 Now, we'll turn the call over to your host today, Michael Thiemann, Vice President of Open Source Affairs at Red Hat. Michael, please go ahead. Thank you very much and welcome to the opensource.com webcast. We are extremely privileged to have Clay Scherke joining us today and talking about his perspectives on open source with his recently published book, Cognitive Surplus. This is a book which I received last year, had a chance to review it, and found that the ideas helped me understand even better the trends and the developments that I have been a firsthand witness to over the last 20-plus years. It's a great opportunity for all of us to hear him talk about this subject and to answer questions. And, Clay, I'm very glad you are joining us. Thank you. Thank you so much, Michael. It's really a pleasure to be here. So, as you said, I want to introduce this sort of basic idea of Cognitive Surplus, specifically with reference to sort of what we've learned from the open source movement, because open source was, in a way, the first great test case of this new resource. After doing that, I want to speculate a little bit about what might come next and about what some of the kinds of issues we face around using the Cognitive Surplus are, because, of course, although we've had this resource 20 years, which is a long time in technical history, it is a short time in social history. And we now face, I think, all of us, the issue of thinking as this network, as the Internet and our ability to coordinate with one another, becomes a more important part of data life. What can we make it do if we put our minds to it? So I will speculate a little bit about that, and I think, you know, questions and conversation, we can take questions and conversation from there. So rather than talk about this stuff in the abstract, I want to just start with a story that I think illustrates the ways in which human society and technological capability are interwoven. And it's the story of my favorite bug report ever. It's the bug report you see on the screen there, slide one. It's a Mozilla bug report. As you can see, it's a bug about password management when different users are on one system, when can they see various forms of password management. And the bug report began with this paragraph. So that's certainly an attention getter right away. And it goes on from there. The original bug report details the way that she went and looked at, she was using the machine after her fiance had logged out. She went in to use the password manager. And it said, oh, here are all of these various sites you have opted not to save your password for. And she found JDate and SwingLifestyle.com and so forth. It was a bunch of stuff that she had not known about her fiance that she discovered through this password flaw. And then she concluded, your browser does not very effectively respect the privacy of different users for one browser. So this bug report was a real bug report. It had obviously an enormously important emotional effect for this user, but it was also a serious technical bug report. And the people in the Mozilla, in the bugzilla form, answered it as a serious question. They went in to try and recreate the bug and so forth. And the conversation rolled on from there. And then you started to get these kinds of responses, where is Firefox installed, trying to elicit from the user the sort of technical details. And then at the end of this interchange in bugzilla, one of the Mozilla Foundation people trying to figure out where the flaw is says, OK, I can't help myself. I also have to tell you, I think you're better off without that cheating fiance of yours. And the conversation went on like this for some time with everybody pouring into this thread doing one or both of two things. One, talking about the technical details of Mozilla. And two, talking to this young woman about her private life and arguing about whether or not she was better off without her fiance. And I love this bug report because I think it illustrates something that is too often overlooked, which is that you can never completely separate the technological and social effects of these tools. They are communications tools, and they have deep technological infrastructure that has to be gotten right in all the ways that we get technology right. But because we pour so much of our lives into them, things like bug reports can ramify in things like breakups. And this is, in my mind, this is kind of a metaphor for what we all deal with all the time, which is that the management of the code base and the management of the community are highly overlapping problems. You can never say, let's get all the technical problems over here and all of the social issues over there, and we'll keep them from affecting one another because we can't do that. Instead, you get this mix of human motivation and human interest and technological capability all kind of stirred together in the same pot. Now, this mixing of human models of motivation are sort of new ways of giving people an incentive to work on something. And the creation of code dates back 25 years, and it dates back in particular to this machine. This is a picture of the most important machine Xerox ever made. It's a Xerox 9700. It's not important because of what the machine did, however. It's important because of where it went. This was the machine that was delivered to Stallman's lab at MIT in the 1980s. You can tell it was the 1980s because of the sexist joke and the ad. But this is the machine that was delivered to Stallman's lab with the source code closed. This is the machine that Stallman wanted to update so that when a print job was done, it would email the person who'd submitted the job to say, your job is finished now. And when he realized that he couldn't, he had the epiphany he's often talked about since, of seeing that this is where software was going, that