinated over the Net. As you can imagine, one of the--the criticisms of open source software is there's no control. It's--the quality is--is poor because nobody is necessarily watching what's going on. How do you contain--how do you--how do you sort of funnel all this kind of bubbling, you know, of energy around the--around the Internet into a particular product that's packaged nicely and that people can use?

And true, that there's--there's a number of ways that that can actually happen. In the case of the Linux operating system, we have what--what we in the community call the benevolent dictator of Linus. And what--he makes the actual final say on everything that goes into--into the soft--into the source code. That's not to say that he actually writes it all, because at this point, he's written less than--by his own account, he's written less than 5 percent of all the code that's in the Linux operating system. But it is to say that he either approves or understands the pieces of code that go in there or has delegated authorities to approve pieces of certain segments of the code to, sort of, his--his top lieutenants, of which there are maybe a half dozen. And they're responsible for certain segments of the code, say, networking or file systems or graphics, things like that. And--and--and--and among them, they make all sort of decisions as to what pieces--what patches go into the code. You know, somebody sends a--sends a--a--a patch on. It's evaluated, and it might be critiqued. It might be sent back to the original author before it gets included. But between them, they--they make the decisions.

Now if we go to Apache, which is again the Web server software that I mentioned, it operates slightly differently. There's actually a group of a--of Apache sort of core team leaders and there are--I don't know how many there are. There may be a dozen of them. But actually make--they actually vote on code before it goes into the kernel--the kernel of the--of the Apache product. So it operates slightly differently, and there's a whole range of--of different ways in--in which open source software projects can be--can be led and managed.

But--but common to all of them is the--the--the kind of requirement or reliance upon the sort of frictionless communication medium of the Internet and the frictionless way in which new pieces of code can be added to the source code and removed. And this is something that doesn't happen in proprietary software. You can't just, as a user, you know, rip out the networking code out of Microsoft Windows and plug in your own version of the networking code and see how that goes. And this frictionless ri--leads to the development of--of almost Ameritocracy of--of the people who contribute to this code base.

You don't use a particular person's piece of networking code because he's the guy at the company who wrote the code who was designated as the networking guy. You use a piece of networking code because it has been proven by the community, or generally accepted by the community that that's the best networking code in the world that has been written for this particular--for this application. And so the result is--is sort of a bubbling up of all the--the--the best components of all this developed software--of this open source developed software around the Net.

And you get to kind of skim the cream and produce this--a wonderful piece of technology at--at very little sort of--of direct expense to anyone. This is, you know, mostly volunteer time, side time. It's time contributed by people at--at universities, at research organizations. More and more, we're seeing now that time and--and--and technology is being contributed by--by commercial enterprises like Intel and Dell and Compaq and so forth because they're starting to see the benefits of this open source model and they're wanting to figure out how they can participate and help out and strengthen that model so that, you know, the first six years of--of--of this, you know, open source operating system development are--are maintained into the next six years, into the next six years.

So it's a--I guess that's the--the fundamental description of the--of the--of the community, and I think that you--you probably notice a number of kind of similar parallels to--to the Lambda MOO in the way this operates and the way people can contribute and build their own parts of this sort of living code base and--and not--and not--well, anyway.

Mr. LESSICK: OK. So a--so right now, you can go out into the World Wide Web and download the source code for the Linux operating system, all of it.

Mr. EWING: Right.

Mr. LESSICK: And if you wanted to hack around with it and decided you wanted to make a better graphical user interface or something, you have the right to improve it. And many thousands of people have done that and then taken their improvements and given it back to the Linux community. It's a kind of gift culture you're describing where people have devoted extraordinary amounts of time improving the source code and then offering it up for others to take. And Linus Torvalds and the committee, at least in the Linux context, decides what's the best that's been contributed and actually incorporates it into this code. And this is made possible because of the communication structure that you're describing, right?

Mr. EWING: Right. Right.

Mr. LESSICK: And also possible because of the law in the sense that the license makes it impossible for you to steal what's public and turn it into something private. And some people talk about the viral effect of open source software and that is: If you have a little bit of GPL or a code that's produced under this public license and it gets into your coding project, it can have the effect of making your software unprotectable anymore because you basically promised not to protect it, so it could have the effect of opening up lots of other software.

Mr. EWING: It's certainly a fear of some companies.

Mr. LESSICK: Some companies. Right. It's the hope of other companies. OK. Now, again, just clarifying questions about this structure that's been described here. Yeah.

Unidentified Panelist: Red--Red Hat, I believe, is a public company now, so what is the profit--where is the profit coming from if you're just giving everything away? From what I understand, it's your--you--you have a l--you have a support model. That's where you're going to make your money, but in seeing as you're now becoming a for-profit company, where--why aren't we--why--why couldn't we be assured that you're not going to--even though it's open source, only make sure that the open source bits that you want will get into the final version of Linux because you've approved it as opposed to letting other innovations that might, you know, work on the sides improve it.

