The Wealth of Networks: How Social Production Transforms Markets and Freedom — transcript
Machine-generated transcript — not human-reviewed
About this transcript
The original audio is authoritative. This draft may mishear words or names, omit speech, and merge questions with answers. Changing voices have not been identified, and meaningful non-speech sounds have not been described. Unflagged text may also be inaccurate.
The full recording was submitted to local speech recognition. Generated text is preserved in its original order, including suspected repetitions. Machine-flagged uncertainty identifies possible errors, not verified corrections. Gap notes may indicate silence or omitted speech; they do not establish which.
Download the original recording (MP3 audio, 1:00:13). Timestamp links open that recording at the indicated time where the browser supports audio fragments; the displayed times can also be used to seek manually.
Send a transcript correction to Joseph, including the timestamp and the words you heard.
Generated transcript
0:00–5:00
And as you now, I'm here to welcome the country of high bankers from Yale Law School, who has just written a book called The Wealth of Nations, the subtitle... Wealth of Nations is a little out of date. The Wealth of Nations is subtitled How Social Connection Transforms Markets and Freedom, and that's his topic today. I'm especially proud and pleased to introduce him to you students because I met Professor Bengler when I was a law school student and had the honor of editing one of his early
works, first major work on the network economy and ownership of network resources and since then he's been one of my role models and mentors so I'm really pleased to introduce him to another round of music and I'm looking forward to his talk. So let's welcome. Thanks so much for doing this and giving me the opportunity to talk.
It's a little narcissistic talking about your own book, but it's OK. You're very kind to let me do it. So what I want to do is talk for about 35 give or take minutes and then have a conversation about what it is that we can say about the emergence of digital network communications that is stable, that is long-term, and that is significant from the way we perceive the future of our
lives together in contemporary societies. Let me start though with this.
Untranscribed interval, 1:54–2:13: possible silence or omitted speech. No description of this interval has been verified.
And you just breathed deep as we're saving And you first act swung running to sing like a stone For the times they are a-changing Come right in and they are due Come senators, congressmen, please keep the call Don't stand in the head or back up the hall
For he the future will be eaten with snow From the battle outside raging We're still shaking with the wind For the time they're out of changing Then the first one now will ever be last Cause the time they are eternal
Untranscribed interval, 3:14–3:35: possible silence or omitted speech. No description of this interval has been verified.
Now, if you think that's funny, you add the blue, clearly the blue of IBM and the story that is produced and presented about five years ago at the Bangalore Linux users group. And you begin to understand that here's a question we need to ask and the one that I'm going to try to answer in the next half an hour or so. what could possibly justify thinking of that trajectory,
relating one's story to freedom in all of these manifestations, concentrated around an operating system? And so I'm going to suggest to you that this is far from silly, though the particular manifestation, you have to admit, is quite amusing. And to do this by talking specifically about the economics and the politics. And the core question with the economics is, what is it about what we're seeing that is stable, sustainable, and different?
And the core question with the politics will be, what is the consequence of the stable patterns of behavior that we see coming out of the effort at a positive description in the economics framework? So here's an image. Between 1835 and 1850, the cost of starting what was understood at the time as a mass circulation paper rose from what was around $10,000 in 2005 terms to around $2.5 million.
5:00–10:00
The critical thing that is the difference between $10,000 and $2.5 million is the $2.5 million you need a business model and $10,000 there's a lot more diverse ways of doing it. And essentially what we saw around this change, and this was a change that was a combination of technology and organization and the sense of what it was to be a news organization in in terms of professionalization, was a stark bifurcation between producers and consumers, passive large audiences to an increasingly large community that was politically
relevant and socially and commercially relevant, and commercial professional elites as the producers. We basically from then on had two major models of production, market-based and government, which was true in general of the economy. and this characterizes and anchors the basic structure of information production for the next 150 years, beginning with radio, which didn't necessarily have to go this way, but did go this way, television, satellites, mainframes, roughly until now.
So here's another image, and it's moderately misleading, but not too much. So in the middle of 2002, a feeling of failure, fear, came across some people in the U.S. When for the first time, a Japanese supercomputer was much faster than its American counterparts, the NEC Earth Simulator. and two years later, IBM Gene Blue just inched out, this is teraflops,
just inched out the NEC Earth Simulator and there was a collective sigh of relief of sorts, at least in some quarters. Now, throughout this time, SETI at home was dwarfing both of these. Now, as I said, this is somewhat misleading because SETI at home is very special purpose. There's all sorts of differences. But the basic point is this. The collective shared leftover computer cycles of four and a half million people were capable of outperforming on a non-insignificant measure the supercomputers produced by the wealthiest countries in the world, engaging some of the most sophisticated companies in the world for purposes of joint public and private collaboration.
Now, this image for me represents the following fairly broad facts. We see a radical decentralization of the capitalization of computation, storage, and communications capacity. The three core components of a computer communications network. It's not that the capital intensity is low, that we are seeing deindustrialization. It's that if with industrialization, capital intensity increased together with its centralization.