it was going closed, and that he was going to devote his life as he has done to seeing that there'd be an open alternative instead. And so we have the well-known story of the founding of GNU, the Free Software Foundation, the GPL and so forth. Less well understood in that story, however, is that the GPL mainly failed for the first decade or so of its life, that there was a handful of bits of software produced, you know, famously Emacs from the GCC. But in fact, in terms of a wave of world-transforming software produced in an open model, very little appeared. And in fact, that period, middle of the 80s to middle of the 90s, was the heyday of commercial and closed software. So as noble as the idea was, it was a better idea in theory than it was in practice. And the GPL preamble makes it clear the theoretical bias. This is in fact a political document. It is a document about user freedom. It's not just about the practical implementation of practical details of making code. And in fact, much of the code made under the GPL never got into the hands of many users. And then famously, this happened in August of 1991, just slightly over 20 years ago now. A young Finnish programmer came onto CompOS Minix and said, hey, I've got this idea. I want to make this hobby operating system. He says sort of hilariously in the first message, this isn't going to be a fancy effort like GNU. But this is the birth of Linux. And this is the beginning of a practical way of implementing some of the ideals in the GPL. But what happened to the GPL in the first few years of its life is that the theory was in place, but the practice of coding was still very much the traditional get a small group, have a leader work together in a kind of tight circle. And what Linus did instead is he said, I'm going to go to the place where people who care about this hang out and I'm going to invite anyone who wants to kick in to kick in. Now, Linus famously in this thread only got about a half a dozen people interested. But all that had to happen was that the work of that, Linus and that half a dozen, had to attract another half dozen people. And that dozen people, all they had to do was attract another dozen and so on. And slowly, and of course, with Michael Teemann on the phone, I don't have to tell this story much, slowly Linux went on to expand until it not only became a viable operating system, but made things like the cloud inconceivable without it. You cannot imagine the cloud existing if you didn't have an operating system built on these principles. And what the practical implementation and the choice of the GPL for Linux, but the practical implementation of this way of working, was in many ways what the original goals of the Free Software Foundation were waiting for. And as Eric Raymond has said, better than anybody else in the Cathedral in the Bazaar, Linux was the first project for which a conscious and successful effort to use the entire world as its talent pool was made. And that observation, that observation that you could for the first time and thanks to not the technological parts of the Internet, but the social parts of the Internet, you could find all the people who cared about Linux like operating systems. And you could address them as a group and you could recruit them as a group. That capability absent any other change altered the way in which software could be produced. And that change, I think, is what is so important, I think for that the open source world has understood for two decades now, is what's so important for the rest of the world to understand. So the question I was asking myself in the book that Michael was alluding to, Cognitive Surplus, is what is Linux made of? And the answer, of course, is that Linux is made of code like any other project. And it gets compiled like any other project and it runs on hardware like any other project. But that answer doesn't explain the difference between, say, Linux and Microsoft Windows. So my answer is that Linux is made of Cognitive Surplus. Linux is made of the free time and talents and committable capabilities of anybody who understands enough about operating systems to offer some addition to the project. And over time, that group has grown to be very, very large, right? So the Cognitive Surplus is made of two things. It's made of free time and talent, right? It's made of the fact that someone who wants to kick in a single bug report can do so and that someone who wants to sit down and work on kernel internals for the rest of their lives can do so. But that we have this group of people who have knowledge and characteristics that would allow them to participate. And two, critically, as Linux relied on, we have a network that makes finding those people and talking to them trivially easy and at zero marginal cost. And those two things together, the presence of free time and talents, which we've had for some time as a society, and a medium that allows for group aggregation at low cost means that these large collaborative projects are now possible in a way that they weren't in most of the last century. So having posited that resource, having posited the idea of a cognitive surplus of a pool of talent that could be taken advantage of in new ways by this network, I needed some way to talk about how big it was and not having an easy metric to hand I invented one. So I started with Wikipedia, which is the collaborative voluntary project familiar to most readers. And with my friend Martin Wattenberg, we did some back-of-the-envelope calculations and figured that by 2008, Wikipedia had taken something like 100 million hours of human thought to create. It's an order of magnitude calculation, but it's the right order of magnitude, 100 million hours. So that's obviously an enormous project. Anything in human history that's taken 100 million hours of time and effort is quite extraordinarily large. But how