Mr. EWING: Right. Well, first I think I should say this is sort of a Red Hat directed question, so I'll try to answer it quickly. But Red Hat has always been, you know, for five years, a for-profit company and, in fact, we're--we're only just now starting to experience real losses. But--yeah. But--but the simple answer to your question is that Red Hat d--does not control this community. We don't decide what goes into the Linux kernel. We don't decide what goes into Apache or sendmail or anything else. These are really community develop--developed projects.

Now what Red Hat adds to the mixture is--is--from our customers' perspective, it's difficult to take advantage of these hundreds of different development communities around the Internet that are developing these great little open source projects. So there's the Linux kernel. There's sendmail. There's Apache. There's a whole bunch of other ones out there. And by themselves, none of them are a complete solution. They're not an operating system. Before companies like Red Hat came along, sort of individuals and hobbyists would try to put things together and--and for their own means and maybe for their friends and for a smaller community. But trying to bring this to a much wider in--base of--of users is what companies like Red Hat are all about. And what we do is, you know, we take and coordinate with all these--these development communities; take their code and integrate it and test it and make sure it all works well together, package it up nicely so it becomes a much easier solution for you to either download Red Hat--Red Hat Linux off the Net or go buy it for whatever than it is to put it together yourself.

Mr. LESSICK: Jay?

JAY: Marc, would you explain the motivations of the people who contribute to something like open source code...

Mr. EWING: Right.

JAY: ...especially the desire to win respect among peers, notoriety for solving technical problems, to make a contribution to others would use extensively and to otherwise distinguish yourself as a creator of something the community is making together? Because I think that's an important part of the structure.

Mr. EWING: Yeah. I mean, I--I think that that's exactly it. That's--that's one of the key motivators for--for any of these developers is that if you write a chunk of code and five people use it, that's not nearly as exciting for you as if you write a chunk of code and five million people use it. It--it--it's much more sort of--you f--not only do feel just like, oh, you--you've given back to the community be--because of all the stuff you--you know, you feel like you should have done because you've been using this code for so long. But it's just a great sort of personal pride in the fa--in the fact that the community has accepted all this code and decided and said--and by using it, saying that you're--you're a--you're a good hacker, you've written a great piece of useful code that I can do. But--and that--and that's--that's sort of the way it's been for--for many, many years. It's actually starting to change a little bit.

The interesting thing about software is that most software is actually not written for sale. Most software is not written inside of Microsoft to sell or--or--or wherever. It's written inside of corporations or other organizations to solve their own kind of internal need. And what you're seeing now--and this is certainly how sendmail came about. This is how Apache came about. And this is sort of how Linux came about, but more on a personal level for Linux. And--and what open source does is it's--it's sort of this--it's a community that allows all these people who have been, up until this point, developing their software--they didn't have really no proprietary interest in. They don't particularly care and to--to maintain it necessarily or to keep it away from other people, but they have had to go through the process of actually, you know, spending time and develop it.

But with open source, now all these people are getting together and say, well, hey, I've got this piece; hey, I've written this piece; or hey, I've written this piece, and I can help you with that one. And they all get together and put it together so it's not a matter of--of people necessarily volunteering huge amounts of time, although there is that component that, you know, for the pride aspect of developing the code that a lot of people use, but rather a sort of realization and recognition of this--this already enormous pool of software out there that isn't really encumbered by--by, you know, restrictive licensing and trying to leverage that and turn it into something much greater.

Mr. LESSICK: Julian.

Mr. DIBBELL: Well--it's--my question is sort of a follow-up to that one. What is f--funding these people? I mean, are they doing it--if they have to solve a particular problem for their workplace, are they getting paid by their employers to write a little bit of open source software?

Mr. EWING: Right. That's it's--for example, that's exactly how Apache came about; is there were some people at--I think it was at Wired magazine or wired.com or whatever their--HotWired or whatever their online service was and they were using a particular--it's the NCSA mosaic server or whatever it was called and it was--wasn't doing what they needed to do. So they started this Apache project and it was funded by the Wired guys. And this was the--the IS guys, the system administrators, the Web masters at that company who made the modifications and, you know, what do they care? They--they--they--they gave it back to the Net and then enlisted the help from other people at other ISPs or other, you know, Web properties who also had the same interests. And now more of what--you're seeing similar things like the--the Dells and the IBMs and the Intels of the world are also starting to contribute their own technology that they've developed. So more and more, a lot of the technology is being written inside of corporations where the people who are writing it have been paid to do it. But that's only--that's only just starting to become the case over the past three years.

Mr. LESSICK: OK. But th--but one common feature here that I think is important answering, again, Jay's point; one of the attra--the attractiveness in both of these contexts is this opportunity to be constructing something, to be building something in a communal way...