And now we're seeing capital intensity continue, but in a decentralized form, so that roughly every connected person on the planet, somewhere between 600 million and a billion people around the planet, has the physical capital necessary to produce and exchange information, knowledge, and culture. So we're in a very new and interesting condition that we haven't been in for a very long time, if ever. The most important inputs into the core economic activities of the most advanced economies
are widely distributed in the population. And the first cluster of these I've already talked about, computation and communications resources. The other component is human creativity, intuition, experience, motivation, and time. And remember, these are not fungible. Any one of you who has tried to leave a job and leave a detailed memo for the person coming after you and assume that when they read it they'll know exactly what you know knows what I mean and I see the smiles and the nods. You know what I mean when I say that human experience and creativity and insight are
not fungible. So both of these components that are the critical components of the information economy are widely held in the population. Now what happens? What happens is that behaviors that were once on the periphery, driven by social motivations, behaviors of cooperation, friendship, human decency, don't suddenly become where they weren't before. But they move from the periphery of economic life, from the subject of anthropology and sociology of the household, to the very core of the most advanced economies.
Two phenomena I characterize, one broad, one a subset, but nonetheless a very important subset. As a general phenomenon, one of the things that emerges from the radical decentralization of capital is commons-based production. That is to say production without exclusion from the inputs, from the outputs, which can can be individual or collective, and can be commercial or non-commercial. The critical point is not whether it's commercial or non-commercial,
10:00–15:00
not whether it's individual or collective, but the fact that production is organized in such a way that the inputs are not appropriated exclusively, and the outputs are not appropriated exclusively, but rather they happen within a social system. A subset of Commons-based production is peer production. And by peer production, I'm going to spend the rest of my time talking about the human creativity of peer production, but we also see sharing material resources. Specifically, I talk about large-scale cooperation among human contributors that is not organized
by price signals in the market or by managerial commands within firms or government agencies. This is the critical defining characteristics, a functional characteristic of the behavior of the phenomenon rather than anything else. And we also see in the economics here are different, and I spent some time in the book as well as in an earlier article explaining why the economics are different. This is important, but I won't spend more time on it here. We are also seeing sharing of material resources, distributed computing. As I mentioned already, Setia at Home is the example, and there are about 60 such projects
now. We're beginning to see wireless mesh networks. This is the article that Molly edited and that actually Mitch, who sat here, put me on in 94. The possibility of shared wireless creating a network. Distributed storage. If for a moment you thought that you would try to tell someone in 99, build me a network
of storage. Build me a system of storage that would store terabytes of data, would be available 24 hours a day, 7 days a week to 100 million users, would be robust to attack from multiple fronts, including people seizing the main server with guns and putting it down, etc. You'd say, you're crazy, it'll take me 10 years and billions of dollars. But that is of course P2P file sharing. When you ignore the content of the data, it's a massive, massively efficient, robust, distributed storage system. And we see mixtures like Skype.
So the poster child, and I'll say no more about free or open source software in this context, the poster child is free and open source software and the critical intervention of that story in my story is something like this. The basic response to peer production by some set of people is, it's a fad, it'll pass, it has no staying point in the market, why do people do it? And anybody who has that intuition has to respond to this problem.
In 1995, two groups of people thought that it would be really important to build web server software. One group was a bunch of people who thought they needed a better web server software. The other group was Microsoft, thinking that this was central to their strategy. Ten years later, one of these groups has about 20% of the market, the other has 70% of the market, and if I told you that the Motley group of collaborators actually continued to hold 20% of the market ten years later, not three months, not six months, not fashion, ten years later, you'd say that's amazing. But of course, Apache is the 70% and the Microsoft servers are about 20.
So this is a fact. We're seeing peer production occur in many more domains of information production and exchange because the underlying economics that enable free and open source software are the economics that I've already suggested. Radical decentralization of the capacities embedded in a system of sharing that allows people to, through social processes, collaborate to produce things they value for a variety of diverse reasons.
So this is NASA Click Workers, an experiment that was run for a while by NASA trying to take images from Mars and converting them to map. Usually used to take a full-time, fully trained PhD. They chunked the work into small bunches like this. They put them up with an intuitive interface that said click on force points and we'll generate a circle. They give it to a few people, 5 to 15. They generate a consensus estimate. They converted to the map.
After six months, they did a study. 85,000 people had participated. The quality of their output, consensus calculated, was, quote, practically indistinguishable from that of a fully trained PhD. Now this is, in some senses, trivial, and in some senses, delightful, because you can actually see how you take the same problem, organize it differently. And through this organization, reorganization are able to tap into a diverse group of people who are willing to spend five minutes playing this instead of watching TV.
15:00–20:00
And you generate a measurable output. I used to have to explain more about Wikipedia. I started writing about it I think in late 2001, about five or six months after the first set of articles. Now I don't need to explain what Wikipedia is, I just need to say one thing. We all hear the anecdotes over time about how poor and how much of a problem there is. I tried to spend a good chunk of time in a chapter explaining this away.