big is it relative to the overall amount of free time in the world? So I needed another thing to compare it to, and I chose television, I chose television, which is the number one use of free time worldwide. So imagine a graphic representation of television. Imagine that white rectangle as the amount of television time watch. So Wikipedia is 100 million hours. How big is that box? And the answer is it's greater than a trillion hours every year. Rendered graphically, Wikipedia is there. The amount of free time spent watching television is such a vast super set of the entire 10-year effort on Wikipedia that you can barely see it represented in the corner. So this is an enormous potential resource. And the fact that it's an enormous potential resource means that it only takes a very small number of people, considered as a percentage of the connected population, donating a very small amount of their time to creative projects to lead to enormous changes in the outcome. And that change, I think, is what we're grappling with as a society. The open source world, as I said, has gotten there first. But the attitude towards problem solving is now becoming general purpose and remarkably powerful. Just last year, DARPA launched something called the Red Balloon Challenge. And they were interested in studying collaborative dynamics. They wanted to know how people would work together to solve a problem. So they put up 10 red-weather balloons all around the country, relatively visible from public roadways and walkways and so forth. But they didn't announce their locations. And they said, essentially, the challenge is who in the next 30 days can find all 10 red balloons and tell us their location within a mile of their actual lat-long? Who can tell us where these 10 red balloons are? So several different teams came up with several different strategies to take this problem on. And there was a team at MIT that said, oh, hey, we know how to do this. We'll just pay people to tell us the answer. If you see a red balloon and you tell us where it is, we'll get the prize money, and then we'll subdivide it with you. So this was fairly obvious way of approaching the problem. But it didn't give that message any reason to spread, right? Because it's easy enough to say, if you know where a red balloon is, you get some money if you tell us. It's hard, however, to let people know that that offer is actually on the table. So then what they said was, if you tell us where the red balloon is, we'll split the money with you. But if you introduce us to the person who tells us where the red balloon is, we'll give them some money and we'll give you some money. And now there was a reason to spread the message because the recruiting logic suddenly meant that this message could spread like wildfire through existing social networks in order to take advantage of the overall observational surplus, if you will, of the United States. So the first interesting thing about that strategy is that it worked. The MIT team was the winning team on the red balloon challenge. The second interesting thing on the strategy in this column here on the number of balloons, the second interesting thing about that strategy is the MIT team was the only team to correctly identify all 10 balloons. But the third interesting thing about that strategy is that DARPA had allotted 30 days for this challenge to unfold. They said, you have basically a month to tell us where these red balloons are. And the MIT team solved the problem in nine hours, which is to say DARPA had overestimated the amount of time this problem could conceivably, you know, could take by a factor of 80. So this is the kind of power, I think, that comes from successfully harnessing these kinds of surpluses, which is that the problem-solving capabilities, if you frame the problem right and describe the problem to people in a way that they would understand why they would wanna participate, you can deploy incredible tools for solving problems like this in ways that we could not have done, we could not easily have done previously. And taking advantage of this cognitive surplus means that we'll start to see incredible social, artistic, literary, and political uses of these new tools. We will also see, of course, lots of lolcats, which are cute cats made cuter with the addition of cute captions. And lolcats, too, are a side effect of the cognitive surplus, right? It isn't just high-minded problem-solving and the creation of world-class software and the creation and distribution of large-scale reference works. You also get a lot of this stuff. You get a lot of silly culture. And the people who are disputing the hypothesis that the internet might be good for society as a cultural force often point to this sort of thing and say, really? This is the medium you're talking about? This thing with the funny cats? Like, that's the thing that's gonna make society better? And it's certainly true that a big chunk of people's free time goes to cracking each other up at work and so on, that these are the kinds of things that get created. But here's the thing, that always happens. And I don't mean it always happens with the internet. I mean it always happens with new media in general. So if you look at the early history of the printing press, it did not take people very long after the invention of movable type to produce the first erotic novel, right? Once publishers had set themselves up in the business of making a profit off printing presses, it did not take long for someone to say, hey, you know what I bet people would pay for? It took another 150 years for people to even think of the scientific journal. So erotic novel was the kind of thing we got because we're the kind of people we are and the printing press is the kind of tool it is. But the scientific journal took inventing, it took a group of people committed to making the world better to