Mr. DIBBELL: Right.

Mr. LESSICK: ...and--and this is a feature that explains a lot of the excitement of--of this; a little bit different maybe from the third perspective from our political consultant from the Internet. So why don't you introduce real-world politics dimension to this community.

Mr. KEN DEUTSCH (Political Consultant, Internet): OK. And I'll just start by saying I--I disagree with the dark side comment. It is the first time it's ever been tried, having been someone who has spent much of my career working for Ralph Nader now to be called the dark side. But at the same time, I sort of take it as a compliment so we must be effecting things.

It--it's very easy for the Internet to become a very negative impact in politics or it's--or it's also very easy for it to become to add something to the political discourse and to add something to communities who want to push particular issues. On the negative side, it would be very easy to go on the Internet, assume false identity, put things out there that are total misinformation to deceive people without people being able to see who you are. And, unfortunately, there's a lot of people on the Internet who would believe it, so that's one of the problems that's out there right now on the Internet.

In terms of the issues of community, in order to--for the Internet to be a tool to actually improve political discourse in the community, the key ingredient that makes it work is its interaction, the online community with the offline community, and the two things working together as one. I'm going to give you a couple of examples of things that aren't doing that and things that are, so hopefully, that will give it some context.

Some people may have seen in the news in the last few days Dick Morris' new project, Vote.com, which is a new project launched about a week ago. And on the site, anyone can go to the site and they can vote on an issue. And what the site then does--it then pops up a little message saying, `If you give us your e-mail message--your e-mail address, we will send the message to your elected officials.' Well, it so happens that it doesn't check the e-mail address; it allows you to vote multiple times. If you keep entering different e-mail addresses, you can put other people's e-mail addresses in, and you don't see what elected officials it goes to and you--and at the same time, the message that goes out, you have no idea what that message actually says.

You just vote on something and move on. You come back the next day and that issue's moved aside and there's another issue there. So instead of building community, it's a one-way communication. You come in, you act on an issue, it pushes something to an elected official and it moves on. So you keep coming back and it doesn't do anything. And I think that talks a little bit about Larry's comment about e-mail and the Web not necessarily being something that develops community. Where it develops community is where the two things work together in a cyclical fashion to--to build support and to build interaction.

Similar things happen with just using e-mail rather than just using the Web. Many of you may have seen a message over the Internet over the last few years about PBS funding. Has anyone gotten that message? I got the message twice in the last couple of months, and one of the versions was sent to me by a relative who I won't mention who said, `Oh, this is really important. If you care about this,' you know. So I went down and I read, and first, I noticed the bill number that's not a real bill number and then I noticed--and this was just about two months ago. Then I noticed that the message then talked about how Newt Gingrich was in the process of doing this. And what happens on these issues--and these are often known as flash campaigns--is they develop a life of their own. Someone throws something out on the Internet, and somewhere it started with a kernel of truth, and because of the nature of the Internet, it takes a life of its--of its own on. So one person passes it on to someone else, and finally it gets to a point--this is just living in e-mail; there's no Web site, there's no reference, there's no source, and as newer Internet users who haven't been--this hasn't happened to before tend to believe it, because it was sent to them by a friend who believed it, and they send it on and it continues to live.

An--another similar one actually has to do with a modem tax proposal that happened about four years ago where there was actually a proposal in front of the FCC at that time and there was actually an issue. A year later, someone took the same message, changed the year from, I think it was, '96 to '97 or '97 to '98, don't--lost track of a little bit of time here, but they just changed it and recirculated it, and all of a sudden, the FCC started getting all these messages, and members of Congress started getting all these messages. In that particular case, there was monied interests on both sides, and it may very well have been done on purpose by a special interest trying to push a ce--a certain--a certain side of the issue. So that's very `one way' and it doesn't really build community because it doesn't bring people back around.

I'll give you an example of a project we're working on, which--because it's going on today and it's for a political client. I can't me--mention who the group is, but we have a client we're working with that has a group of people who agree with them on their issue. And this week, as the budget's being debated, they found out on Friday that a certain senator was going to pu--try to push something into the budget that was going to affect their constituents. Now they had already set up a Web site. They had already identified who it was that agrees with them that--as part of their community. These are people who they've worked with on this issue offline. They recruited them to come online.

And when they came online, they identified their actual physical state so they knew--physical state as in the state they lived in, so they--they were able to identify from the people's address what--who their members of Congress were, stored that information, so when this happened this morning, an e-mail message went out to all the people whose profile said, `Yes, I agree on this issue, and yes, I live in this district,' and they were asked to come back to the site. When they come back to the site this afternoon and this evening, because this is going to happen immediately, they're going to be asked to make a phone call. And so it's taking online action and online community and spinning it back offline where it started. And so it's a cyclical effort of taking the online activity, building activity offline, building activity online and building a closer relationship between the organization that represents that interest and the people who benefit from that interest.