But let's remember as we measure, as we look at anecdotes, we have one study. It's self-contested from nature that actually tries to look at science definitions. And they took the definitions, they sent them out to scientists, and they came back with responses. And the response was, as you would expect from fully trained professional scientists about encyclopedia definitions after all, they're both roughly equally crappy. Which is okay when you're no crappier basically than Britannica, and you're produced by 60,000
volunteers where anyone can edit what they want. Not too bad. And we have a variety of other things. So Yahoo, the human editor directory, was its original, the original invention was you come here, you see what actual human beings as opposed to an algorithm chose many years ago. Now we have the open directory project with another 50,000 or 60,000 volunteers doing the same thing when you have somebody who is intensely passionate about a subject looking at websites and saying this is worthwhile putting in the directory, this is not.
Basically, at a time when capital cost was very expensive, we had three large forms of transactional frameworks mapped along whether they're centralized or decentralized and whether they're market based or non-market based. We had decentralized market based system, that's the price system, the classic market. We had centralized market based systems, these were firms. systems, and we had centralized non-market based systems. These were government and traditional non-profits. And this is all transactional networks for the economy.
What we have now, because of the decentralization of the technical and physical capacity to contribute, is the emergence of this fourth transactional framework, social sharing and exchange, not that we didn't have social sharing before, but the salience as an economic phenomenon, as a source of flow of information, knowledge, and culture in the total economy of information, knowledge, and culture has increased. Increased to the point that it's affecting everything.
Affecting the economics of these other models and affecting the political system that in the past was dependent on these three and now can depend on these four. So it's always important to remember, my claim is not, repeal all economics, we're all gonna hang around doing fun stuff. My claim is that hanging around doing fun stuff has a bigger impact today in these domains than it was able to do before. And the political mapping is what happens
when you live in such a system that has a higher salience for social sharing and exchange. So the easy things, the things that are most visible. New sources of competition. P2P software competing with the recording industry. Free and open source software competing with Microsoft. Wikipedia, certainly with Grolier and Encarta, the paid for. Maybe even Britannica, if enough times respectable places like Nature say that they're equally crappy, maybe there'll be competition there too. Skype with telecommunications.
In addition, we're also seeing much greater recognition of the new opportunities. The shift from the value we place on well-behaved appliances like TVs and CD players, versus the value we place on PCs, handhelds, wireless, various production tools. The strict bifurcation between consumer and infrastructure goods shifting over to dual-use machines like soft infrastructure, like Wi-Fi devices and Skype. Most important for purposes of the rest of the talk today, we're seeing a shift from purely focus on finished information and cultural goods, like movies and recorded music, to
platforms for self-expression and collaboration. Be it GarageBand, be it something as simple as Craigslist, be it Technorati, be it the whole framework that now is beginning to understand itself as Web 2.0, largely understanding itself as trying to build business models around building platforms for people doing things in new ways as producers. And the final model is in some sense is quirky. It has the following model. It's stuff will flow out of connected human beings and the probability that it will flow
20:00–25:00
as an input into production is sufficiently high that I will contract with someone else over here and say, I promise you to deliver this service on such and such a day because Because I rely that stuff will flow out of connected human beings. And not only that, I'll pay people to participate with this. And I might add political support. This sounds crazy until you look at IBM's revenues, the largest patent holder in the United States. The green is royalty and licensing related revenue. The blue is Linux related services.
And maybe it looks a little less crazy than before. The measure is thousands of millions of dollars. So, social production, a real fact, not a fad. It is, I suggest to you, the critical long-term shift caused by the internet. In some contexts, it's more efficient and more innovative than markets and firms, not always. But what's critical from the perspective of everything else I will talk about is that even when it's not more efficient,
It is sustainable and growing fast. And from the perspectives of political implications, the fact that it's real, stable, and growing is the central point. The fact that it's efficient is a great bonus. From the perspective of economy, obviously, it's otherwise. needless to say, as we already saw from the sources of competition, this sort of behavior is a threat to and is threatened by incumbent business models. A lot of what we're seeing
today around battles around intellectual property, telecommunications, a variety of other new and funky laws is basically the battlefield where the incumbents of the industrial information economy are trying to make sure that the law makes it harder for these new modes of production to exist, and those both in society who are enjoying these new freedoms and these new capabilities, and those who are beginning to see that there are business opportunities around supplying platforms for them, are pushing on the other side.
And the very last two slides will actually suggest to you that it's not that there's one way or another, there's a real push and a battle between social, economic, and cultural and regulatory forces on both sides trying to battle which way the world will go. But first, let me spend almost the entire rest of the lecture talking about the politics or why we should care about the outcome. And I'll talk about, in about 15 minutes, I'll talk about autonomy, democracy, justice
and development. and this will all be very relaxed and calm and detailed. Autonomy. It's a hard concept. I spend more time on it than I will spend here. The core changes, and these actually enable both the economic production and the political and justice aspects, is the change in the practical space for human agency. Individuals, practical as they are in the world, have new capacities to do more for and by ourselves,
and we have new capacities to do more in loose association with others. The looseness here is important because traditionally when there were things that required cooperation, because of distance, because of capital constraints, we usually had to enter into very strong commitments, be they co-worker commitments, neighborhood commitments, family, close friendships, that limited the range of things we could contribute to. There's a lot of decrying of the breakdown of social ties on the net,
and I tried to spend a chapter to suggest why it's less of a problem that people are concerned, which I won't talk about here. But the looseness, the very ability to cooperate with many people in many different contexts on a limited basis also creates completely new capacities. So just two quick stories. If you go on Google, and I've been doing this for five years since my 10 year old was five and I first came back from seeing a replica Viking ship and we looked for them.