figure out how to use that particular medium to achieve that particular outcome. Long cats are the kind of thing we get because we're the kind of people we are and the web is the kind of medium it is. But when you look at things like Linux or like Wikipedia, what you see is they didn't just happen, they came together because a small group of people committed themselves to using these tools in a way that would let them accomplish something more than just cracking each other up at work, right? So the question then comes about, what kind of thing can we expect from this particular media revolution? We have some sense of what happened with early media revolutions. What kinds of things can we expect if we have this participatory medium that is now spread worldwide, right? And when we look back at previous media revolutions, we can see a sort of a common set of predictions. There were predictions made about the rise of the telegraph, as you can see. There were similar predictions made about the rise of the telephone, where the ability to communicate worldwide was going to lead to desirable outcomes. There were similar predictions made about the spread of television, right? Where the ability to see how other people on the planet lived were going to create this value. And the commonality to all of these predictions, of course, is that they were all completely wrong. Every time, every time a new tool comes along, someone will line up to predict an imminent outbreak of world peace, right? Once we can all communicate with each other, then we will all see eye to eye and then we will all finally exist in the brotherhood and sisterhood of man. Never happens. What happens when we get new media is we get lots and lots and lots of arguing. We get more arguing after the new medium than we had before. So, back to the scientific journal. The scientific journal was an example of this increase in arguing, not an increase in agreement. That prior to the printing press, the attempt to understand matter, which was conducted under the rough heading of alchemy, was a private activity in which one or a small group of people generally sworn to secrecy would do their work and document their results in ways that are incredibly difficult to find. And if you found them, they were incredibly difficult to read. And if you read them, they were incredibly difficult to interpret. And what the people who invented chemistry did was they took that practice and they changed it. Many of the early chemists were alchemists. They used the same tools that alchemists use. They worked in the same laboratories that alchemists worked. There was no practical difference between alchemy and chemistry in terms of the ability to manage the day-to-day life of working with chemicals. The big difference is that they all agreed to publish the results in language that they could all understand and in ways that would let any of them recreate the others' experiments. It was the birth of peer review. So the printing press didn't lead to more agreement. It led to better arguments. And this is over and over the pattern we see with successful integration of media, which is that better kinds of arguing, and anyone who's been in an open-source community has seen that highly structured argumentation is actually how those projects progress, that better kinds of arguing turn out to be the thing that makes a particular medium particularly useful. So when you take a look at what is true of successful uses of large-scale collaboration, one of the biggest differences you can see in the participatory logic is how incredibly wide it is, right? That this is in fact not just a new tool for the same old people to do things in a new way. It's actually a way of inviting an enormous number of new people into the process, which is why I think the Cognitive Surplus becomes a way of imagining what's going on precisely because of this inclusion of so many more people than had previously participated. If you look, for example, at the article on Linux on Wikipedia and you ask yourself who has written it, the answer is there have been something like 10,000 edits by not quite 5,000 users. So it's about two and a half edits a person. And so the answer is 10,000 people have written the Linux article on Wikipedia. And it seems when you look at it this way, like all those people are coming in, everybody's kicking in two or three edits and the whole article kind of takes shape that way. When you look at the actual underlying distribution of participation, however, all of that idea of two and a half edits goes away. That's a mathematical artifact of the system, but it doesn't say anything about the particular user. If you graph user participation, edits per user in rank order, you can barely see the red line, it hugs the X and Y axis so closely. There are a couple of users who've done a couple hundred edits each. And then there are an enormous number of users. In fact, the most active users done almost 500 edits on their own. The 10th most active is done over 100. And then there's this enormous long tail. By the time you're down to the 100th most active user, only 10 edits. And then you have this huge tail of 3,000 of the 4,000 some odd users who've done only one edit each. And this is also the distribution of tags on Flickr. This is the distribution of check-ins. The Linux kernel is a very common pattern called the parallel distribution. But what it says is the ability to bring multiple people into the participatory environment to let multiple people participate in this project doesn't mean that they're all gonna come in and think of themselves as active collaborators. In fact, in a Wikipedia article, as in so many kinds of large collaborative projects, the argument that goes on, the people making the key decisions or the people making hundreds or in some, in some handful of cases, thousands of edits or commits