What's actually happening is they press on something on their screen, their physical phone line will ring, they press one, and it'll patch them through to the right targeted member's office. Then they can re--type into the screen what was told to them, and the lobbyist representing that interest can then see that information in a real-time database. And if they're hearing something different from the lobbying or from the messages, they can then change the message that the next people see who will continue to make the phone call. So it--it continues to evolve based on what's happening both online and offline.

Mr. LESSICK: So you--you've done this technol--this technique before in other contexts?

Mr. DEUTSCH: Yes.

Mr. LESSICK: For other campaigns?

Mr. DEUTSCH: Yeah.

Mr. LESSICK: And--and so what you're describing is an automatic--automated way to identify a community and get them to engage in political action in a real-time fashion, right?

Mr. DEUTSCH: Correct.

Mr. LESSICK: And so--again, to make sure you get this picture, you touch something on the screen, your telephone rings because they have your telephone number...

Mr. DEUTSCH: Right.

Mr. LESSICK: ...you pick it up and then your telephone is connected to a Capitol Hill telephone...

Mr. DEUTSCH: Correct.

Mr. LESSICK: ...that says, `Congressman so and so's office,' and then you read what's on the screen that you're supposed to say to that Congress person, right?

Mr. DEUTSCH: Correct.

Mr. LESSICK: And so you say, `So Congress person, I was just--happened to be calling on this particular issue, and what do you think about Bill X?' And then you report this information back.

Mr. DEUTSCH: Yes. And--and the--what--what happens is that the congressional offices do not see that it's coming from the Internet necessarily. And unlike traditional ways of mobilizing people, you--in typical grassroots, it's working on big ticket items, you know, we support this or we're opposed to that. Now what's able to happen is that grassroots politics, because of the speed of the Internet, can respond to things that grassroots were never used before.

Mr. LESSICK: So--so what's the numbers? How many people are we talking about that you're able to mobil--mobilize with?

Mr. DEUTSCH: You--you know, it depends on the issue. There are some issues that are hot-button issues that have a constituency offline, that the group existed beforehand. There are other organizations that start on the Internet and the community starts on the Internet and it moves off the Internet. One example of that is a project we worked on called gunfreekids.org, which started in late--I guess, in June when the gun issue came up in Congress, and it has been building about a thousand people a week. It's about 16,000 people now. There's no--there's just starting to be an offline organization. It started as just Internet users then telling other Internet users and people identified with the organization.

And so when we went to see where people were coming from--you can actually on Internet sites look at logs of where people--how they followed to get to that site, known as referrers. And what we found was that people were posting things to the Internet on completely unrelated issues and they so identified with this organization that they put in their signature line, you know, `I'm a member of gunfreekids, visit it at,' and the postings are on completely unrelated issues, but the people are identifying themself with the organization and, in essence, that's what's creating the organization is that actual dynamic.

Mr. LESSICK: OK. But then again, still broad numbers. What's the size that we're talking about?

Mr. DEUTSCH: Broad nu--how many people have taken political action on the Internet?

Mr. LESSICK: Well, how many people...

Mr. DEUTSCH: Or the different cases?

Mr. LESSICK: When you--when you try to brag to a client, you say...

Mr. DEUTSCH: Right.

Mr. LESSICK: ...in case X, we turned out...

Mr. DEUTSCH: Well, on that particular case, there's been 16--gunfreekids, there's 16,000 involved people that have taken at least one action. At least a third of them have taken three or more actions.

Mr. LESSICK: Is this the biggest you've done or is it...

Mr. DEUTSCH: It's--it--it--it really varies on audiences. I think from--that's the biggest, in terms of creating something out of nothing. We've also done some work for organizations that already exist, that already had a base, as well as a lot of work for corporate intranets where people are trying to mobilize their employees and can easily get, you know, thousands of people to participate because you're dealing with employees sitting at a desktop computer. You know the environment they're at so it's very easy to mobilize them.

Mr. LESSICK: Joyce?

JOYCE: ...(Unintelligible) in politics. I am in west Los Angeles, and in the 310 area code, the telephone is critical to this story, and we had introduced to us an overlay system in which we had to dial our own area code for every local call. This is a real annoyance. And so instead of dialing seven numbers, you have to dial 11 numbers every time you make a telephone call. A--a reporter for the Los Angeles Times named Robert Scheer, who's--has a column and is very well-known, became interested in this issue because a person--actually, he's a cosmetic surgeon--was just furious about this and had called his local assemblyman and gotten nowhere with the local assemblyman, so he realized that there was no two-way street with his being a constituent and assemblyman.