Wonderful source of materials on Viking ships. It's been first, second, third on Google for the last five years. It's produced by a fifth grade teacher in Gander, Newfoundland, out of his classroom. And if you can for a moment imagine what it would mean for a fifth grade teacher in Gander, Newfoundland to say, what I want to do is build, I'm passionate about Viking ships, I want to build a learning tool for children around the world. So you want to do it, so what? You simply couldn't. Now you can.
25:00–30:00
The range of things you can imagine for yourself is different. Similarly, the range of things you might not be able to do by yourself, but you could do with the help of others changes. So Michael Hart for a lot of years tried to create an online library, but only when things really get going and the networked environment allows him to collaborate with many other people do we get something like Project Gutenberg. So the same constraints that were there because one person can't do them alone, again, are lowered. So that's autonomy.
Let me talk and spend most of my time, I think, with regard to the politics on democracy. And the thing to remember is that our experience with the rise of democracy, first its victories over monarchies and then its victories over authoritarian regimes in the 20th century and its generalization over the last 50 years, has all occurred in the context of mass media in the public spheres. We have the early Dutch Republic, we have the early American Republic, we have three years in the early French Republic. There are examples of non-mass media based republics.
but largely the growth of democracy are experiences within a context of mass media. And so the basic question is, and we have many books criticizing the mass media as a platform for the public sphere, and I spend a chapter bringing those out a little bit. But what I want to talk about here is, what does this network public sphere look like? What are its components? What are its constraints?
So let me tell you a story and let's start with this, the greatest of the high moment of the role of the press as a platform for the public sphere, the Pentagon Papers, the moment at which the American public begins to learn in some deep sense and broad and accepted sense that Vietnam was a mess from the beginning. Ellsberg brings the papers to the Times, the Times pays people to sit in a room for weeks and figure it out and then publishes it.
There's a U.S. injunction. They shift it over to the Washington Post. The injunction keeps going. They have enough money and clout in society to take it all the way to the Supreme Court, and Deus Ex Machina, the Supreme Court, pushes away the injunction, and the American people get their information. Now, this is the fourth estate function. This is the way it's supposed to work. What's the alternative? How does this happen in the network public sphere? So, Diebold is a company that makes voting machines, and as you may recall, in 2000,
the United States had some disagreement about the quality of voting outcomes in some places, and so one of the solutions was supposed to be mechanical voting machines. 2002, we see the first efforts at these kinds of machines in Georgia. The mass media, almost no reports. Two or three reports, it's technical, it's hard, there might be a little technical issues, but we're working on solving it and we have the best people working on solving it. These two quotes coming from the state officials who bought the machines and the vendors who sold them to them. And there are a few people shouting on the side.
Now, you can't blame them. It's hard. It's proprietary code. It's technically difficult. It's expensive. There's a lot of obscurantism. But that's what we want the fourth estate for. So how does it happen in the networked public sphere? One activist, Bev Harris, finds the source code to the machine. and posts it on her own site, as well as on a site in New Zealand, Scoop, that says look at it. And now it's very interesting to see what the editorial was around it, because it reveals some of the basic components of the network public sphere
and how they differ from the mass media public sphere. So the first thing is, as we anticipate attempts to prevent the distribution of this information, we encourage you to make copies. So it's not about hiring a very good lawyer and going and lifting the injunctions out. Distribute it out there. The second thing is, here they are. They're here for you to read. See for yourself. Not trust us, we're the reliable media. It's here's the raw materials, let's work on it. Someone competent, perhaps more than one person competent, will be able to find things out.
Let's talk. By the way, the files, some of them are password protected. But here are the nice people from lostpasswords.com, and they'll help you get around that. And some of them might be corrupted. And here are those nice people from ZipRepair. You can get something from them. And we don't have money to pay anybody to spend weeks going through these. But it's really important. If you find something, post it here. Completely different model. Now what happens? What happens is, in fact, one of the people who looked at this, Avi Rubin from Johns Hopkins, came up with his group with a set of problems with the machines.