or what have you. And the large bulk of participants are just there to say or do one or two things and have that work integrated. So to say that these tools tend to improve argumentation doesn't mean that they make many, many more people argue. In fact, what it does is it concentrates the argument to that handful of people most committed to the project while letting a huge army of people contribute in ways that are useful, but that don't bog the project down. And this, the style of management that's required for this kind of participation is so radically different from the style of management that's required in a traditional commercial software house. That is, I think, the biggest difference. It's not ultimately what the source code looks like, rather the process by which the source code comes together that is open source is great, great contribution to the world. So if we assume this, if we assume that these tools are there, that the effect of these tools is to bring more people into the environment and to allow the core most committed of them to argue their way to success, then what are the open questions for the environment we're heading into? And I will end here with an observation that I think is profound and is, for me, I think now the core question for a lot of these large-scale technical communities. And the observation can, I think, best be illustrated with a game invented by a philosopher named Peter Subur. The game's called NOMIC. And he wanted a game which would show his students how self-modifying systems work. So NOMIC is a game where the rules of the game can be changed by the players. In fact, changing the rules is one of the moves in the game. And NOMIC starts out with a very complicated set of rules. I've just reproduced it in a stylized fashion here. And there are A-level rules and B-level rules. A-level rules can't be changed and B-level rules can be changed. So this seems originally to be a system that has some fixed points and some flexible points. But Subur's essential contribution to observations about self-modifying systems is that if you make the rule that says A-level rules can't be changed, if you make that rule a B-level rule, then the first thing you can do is you can change that rule and then you can change the A rules afterwards. Which is to say, in the design of collaborative systems, in the way that we think about bringing people together to collaborate, it's actually possible to have a system in which all elements of the system are open to change, but some elements are harder to change than others. And if you think of the US Constitution, it is much easier to pass a law than it is to amend the Constitution. But it is possible to amend the Constitution. And in fact, one of the theoretical amendments, you could amend the Constitution to make it easier to amend. So I think the great open question for these large-scale participatory communities, particularly around technologically convergent conversations like software where the source code has to be a particular thing before it hits the compiler, is this. We had 25 years of free software and open source. And in most of that time, for most projects, there has been a charismatic leader or at least a leader with obvious final say at the helm, whether we're talking about Stallman or we're talking about Torvald, Guido Van Rossen, with Python, whose name has been Everland dictator for life, Larry Wall and Pearl, on and on the list goes even Jimmy Wales as a kind of eminence gris of Wikipedia. And yet we all hope that these projects outlast their founders, which means at some point that we have the problem of the succession crisis. How do you move from a single leader who owns the project to hand over control of the direction to a group? Now in many cases, the kind of open projects we look at have already had that happen to some degree. Linus is certainly less involved in all of the details simply because he can't be, the project is so enormous that there's a higher degree of delegation. On Wikipedia, an enormous amount of control has been handed over to the users already in various forms, like the arbitration committee. In some cases, like the Apache Foundation, it has been set up explicitly with an elected group rather than a single leader. But even given those early experiments, the problem we all face and the opportunity we all have, analogous to the founding of scientific journals as a way of harnessing the printing press, is how do we use this medium, not just to share source code and argue about how it should be before we agree what it is we're going to do. But how do we use this medium to actually articulate the social agreement among a group of people so that they can have profitable arguments with one another, so that they can hash out their differences and come to some kind of either agreed-on best model or at least a compromise? And as I was saying to Michael in the conversation before this call started, what I think this really means is that the open source movement is in many ways the experimental wing of political philosophy. That the question of how a group of people should arrange their affairs so that they can have the most profitable and useful arguments amongst themselves is not itself a technical question. It will take some technology to solve it, but the real question is, how do we take these tools and press them into service so that we can take advantage of the cognitive surplus in a way that is long-term, large-scale and ongoing rather than just doing it on a short-term and project-by-project basis? And there I'll end and hope or assume that we can have questions in conversation about that. So Clay, I want to thank you very much. Obviously, as a professor, you have given us an entire semester's worth of ideas in one single session. So in some ways, I've already envisioned the next six webcasts that we could do on these