So bombarded Scheer, saying, `You've got to do something about this.' So Scheer began writing articles about it in the west Los Angeles edition of the LA Times. At that point, several other people became involved in it, among them three people who were very--who were masters at setting up a Web site, and they set up a Web site. And on this Web site, they did the essential reporting and explaining of why it was that the telephone company had run out of numbers. And it wasn't that they'd run out of numbers, it was the way in which the PUC allocated numbers in these very inefficient batches of 75,000 at a time. So there are telephone companies all over America that have great batches of numbers that aren't used, which has created this artificial shortage.

Well, at that point, an assemblyman, Wally Knox, got into it because it was his assembly district. He got out on e-mail and said, `Are you, you know, mad as a bee about having this dialing system overlay?' Well, needless to say, the people who are on e-mail are the kind of people who want to save their time. So, of course, he got this tremendous response from people who were interested in--in doing something about it. And they began e-mailing and phoning and writing the PUC. And it became a hot issue and Scheer wrote column after column about it, but there's one other technology in this that's just fascinating.

Fin--Wally Knox introduced a bill that said they were going to repeal this overlay system, and they were going to stop it as it was planned in several other places in the country. And the bill was on the floor of the Assembly, and I guess the lobbyists who were affected felt that they had defeated the bill because they had put in a poison pill provision. Well, Scheer was on a cell phone and called the various assemblymen who said that they were going to vote no, and said, `Why in the world are you voting no on this?' And they said, `Well, we've got this provision in there.' So about the last minute before they were going to vote on it, he telephoned Wally Knox, again from his cell phone, to the floor of the California Assembly and said, `Get that measure--get that one item out of the bill because we can pass this if you do it.' He took it out, they had the--the vote, it was overwhelmingly successful. We're getting rid of our overlay, and now this is being sort of a model of what you--how you can be a p--an activist using these multiple technologies, but it was essentially a cyberspace story of e-mail and this Web site.

Mr. DEUTSCH: And I--I think one of the things that really makes stories--the--cases like that work is that it's not just happening on the Internet. It's not an Internet community by itself. It's coordinating with people who have connections and with press and other things...

JOYCE: Right, the newspaper and...

Mr. DEUTSCH: ...that are integrated to make it--make it work. There's a couple of cases of huge campaigns on the Internet that brought the Internet community together, most notably the Communications Decency Act. And when the first round to that happened, there was, you know, so many Web sites involved and e-mails going on, but it didn't get beyond the Internet. So everyone who was involved in discussions on the Internet knew about the issues, but the groups that could have helped turn the issue, if they had been educated to get involved in the issue, were never reached out to because they weren't on the Internet.

JOYCE: The last e-mail I got on it was we're having a great big celebration at the Beverly Wilshire Hotel.

Mr. DEUTSCH: Congratulations.

Mr. LESSICK: Joel.

Mr. JOEL FLEISHMAN: As I listen to you-all speak, what struck me about what you said was that--that the--that the Internet was very good at creating communities of like-minded people. I heard that essentially from everyone. Those were Ken's terms in terms of the--the way in which a facility--and I understand that. In most of our deliberations, we have been concerned about how you facilitate the creation of communities of where there are differing minded people.

I mean, the typical geographical area has people who see--have different views on everything. We've been very much concerned, whether it's--you're talking about a town or institution or you--or a university or what have you or the nation, of how do you foster a greater sense of community there in overcoming differences of opinion? Where do you--where do you make that connection between people who have--who--who are drawn to the Internet because they ha--they share the same views on a particular issue and--and between others who don't, who differ on those issues? How do you create community across those groups?

Mr. LESSICK: Julian.

Mr. DIBBELL: I--I would say actually I didn't--in--in my and--and in Marc's presentation hear anything about like-minded people. I--I--I heard about people coming together around shared--certain shared interests. In--in Lambda MOO, it may be cre--actually there aren't too many shared interests in Lambda MOO. Some people might be building the place. Some people might be there just to chitchat, but they're all there because of these things that draw them there. And so you actually get a very robust argu--argumentative culture there because people have dif--very different perspectives on everything else.

Now this is leaving aside the eternal question of--of universal access and the fact that, you know, the people that do not have access to the Internet right now are very demographically specific in--in disturbing ways. But within, you know, the realm of--of--of online community is you often get people drawn by one particular aspect of what they're doing there, but who come from all kinds of different perspectives. Otherwise--and...

Mr. LESSICK: But there--there is some--there's a feature of Joel's question I want to make sure you're focusing on, and that is, Joel, you're--you're confronting a context where people have no option to leave. They have to live with their neighbors. They have to get along. They have to live in the university. They have to get on.

Mr. FLEISHMAN: That's right. That's right.

Mr. LESSICK: And a community in that context is learning to get along with and dealing with people. Now--and the perception, as I hear, as an undertone to your question is that in--in these contexts, people can just up and leave. Now I don't actually agree with that, that that's what can happen in here, particularly in Lambda MOO. So do you find--do you find people escaping when things get rough easily, or what's the cost in leaving? That's a way to...