They were respectable enough that they got media exposure, Diebold responded, there was peer review. This prompted a set of studies in Maryland, one of the states that had bought these machines, and they required modification. At that point, Bev Harris becomes a bit of a node. And here's another point. So an anonymous whistleblower, it's actually hard to nail down how these came out, but an anonymous whistleblower took the cache and sent it in two directions. One into the hip version of the mass media, the other back to Bev Harris. So Ward tells a story, can you believe these idiots lost
30:00–35:00
their source code again? This time they lost emails. And that was it. Bev Harris, she knows what to do. Put it on. Let's see what happens. Except this time Diebold has a different tactic. This time they go after the ISP based on the Judgment and Copyright Act and say, the emails, they're ours. And you can't have them. Copyright. You have to cut it off. However, we know what people do. So some Swarthmore students reposted the materials. Well, Diebold also knows already what to do, and who's the ISP for the students? That's the college. The college gets a DMCA request, shut them down, and that's the end of the story.
No. Because at that point, the materials have been distributed, and this is what's fascinating about the ecology of the networked public sphere. What do we have? We have individuals in all sorts of other campuses. We also have a very motley and bizarre interaction between the legal and the illegal, the commercial and the non-commercial, the proprietary and the open. Because what do we have? Some of the email goes into Freenet. Free software developed to deal with censorship, not at all with music.
Some of it goes on Overnet, developed by a commercial firm, making money off people, largely doing something that's illegal, using a proprietary framework. but also to which there are free software interfaces as well. So what we see is a very curious set of interactions, ultimately bringing to the point where the emails make their way to the California State Commission. And it's quite wonderful. You read the transcript of the California State Commission,
and it's literally one of these Perry Mason moments where, well, we were going to approve these machines today, but new information has come to light, and let's do a study. And then within a few weeks, a couple of months, they come back and say, well, we did a state study, and it looks like a large number of these machines that we supposedly certified are different from the ones we actually certified. Why don't we decertify a whole bunch of them and have you deliver an use? So genuine, effective political action through information identified in a decentralized form
without the mass media. Now, it's not that courts were irrelevant here. The students actually went with the EFF support and went to get a declaratory judgment from a court and they did get it so badly that Diebold had to pay their fees, which is not the common American way of doing things. But it happened a year later when it would have been too late to make a difference to the election. So the network public sphere solved the problem for itself in a completely different way.
Now the argument that the internet democratizes is very old. As old as more or less people outside of the three and a half people working on it became even cognizant of the internet and maybe older than that too. So the theory of everyone a pamphleteer is hardly new. And it's been subject to a series of critiques, the first generation critique largely located, for example, or most prominently in Cass Sunstein's republic.com.
Fragmentation, the Babel objection. Everybody talks just to themselves. Nobody really has a common public sphere. There's polarization. we know already who will play the watchdog. The second generation critique, in a sense, answers the first generation, but creates its own problem. And that's the power law distribution of attention on the net. The fact that a tiny portion of websites are linked to by thousands of sites or millions of sites, and thousands or millions of sites are linked to by no one.
And here the problem is, you can talk, but no one will hear. I spent a good bit of time in the chapter actually trying to look through the empirical literature on power law distribution of links. I won't go into it in the talk now, though I'm happy to talk about it later. The basic picture we have is actually that neither of these critiques is empirically correct. What we see is we see sites clustering. Political blogs are more likely to link to each other, even across the political divide,
than either one of them is likely to link to laundry detergent or purchasing sites, which creates lumpiness. At small scales, when you go to topical clusters that are sufficiently small, it's no longer the case that the tail really is that, that, that long and nobody looks like it. It looks like there are dozens, sometimes even low hundreds of numbers of sites that have five, ten, fifteen links, which means dozens or even low hundreds of sites where if you show something, they could be seen by a community and if they're intensely interesting,
35:00–40:00
each of these communities has somebody who's really seen. That ends up being the antenna to transmit it across the attention backbone up to the rest of the web. All of these connected in a hugely interconnected, strongly connected core where there are multiple redundant paths from each place to every other. So no, not everyone a pamphleteer, but we're also not intellectual lemmings. The structured web offers more visibility to more people in accreditation and filtration clusters, free, relatively speaking, of the financing constraints of speech because it's so relatively cheap,
Emphasizing, and this is critical, emphasizing what's critically, intensively interesting to a small group of people on the order of hundreds or thousands of people who link to each other as opposed to what's generally inoffensive or moderately acceptable to millions of people. Inverting what it is that counts as important from the mass media environment. We see strong linking practices in the major sites. We see many hubs themselves that are cooperative sites like Slashdot and
the Daily Cost, even TPM is beginning to use readers or users in this form. A very strong of see-for-yourself ethic as opposed to trust me, relatively easier for each individual to be heard, easier to form purposive association as we saw in the Diebold case, global not only local. important in all of this is to remember. If your baseline is utopia, this is not a big success. If your baseline is three networks, this is amazing. And so we're not at the three
networks. We're, let's say, at the 500 channels universe. But if that's your baseline and compared to it you see something like half the sites on the net are capable of being reached out from any point on the net, these kinds of models, it's an improvement. Now, politics is not only about politics, it's also about culture and the way we make meaning and understand. So one of the things we see is the plasticity and malleability of culture.
And so this is an example. Raise your hands if you're bored and you want to move on.
Untranscribed interval, 37:25–37:49: possible silence or omitted speech. No description of this interval has been verified.
I'm going to be coming through. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please. Please.