subjects. So thank you very much. Sure, of course. And I'd also like to just make the observation that I think that at the end of the day, you are more of a crocheter than a weaver because you did such a... Actually, bringing the whole discussion full circle, I had prepared a chain of things that would have completely popped the stack back to the beginning, but you circled the square. You could never pop the stack too many times, so. So, but let me just pick up a couple of things while we wait for questions to accumulate. And I want to start with the famous quote from Gandhi, which was, whatever you do will be insignificant, but it is very important that you do it. Nice. I use that quote every time I give higher orientation to the red hat, and I use that quote to explain how the individuals at the far right-hand side of the power law actually create the power of open source. Right, right, right, absolutely. One of the things that, and this is not so true anymore, but one of the things that used to be hard to explain about open source was the idea that the 80-20 optimization, beloved of management from, since time immemorial, is actually not a universally good idea, right? And so everybody looks at this, oh, 80% of the books sell 20% of the copies, and 20% of the books sell 80% of the copies and so forth and so on. And when you look at the right-hand side of the power law distribution, if you're looking at it with an employee-employer mindset, and you think, oh, we hired that guy and he hung out in the break room and drank our coax all the time, and he only ever committed one thing to the code base, you're thinking, you know, bad hire, right? But if that one thing was, say, a patch for a buffer overflow error, then you might actually wanna take that code even if you never hear from that person again. Well, but I would, I would, I would recap that a little bit and say that if the guy inside drinking the coax has relationships with 10,000 people who can make that one change, then you made a good hire. And so there are, yeah, there are all kinds of ways in which the value doesn't show up, can exist without showing up directly in the commit, in the commit list. In fact, one of the funny things Apache has started to do is mine its mailing list to find people who are active on the project in terms of, you know, cat herding, but are not themselves having their name on the change log. But the key thing I think to understand about this sort of individual contributions is 80-20 optimizations are a function of economics. Why give up the contribution, the individual contributions of individual committers if you don't have to? So in a way, the commercial model for creating software makes 80-20 optimization seem like a good idea only because it forestalls taking in those individual contributions in the first place. Exactly, 80-20 is for the people who have not read and understood your book. Well, you're sweet to say so, but also free or the people who have not hung out on, you know, a Linux mailing list or watched a Wiki work for a while. And I think that business of understanding, you know, exactly what you said, I've not heard the guy quote using that way, but that is a brilliant way to illustrate the issue, which is it is partly cumulative, right? It's partly about adding up all these little contributions that you want, but it's also partly cultural. It says we are open to a good idea wherever it comes from and whether or not we have a long-term or a short-term relationship with a person who had the good idea. And once you can say yes to those two things, then you have a degree of power and leverage that you don't get when you say, this is a closed shop, if we don't know you, you can't participate. Right, so there's a question that came in from Bruce Novell saying group collaboration seem also to benefit individual contributors as well as a larger population of all potential users or society at large. What do you think? And I think both of us agree with that statement, but I think we may have a different angle as to why that is such an important observation. So I'll let you take the first swing at it. Well, so, you know, I think a lot of it is sort of what we were talking about before, which is that there's a cultural advantage to having a group of people think that their contribution could matter. Linus has often said it takes about a year and a half for someone working on the kernel project to reliably get their patches accepted. Just that is the learning curve and it's a very, very steep learning curve. But when that person gets their patches accepted, they have become a much better programmer than when they started and the community's made stronger by having those kinds of programmers work on this. So in a way, the kind of worship of the individual, the individual at the helm of the enterprise, understates the degree to which if you don't have a culture that produces good work, that individual can't do much by themselves. We've always had, you know, whether it's Tom's Edison or Steve Jobs, we've always had people who are credited by the population at large with the work that's actually mainly done by their employees. But I think it's really important to understand that the culture underneath a piece of technology, the culture of the users matters as much to the long-term viability of the technology as does the value of any given iteration of the technology itself. I mean, Linus says this a lot that actually managing the user basis is, you know, both kind of harder and more important than managing the code base. Yeah, so let me just share some of the wisdom that I've received. James Herbslip from CMU did an analysis of the Apache project among many things which not only won the best ATM paper on software engineering in the year 2000, but also the eventual paper of the last decade in 2010, and one of the observations is that the