Mr. DIBBELL: Well, the cost--usually by the time you leave, you've paid so much cost in torn hair and, you know, tears at your keyboard, that there's not much of a cost of leaving, but there is an investment, and there is a real gravity to these places that keeps people there in spite of whatever differences they may have. Similar, I--I think--and Marc can speak to this. I mean, there are huge ideological debates that rend the open source community, but there is an investment--they--they can't leave because that's the community. I...

Mr. LESSICK: But--but focus a little bit more on this particular feature. If you've spent 5,000 hours building a character in Lambda MOO, it's not like real space where you can just sell your house and take your assets and move to another community. You can't move that character to someplace else. So if this is the kind of life that you like, then it's hard to transport elsewhere, so you have this feature that you're describing. Now there were times when you could move characters; you know, there were facilities for moving characters. But the fact that you can't leave is, I think, creating the feature that you're pointing at.

Mr. DIBBELL: I should point out, though, that--that now from some--on some fantasy role playing games in which you do develop very programmatic characters by winning a certain amount of--of contests, people are taking to eBay and auctioning off those characters when they leave and--and liquidating them into...

Mr. LESSICK: that's great.

Mr. DIBBELL: ...into real assets that the other players will take more seriously.

Mr. LESSICK: Cass?

CASS: You know, these are two not extremely closely linked questions, but they directly bear on what we've been doing. There's a psychological phenomenon called dupe--group polarization in accordance with which, if you've got members of a deliberating body, by the end of the deliberation, there'll be a shift in the more extreme version of the direction indicated by the initial predisposition. But if you've got a group of people who think Clinton's kind of bad, by the end of the discussion, there'll be a consensus that he's quite bad. If you've got a group of people that think affirmative action is a good idea, by the end, they--they'll move to the direction of thinking it's a very good idea.

It's a statistical regularity. And there's been recent work suggesting that on the Internet, this is a potentially dangerous tendency because it can lead to national vulcanization. With deliberating groups of individuals who have an agreement, let's say, on something called X, and by the end, they'll think a very firm version of X simply because they haven't been dispo--exposed to competing views, they'll be uninformed and extremists. OK? So my question I guess for Julian and Ken is whether they've seen anything like that in the--in the work which they've--you've done? I guess when you describe the Communications Decency Act, in a way, that's a--that's a story of popular mobilization against a crazy law, but one might worry a little bit too much about government by referendum. The issues there are actually quite complicated. Some of the e-mails I got...

Mr. LESSICK: Right.

CASS: ...about the Communications Decency Act was just crazy stuff where people were stirring themselves up into thinking Stalin had come into Congress and was running it. OK. That's one point.

The other has--has a link, has--has to do with open software. And that is the suggestions you were making were really about the advantages for--for consumers of having something where they can remanage their own code. But there's something that policy-makers, Larry particularly is interested in, which you didn't stress--I'd be interested in hearing your observations about, which is not so much the consumers' advantages of open software, but the kind of civic advantages. And here, the notion is that Microsoft or Netscape or something does a certain steering, and what's pernicious about it might be thought is that it's invisible steering. So that, you know, you've got things automatically on the screen that you can't a--alter very much.

Now some people think that these are kind of overstated concerns because the political and market pressures on Microsoft, even if it is a monopolist, are such that the steering is going to pretty benign. It'll be NBC rather than ABC or one flower company rather than another. But is--this is really a question about your conception of what open software is about. Do you see it mostly as producing more consumer choice in a way many but not all consumers will like or do you see it as having important kind of civic function? And--and maybe you could be concrete about that.

Mr. EWING: Well, let's see how concrete I can be. I--I would guess I would see it as--as being both in choice in the sense of--of allowing innovation both more rapidly and just simply allowing more of it than you have in an enclosed environment. When you've got--infor--information technology infrastructures are--are--are not necessarily unique, but they have this--sort of this network effect where, you know, if everybody's using Microsoft Word, I've got to use Microsoft Word, and these things have been talked about a lot by economists.

And it--it--it becomes--the result of it is one--one corporation ends up owning this infrastructure. They get to own the innovation of that infrastructure or and--or lack of it. If they don't innovate, then nothing much happens. And so what open source does is really open that up so that a--that the--the infrastructure is really all laid out there for people to participate in either developing or not developing if the public sees that it doesn't need anything. But what I think is dangerous is to let--let the--that decision be made in one place in--you know, in--in a single corporation, not have the public be able to participate or--or decide that, you know, that--that a certain new feature needs to be developed or--or whatever. That--I--I'm sorry. I missed the second part of it. Wha--I was arguing. I was going to say that both of yours were true.

CASS: Well, the--the other had to do with civic, and really what you're emphasizing is the advantage for consumers of flexibility...