Untranscribed interval, 38:08–38:26: possible silence or omitted speech. No description of this interval has been verified.
Thank you.
Untranscribed interval, 38:28–38:57: possible silence or omitted speech. No description of this interval has been verified.
Thank you.
Untranscribed interval, 38:57–39:19: possible silence or omitted speech. No description of this interval has been verified.
What you need me to do, what I need you to do if I'm doing my job, I will take you. You need me to stop playing. You know, you're still a big, I can't take a moment to take, I need you to remember what they, I can't, I'm still a perfect, I can't tell you something, I'm a perfect, if you want to stop, I need you to stay to your own, so. Roll up, roll up, roll up, roll up, I heard all the leaves, go rock, roll up, roll up, roll up,
Untranscribed interval, 39:48–40:06: possible silence or omitted speech. No description of this interval has been verified.
40:00–45:00
So, just to compact it all into two words. The 20th century saw a period during which around economies of production of culture, We saw people de-skilling themselves from making folk culture and turning themselves over to passive reception of the symbols through which we understand how our lives are, how they can be, and why it matters. And what we're beginning to see now is the re-emergence of relatively high-production folk culture
for the retelling of narratives about what it is that matters in our lives. Now, in addition, we also see greater transparency Critical evaluation of culture moves from academic seminars to blogs to Wikipedia. The practice of production makes people better readers. So this is what you get from Encarta about what a typical Barbie looks like. This is what you get from Wikipedia with distinct descriptions of the controversies
over body image, with claims about the relative role of the doll early in the 60s when she was replacing all babying dolls with explicit talk about what the advantages and disadvantages are about talking about this way or the other. Of course, if you go to encyclopedia.com, you'll see Barbie Klaus. Similarly, if you look at Google, immediately after Barbie.com, you'll have adios Barbie, a body image for every site and a variety of others, a contested image. If you go to overture, where basically you get placed based on how much you paid overture
rather than by how many people thought you were good. By the time you'll get to these once you run out of people who paid. So finally, in order to leave a little room for questions, I want to talk about the small subject of justice. And here, it's important to recognize that more of what makes for human welfare and development depends on information, knowledge, and culture today in the globally networked information economy. If you just look at the Human Development Index, The three major components, health through life expectancy, education through literacy, and GDP per capita growth,
each of them is critically dependent on how we produce and manage the information inputs and outputs of our economy. Food security, medicines, research in general, and outcomes data are all information goods that go into the question of human welfare and development. Books and teaching materials, computation and communications, libraries, academic centers and journals go into education and literacy. For over 50 years we've already known that growth is driven primarily by innovation rather
than efficiency and innovation and the way that particularly second comers can access that innovation becomes central to the degree to which they can participate in the global information economy. So, commons based and peer production are not a solution. And in this regard, I make a much more tentative set of claims about justice, in particular about development, than I do about democracy and autonomy. Because here, the way we manage information instrumentally affects access to knowledge.
Although there are aspects of access to knowledge that are intrinsic. We're beginning to see at the level of infrastructures, free and open source software, playing a role in development. We're beginning to see open academic publishing making access to knowledge and participation better. We're beginning to see interesting study, interesting efforts in open source agriculture in order to avoid biological innovation and agronomic innovation in order to avoid the capture and the industrialization of farming.
we're beginning to see on the health side in bioinformatics an open source or open bioinformatics movement like the HapMap project, much more ambitious and not in any way, shape, or form close to being solved, but very important, moving that open development methodology down to the wet lab. So, lots of caveats, they're all there in the chapter, but just a little bit of images of what might be. So here's the free high school science text anchored in the University of Cape Town graduate students, peer produced textbook for South African high schools
to be available once approved free. The physics textbook took about three years and it's about done. Chemistry and math are taking time. It's not simple, it's not solved. It turns out to be harder, for example, than something that's much more modular like an encyclopedia, but this is roughly the idea. HapMap project we talked about. Perhaps the most advanced model of this is the BioForge and patent lens run out of BIOS and Cambia in Australia, an effort to create a major patent mapping tool that allows you
to find what's constrained and what's not constrained and develop through it, as well as a BioForge model on OpenForge and seeded with technologies that came out of Cambia that allow people to do basic research tools in exchange for a GPL-like model. You can use these tools if you contribute your improvements as well. And perhaps the quirkiest but farthest from reality, William Scott from Indiana University, Purdue University, putting together kits that then, for undergraduates, when they synthesize something that otherwise they would put in the garbage,
45:00–50:00
are actually synthesizing targets generated by distributed computation as potential compounds that might be interesting for purposes of neglected diseases. So, in a minute, as I said, none of this is uncontroversial. What we're seeing is a push-me-pull-you battle. between trends that are legal and not only legal. We see the broadband duopoly that's almost unregulated being pushed back by the development of open wireless networks to some extent and municipal broadband,
and that's a major point of argument. We see the push, primarily pushed by Hollywood, towards trusted systems and computers that can be trusted against their owners to do what the content providers want versus general purpose devices. There's huge market demand for general purpose devices as opposed to glorified TVs, and that's a real push there. Software patents and proprietary software and free and open source software and open standards, another push. And remember this is not about commercial, non-commercial, it's alliances on particularly the openness side. We see lots of law trying to lock things down, together with
the increasing success of sharing practices as well as a good degree of disdain for proprietary strategies that doesn't help those who are trying to appropriate other. In copyright, we see a wide range of changes intended to make things harder to copy, harder to share, harder to use as inputs, while at the same time we see an increasingly growing movement like Creative Commons, the free culture movement, the access to knowledge movement, as well as, as I said, increasing just plain practice.