contribution from the long tail, the availability of an infinite number of individual patches which correct errors creates a systemic level of quality that simply cannot be achieved by the manager who basically says, that's a one-off, we can ignore it. And in fact, I think it's interesting. Independent software quality metrics of both initially developed by industry and later funded and evaluated by the Department of Homeland Security find that open source software has achieved 100 times better, i.e., lower defect densities and proprietary software and makes double digit improvement in its average software quality every year compared with proprietary software, which is as bad as it ever was. That's the subject that we discussed. That is, yep. We, there's the opensource.com reference to that is called integral innovation and it talks about the importance of the incrementality of innovation as opposed to the giant leap. Right, right. Well, we were talking about this before the call. I think Stephen Johnson's most recent book on innovation is really significant for saying the whole argument about public versus private support for innovation misses the fact that the real, the kind of deeply important dynamic is between solo and network innovation. And in fact, network innovation, the ability of people who are working on a project to talk to people who are working on related or in some cases, even seemingly unrelated ideas that nevertheless create some kind of spark or synthesis. That turns out to be more important than either the source of funding or the model of creativity you're using. Yes, there is. Open to the network is the hallmark of the, you know, openness to other people's ideas is the hallmark of the kind of open source model of listening for, you know, listening for good ideas. A fantastic example of what is network innovation is the recently published, the accidentally published rant of a Google employee. And I read this article yesterday about a Googler explaining why Google Plus was a failure and how to understand the success of Amazon. And you could read that article and gain a tremendous insight into these concepts of network innovation. But I promise you that if you appreciate the thesis of cognitive surplus and how it applies not only to super big companies like Amazon.com, Facebook and Google, but can apply to socially interesting projects, whether they are history or literature or criticism or in a recent case, the discovery of molecular models for some important classes of drug treatments. There's a project which was, there was a great challenge that had stumped scientists for 10 years and somebody cleverly turned the nature of that problem into a game and game found the problem faster, like in 30 days, compared with scientists who with all their funding could not find the solution in 10 years. And so if you read that Google rant, you will see aspects of this problem, but if you also do that with the understanding of cognitive surplus, you can see how transformative this network innovation can be. So, let's see, we've got a bunch of questions queued up and let me give you a chance to pick the next question and respond to it. I'm sorry, I'm not sure that I can, hang on a moment. The Q and A section, all right. Well, so actually- Yeah, so why don't you pick one because I'm not sure I've got my interface set up right to do that. Let me pick a good one. So, Rebecca asks, it seems like the line between leisure and work has blurred, for example, scientists who contribute to Wikipedia article in their spare time or doctors who contribute to websites. Why is this happening? Right, well, I think the question has two parts. The short answer, the near-term answer is, because we like it, one of the things that happened with the rise of the kind of individual actors of the core economic unit, the sort of neoclassical model of economics is we assumed that all motivation was personal and we assumed that it was largely optimizing, right? I was doing things in order to get more of whatever it was I wanted. And so when I worked, I was building up money to spend and then when I went out in the world, I was then spending that money on whatever I was interested in. And what now 40 years of behavioral economics and various branches of psychology have shown is that people have deep intrinsic motivations, motivations to do things that make them happy because they're doing them, that don't fit well into that personal, rational, optimizer model. And that there are some of those things that are intrinsic and personal. I wanna feel autonomous, I wanna feel competent. But that there's actually an enormous collection of social motivations that make us happy in ways that cannot be resolved to the standard analysis of you have a job and you have a salary and you get the money and so on. So if I do something that gets me recognized for being a good contributor to a community, I can feel a sense of membership and a sense of generosity that literally cannot be bought on the open market at any price. And so a doctor or somebody who makes architectural models out of Lego or somebody who has a carpooling opportunity that they can offer any of these examples involve people getting positive social feedback in ways that can't be made coherent if your tool of analysis is simply a question of work and salary versus free time. The other longer term thing to observe is that the separation of the social domain into work and leisure is itself a relatively new phenomenon. And that if you look at society essentially prior to the idea that you had to go to the factory because that was where the steam engine was, what you see is that the so-called cottage industries had work and life intermixed fairly coherently. So I think the idea that the mixing up of work and life is somehow a historical novelty is true if you look back decades, but it's not true