Mr. EWING: Right.

CASS: ...and do you--rather than the citizenship values, let's call them, of decreasing private authority to steer via closed software.

Mr. EWING: Well, that's one that's harder for me to answer.

Mr. LESSICK: It's interesting to note the origin here. Th--before the open--there was open source, there was a movement called the free software movement run by a guy names Richard Stallman, one of the MacArthur genius people. Free software for Richard Stallman meant, in one sense, software you didn't have to pay for, but that's not what's important to him. He meant free software in the sense of free speech; that you build an architecture that is free of control, and anybody can develop in any way that they want. So for him, this civic feature and for people in the free software movement that still lives today is extremely important. And...

Mr. EWING: Yeah.

Mr. LESSICK: ...are you--are you denying this importance?

Mr. EWING: No. I--I would say that's true. But I think that, Cass might argue that that's the same thing all over again. That's ju--that's not civic, but that's--again, that's just consumer control and consumer choice.

Mr. LESSICK: Yeah.

Mr. DEUTSCH: Ju--let me address the first part of that question. I--I have seen it to an extreme in terms of the polarization that happens, and it's mostly because people do get into these groups by themself because it's very easy to find discussions just on your perspective on the Internet. There's actually a very dangerous phenomenon being developed. There's a new term--I don't know if many of you have heard it called--hacktivism, which is--literally there are discussions on it and there are people who call themselves hacktivists. And there are people who think it is a legitimate form of political expression to silence their political opponents. At one point, when the--during one of the votes or at one point when the gunfreekids site was very active, it was attacked by what is known as a denial of service attack where someone on the other side of the issue decided they didn't want these people to speak and sent so much messages to the server that it brought it down.

I think the--in terms of how to counter that, you know, one--one of the things very clearly is just that there's a problem with new users to the Internet not knowing how to evaluate the type of speech that's there; being very careful with my words here. But it's--it's a phenomenon of changing the type of media. We're used to, as a society, edited media. And however--what kind of ever--influences there are with the press, we already have our degree of cynicism about the press. People aren't as cynical as messages they read on the Internet, even though they should be. And so it's very easy for someone who's got a hard-line position to drive the debate. And I think the--the biggest improvement that needs to happen is to train people, whether it's in the schools or at a later age as well, how to critically evaluate where that message is coming from. Is it coming from a special interest group? Is it coming from a reliable source? And then to--to act accordingly.

Mr. LESSICK: But--but this is a m--market opportunity for you. I mean, Cass' phenomenon is what makes it possible for you to really deliver Internet campaigns that are powerful.

Mr. DEUTSCH: Yes and no. I mean, at the risk of hurting my own business, which I--I--I often do--I mean, one thing I just did is admitted one of our sites was--was attacked in that way. And I think that is actually one of the problem, is the--the self-interest of most companies. Hacking is a phenomenon that there's probably not a corporation, a political group, a government agency that has not been attacked successfully. But no one wants to talk about it. And the fact that everyone's trying to deny these problems and push them under the rug is why they continue to exist. Mr. LESSICK: OK. We're running into a time problem. We have two questions here. Jay, and then a question in the back.

JAY: I read the--Microsoft's panicky memo about open source software and found it quite interesting. But I thought it could have been a panicky reaction virtually from every Fortune 500 company and hard-core capitalist in the country simply because of something that Larry mentioned in an aside, the aspect of the gift economy that operates in actually both of your cases, where there is competition, there is an attempt to excel and to stand out, and there is even a kind of scarcity in certain respects. But you win not by accumulating the most; you win by giving away the most. And so that's a gift economy, right? And the people who succeed are the ones who create the best things for the community, and those are the people who are the winners in that form of exchange, which is the total opposite of commodity exchange. So if I was Microsoft, I'd be panicking for more than one reason, because that's obviously a very different model for how an economy can work, a gift economy.

So--and--and the reason this is important for us as a commission is that it underlines why we're interested both in public discourse and community. When the underlying purpose of the community is to build something of lasting value--and the subtitle for both of your presentations would be: If They Build It, They Will Come. Right? If the underlying premise of the community is to build something of value, then the discourse for that community has to be different than if the underlying purpose is just to outdo everybody else or to get a scare resource for yourself or to grab power or something like that. So th--the discourse in the MUD and in the open source community is it's task-oriented and it's oriented towards building, but it's--it's different because everybody there is contributing to something lasting and excellent, which is actually another model of public life, not just a game or a software development method. And that's, to me, why these two presentations are really important to us, which is not so much a question--I apologize for that--as a comment. But that's what I heard in them.

Mr. DIBBELL: But I--yeah. I want to underline your use of the word `model,' because, you know, what--what I sometimes get in talking about the possibilities of online communities of various sorts is, you know, `Well, how does it translate into offline action?' you know.

JAY: Right.