Here, you'll like this. To wrap up, technological threshold conditions enable greater individual human agency. As a result, social sharing and exchange emerge as a major modality of economic production, as opposed to only of social relations. We're beginning to practice new ways of being free and equal human beings, but these practices are subject to a global and persistent political and regulatory battle. And what I want to emphasize in this book is that the choices that we're making today,
that may sometimes seem technical and completely unrelated to each other, are precisely choices about how to be productive, free, and equal human beings for the next 30 to 50 years. That seems to be a subject sufficiently important to be engaged in. Thank you. So we have a few minutes. Yeah. So I was one of the students that was cease and desisted in the D-Bold memos case.
I can tell you that making the decision to post those memos was not something that I did lightly. But none of us saw it in this beautiful way that you've explained the story. And so I'm wondering, there seems to be an axis of risk in this, that some of these things like writing patches for the Linux kernel, for example, aren't as risky as some action. And I'm wondering about direction and what your thoughts on risk and directing these things to come out with what ends up to be sort of a very socially beneficial outcome.
Yeah, I mean, this is actually a very, very good question. Unsurprisingly, coming from somebody who lives through this. the element of risk and the threat is an intentional outcome of the battle to change the laws in such a way as to make it illegal to do things that might not have been thought illegal a few years ago or may have been impossible a few years ago
I have no real solution in the sense that the risk is not inherent in the technical change or the social and cultural possibilities. The risk comes out of the fact that you're in the moment of a battle. And you're absolutely right, some fronts of this battle, like development of free and open source software, are not at the moment points of risk. Even in the context where you might have imagined that doing that would render you susceptible to economic ostracism.
That's not the case, whereas here you do see it. It is a real drag on the possibility of these practices. It is the core purpose of the campaign to render illegal any of these practices to create risk and thereby to stop these practices and shift people over to, okay, let's just read the blogs as opposed to write them. Let's just read the New York Times as blogs as opposed to write our own.
50:00–55:00
Let's just download film as opposed to mix and create our own because it's risky. Precisely that risk and generating that risk is a core component of the battle. Yeah? I actually wanted to point something out that I think you will better support your thesis.
You talk about the power law and how more people at the top of the curve get linked to and other folks were then. And I recently did a blog post comparing making the linking statistics from Hethnirond 18 months ago to the current statistics that were published in February. And actually, there's a 3x reduction in linking to the guys at the top. And so I talked about the maturing of the atmosphere and that there's a sort of conversational middle
that's going to be towards lower down that where people are looking at each other and there's now beginning to be evidence as there's 30 million blogs that go around. Very cool. I'll need to, the book is up by the way on a wiki so I'm going to have to immediately link or maybe you could to that because that, I mean the whole point of the wiki was exactly, I must have missed millions of things and I should have known of course when I come here to talk I would find three of them in the first two comments. But that's exactly the idea. Yeah, I don't know, this is a platform for starting to talk about these things.
So that's actually incredibly helpful. But you said there were two things. No, no, no, I was too sensitive to two of the items that I didn't know. I'm just letting you know, there's actually probably a bunch of people in the audience. So part of that is... Yeah. Yeah, I'm wondering how the traditional acting, the enforcing quality in our society, or the green quality map to this? Because you have, in fact, now we're excited like 600 million people have access
and 6 million people on the planet. Also, with the increasing amount of time people are spending working to support these houses and household production around the market, how does such a green quality of access to participating in the public sphere? And so how do you think about that? Okay. Okay, the first time you said equality, I thought you said quality. So I was listening to a completely different question which has answers in these chapters as opposed to those chapters.
So now I sort of need to rewind. Okay, traditional equality concerns in terms of people outside the 600 million to a billion. I don't have a good answer. The best answer that I have is what is the baseline? And for people who are poor and who either can themselves aggregate through, if you're talking about outside of the United States, through cyber cafes, through telecenters,
if you're talking about in richer countries through NGOs and community facilitation, the difference between the cost of being able to produce a tool, produce a broadcast station, do something in the old set of costs relative to now, doesn't solve the problem, but one of the sectors, and I didn't actually emphasize this here, One of the sectors whose efficacy increases because of the current cost structure is not only individuals and their disorganized social relations, but also organizations, like community organizations, organizations of civil society.
So what you need and what you're beginning to see is efforts at more traditional organizations focused on serving people who are off the net, becoming more effective using these tools. So you get some improvement, but nothing like this. And for that, what do you do with countries that have 40% illiteracy? And the answer is actually not nothing. Because one of the things that actually Brazil, for example, the Ministry of Culture is doing is actually putting together telecenters from one ministry, but the Ministry of Culture
is putting together multimedia centers that are intended to capture and communicate folk culture in the Brazilian sense, not in the sense that I described here, and trying to capture things that are not, but I don't want to wave my arms too much and say this will solve inequality at this basic level. It improves. Yes? I just wanted to jump the other way and get into the quality part. I noticed you mentioned that the open textbooks were peer reviewed and the concept of the peer review process reinforcing an existing paradigm
55:00–1:00:00
versus the fully open comments production that we talked about. I just wanted to get your feeling on if there are conflicts between those two. That's a great question again. Some kind of peer review is present in practically all these successful projects. Wikipedia is a form of peer review in the sense that the peers are not, we call it peer review in academia and really what we mean is review by seniors.
At least at some state, although that's actually, I take that back. That was a nasty joke at academia that's unfair. That's true when you're junior. Then it does become peers. And it's not hierarchy all the way up. It's more of a two-tiered hierarchy than hierarchy all the way up. And so it is in some senses much more peer reviewed. If you look at NIH grants by comparison to other kinds of government contracts, those really are peer reviewed. That is the academic literature. So we have peer review happening in real peer systems in the technical sense of the peers
decide what is quality. And we have peer review in the traditional academic sense, which people with appropriate accreditation decide what's good, enabling people even sometimes without proper accreditation to prove their quality through this accreditation system. To some extent, these are complementary. There are places where they're competing and that's in open academic publishing. Because universities are so traditionalist and so conservative in what counts for promotion and tenure that no, so for example, Harold Varmus led a whole bunch of other people who
were saying we're going to sign the PLOS pledge, the Public Library of Science pledge, we're only going to publish in open source journals. And then, you know, their grad students knocked on the door and said, that's fine for you, you already have your Nobel Prize. I need to publish in Cell and Nature. And then they backed off and they said, no, let's go over here. But, on the other hand, it's a system with a good ethos at baseline. So the internal political economy of universities, once you see the NIH requiring free publication
after 12 months, once you see the real problem generated by open bioinformatics where you You have scientists saying, look, we can't annotate the genome in this particular way. We can't annotate haplomaps thinking that everything will come out as a paper. We've got to find a way of crediting little annotations. And so you're beginning within academia to see primarily in science and primarily still yet in the biological sciences, although that's actually not true because physics was way ahead of everyone with arcs in really distributed systems. and read. But that has to do with the fact that it's, I think, that it's such a small
and highly internally limiting and limited discourse within itself. Whereas bio is much broader audiences. So the answer is, in some cases they're mutually supportive. For open publication itself there's a real tension but there's also political openness. But it will require some political play within the academic system to say, look, if you treat a publication in PLOS Biology the same as you treat it in Cell, then people will go,
open access immediately. One last question. I wanted to go back to the slide where you had a cloud and stuff was produced, because there's been a lot of discussion about motivation, right? And I wondered what you think of one thing that I thought I heard you say that I think is an underexplored motivation, and that's that people do click works instead of watching five minutes of TV or whatever. And I wonder, to what extent are we seeing this as an emergent property of the network, as Moglin says or whatever, simply because as humans, we have
unused processor cycles, much like in the distributed computing. I do spend a good chunk of time talking about motivation, but I do also emphasize that the The 20th century model of industrial culture utilizes our cycles as consumers, as couch potatoes, enormous number of hours per year. And so if you tap some, before you even touch what we do as market actors, there's just
enormous untapped capacity. And unless you have some deep attachment to the idea that sitting in front of a TV is more efficient, more productive, more innovative, more anything than reading and writing blog posts, editing in Wikipedia, or clicking on Mars workers, then that's just a huge source of untapped capacity in our economy that we're just wasting. Well, we're not wasting, but we're thoroughly underutilizing. Anyway, thank you very much.
Generation and audio provenance
- Generated
- 2026-09-09T23:19:43.670896+00:00
- Speech-recognition engine
- mlx-whisper 0.4.3
- Model
- mlx-community/whisper-large-v3-turbo
- Model revision
a4aaeec0636e6fef84abdcbe3544cb2bf7e9f6fb- Model weights SHA-256
951ed3fc1203e6a62467abb2144a96ce7eafca8fa77e3704fdb8635ff3e7f8a6- Original audio file
- Yochai_Benkler_Boalt_27Apr2006.mp3 (46220358 bytes)
- Original audio SHA-256
a60fd0a0dd11512c486f54c30f544c8c8f40dc9a18bc02840cd5128df865ff43- Completion receipt SHA-256
a2d969b1349f4e00da62ce163791fb18be4393ac6540da8d3b05abb292156793- Normalized generated text SHA-256
11bb02cf8b9ffa9c64ba9bcc11e94cd09964f62ff39f29aab8e4cc56412e44a7
Audio stayed local during generation. This provenance identifies the source and process; it does not certify transcription accuracy.
Back to the top of this transcript · Return to the original talk post