Mr. DIBBELL: And--and the kinds of things Ken are--is talking about are really interesting, but it--it doesn't invalidate situations that are modeling things that might translate usefully into other situations. I--I mean, the MOO is really, I mean, entertainment. I mean, this goes back to Neil Gaylor's presentation. I mean, it's a--it's really ultimately a kind of entertainment, but, you know, if the movies are important to us to talk about, then, you know, in--in--in a political sense and in ways that they model political discourse, then this certainly is such. And--and the online--the open source community even more so, because, I mean, their model translates directly into so many other forms--realms of discourse that are threatened right now by the increasing anxiety of corporations about their intellectual property and that are--that are threatened by other models, by an old model that's trying to, you know...

JAY: Right.

Mr. DIBBELL: ...assert itself.

Mr. EWING: I'd like to come back to your--your initial comment about all the companies being afraid of open source. And you'll probably find that most companies--and I'm talking about the IT industry--most companies actually aren't that afraid of it. Th--they see it as liberating in a way. And the reason Microsoft is afraid of it is because th--they do at this moment sort of own that infrastructure. And that infra--getting that--freeing that infrastructure and be--making it become a public infrastructure rather than a p--a private one is what all these companies see as--as a great opportunity for them to be sort of lifted out from under th--from under Microsoft. Maybe this gets back to Cass' civic thing because of--we're talking about something as important as the communications infrastructure of the future. We're talking about the Internet. We're talking about operating systems, the basic communications stuff.

JAY: Just to clarify, the thing that's threatening to them is not open source software. It's the gift economy. That's...

Unidentified Panelist: ...(Unintelligible) I didn't understand this as--this as necessarily as a gift economy. I mean, that it's--by putting your software out, you get other people to work on pro--you find out who's working on problems which are similar to yours.

Mr. EWING: Well, yeah. You see...

Unidentified Panelist: And, I mean, there's--there's--I--I don't want to say it's--there's no altruistic spirit in this, but one could see an ec--a clear economic...

JAY: That's why it's a gift economy.

Mr. EWING: Yeah.

JAY: I mean, the point is it's going both ways, right? I mean, it's just--it's not a direct compensation.

Unidentified Panelist: But there is an ec--economic...

Mr. EWING: The gi--the giving it away...

Unidentified Panelist: It's consistent...

Mr. EWING: ...what it--what it does is to serve is--is build you up--your--your user base, right, or your constituents or whatever it is. It doesn't necessarily get you the money. And that doesn't service all of their needs, either. And what it does is just sort of shift the economy away from an intellectual property economy to a service economy. And th--this is what companies like Red Hat and others are doing now and--and--we're not unique in this respect. Most of the software companies around the world are making this transition now to be more service-oriented as opposed to this--this kind of str--kind of licensing model.

Mr. LESSICK: OK.

Unidentified Panelist: But Marc made another point. You--you said it was a meritocracy, that the best ones get used most widely. And I think the--the capacity for--to return a meritocracy to political life is something that people--or economic life perhaps, too--is something that people yearn for and haven't found the capacity in other mediums.

Mr. EWING: Right.

Unidentified Panelist: And so if this is really providing that, it's another reason why it's of interest to--to us and the kinds of questions that we're dealing with.

Mr. DEUTSCH: And I think one thing, just to be really clear with that, is that I agree with you. The danger of putting values on top of the technology is--you know, the--the technology doesn't have values, the technology leads itself to community. It gives a whole new range of communication tools. And what types of laws and other things are done are going to direct whether or not it moves in that direction, or whether or not this technology is used to control people.

Mr. LESSICK: OK. So now this is a closed technology I'm exercising, and here's the last question. Over there.

Unidentified Student: Actually it's not a sub--it continues. It's not really a...

Mr. LESSICK: Great.

Unidentified Student: But, first of all, I want to say thanks for letting our class come. We've enjoyed it. And we'll have a lot to talk about. My question is for Ken Deutsch. I'm just wondering if, since you started lobbying this way--a radically new way where you get to actually have everybody be a professional lobbyist because they have a script in front of them--have the congressional staffs changed? I mean, have they become more sophisticated technologically? Have they become more sophisticated in dealing with what used to just be a constituent now is sort of a well-trained, well-versed lobbyist/constituent? And I'm wondering if it fed back and modified the way the staff chain worked?

Mr. DEUTSCH: It--it is affecting congressional offices. And it's affecting them in a way because the volume has increased and it's not just because of the Internet. And it's--basically, there's a, you know, arms race going on between the different special interests to increase what they're doing. And what it's done is it's increased the stakes. So what happens now is that most e-mail messages, most faxes, are just cataloged and they're not necessarily read. So it goes to a point where the personal contact becomes even more important because of the increased use of the technology.

Mr. LESSICK: OK. Thank you very much.

List of Transcripts