if you look back centuries. So you promised that one of the takeaway messages from this whole talk would be the idea that technology and the social environment cannot be separated. And actually, one of the questions that you raised which made me write a note, when you talk about the importance of the charismatic leader and whether or not we are evolving beyond the Lina's Torbols and the Larry Walls and so forth and so on or not, I would argue that one of the functions of the charismatic leader is to represent the trust of the community. And you can appreciate the quality of the human who stands at the top of that particular domain and you appreciate their values, you appreciate their integrity, you appreciate their sense of humor, such as Lina's Torbols, all they're calling himself an asshole. If you appreciate this charismatic, that is the conductive wire that retains. It's the way that when a conversation deadlocks, if you've got somebody to come in and say, at this point, given that the technical argument to have it on the day, somebody just needs to make a decision and that someone is me, you have a way to go on. You know what, I did find the QA window. I also would like to address the question Guy Martin has asked, which I think is a very good one. When I say committing free time, I don't mean committing time away from work. I'm sorry, and the question is about why is it that such an enormous number of contributions to Linux kernel have come from organizations that pay their engineers to contribute, whether it's IBM or Novell or what have you. When I say free time, I don't mean free in the sense of now I'm off work and I can kick back my heels. I mean free in the sense of it is up to the individual to decide where that time will be committed. So the big difference between IBM paying someone to work on Linux and Microsoft paying someone to work on Windows is that IBM thinks it will be good for our business if this person works on that piece of software, which is a classic commercial calculation. But IBM cannot direct that person's work to the point where they guarantee it will be accepted into the kernel, nor can they expropriate the results and take the resulting revenues. And that is I think pretty extraordinary in the history of market economies. That it would be worth paying someone to contribute to a project that you can't control and don't own. And I think that IBM's shift to support for open source says how much it recognizes the positive ecosystem dynamics that it benefits from, that it will give up the two classic prerogatives of management. I will tell you what to do and my company will capture 100% of the value of the results and is still willing to pay people because the overall value of the ecosystem is high enough to them that it is worth it. In a sense of arbitrage, it is worth buying progress and selling exclusivity. That's interesting, yes. So we have a very short period of time left, but we've got these protests going on now in Wall Street and somebody has somehow attached the label of open source that somehow the social phenomenon of people bringing their complaints to the bankers who brought down their world has something to do with open source and clearly there are a lot of college educated people there with a tremendous amount of cognitive surplus and enormous amount of incurable financial debt. What are your, and you did mention this philosopher as your last slide as a potential way of understanding democracy. Can you tell us how this open source and cognitive surplus model could have bettered them? Well, so the Occupy Wall Street protests are very interesting because they are in a way an almost literal embodiment of this surplus. There is an army of the unemployed in every city in this country, but they need these tools in order to be able to take coordinated action. And the kind of extraordinary thing about the Occupy Wall Street movement is that they have hit, at least in the New York camp which is the only one I've visited down at Zuccotti Park, but their technological tools of choice are not tools designed for left wing dissent and they're not tools designed for political engagement. It's tumbler and paste bin. So you've got one tool that's for 13 year old girls to trade pictures of cute haircuts and another tool that's for sharing source code among programmers and those two tools because they worked for other groups have been adapted by OWS for coordinating and spreading their message. So I think one of the big lessons of OWS is that once you get the social pattern of coordination and participation rate, the number of different ways in which it can be pressed into service is much larger than the number of ways for which the tool was originally optimized. The other thing I will say about open source is that the idea of open source has been so powerful that it has jumped the banks and it has now become essentially the brand name for a certain attitude towards participation. It's not at all clear, in fact, I think it is clear that it is not the case that the Occupy Wall Street protests can be looked on as an open source project in the kind of classic way. Open source requires a kind of focused agreement simply because there's no arguing with the compiler. And so there's a kind of a non-human member of every open source community that gets a vote, something either compiles or it doesn't. Occupy Wall Street has a much looser set of goals but I think the tribute they pay by calling themselves an open source movement is that they recognize that there are other places in the world where large-scale voluntary and participatory models have yielded tremendous results and that's what they hope to create. So earlier you referenced that the medium is the message and perhaps the way of understanding not just Occupy Wall Street and not just open source but also cognitive surplus is that the community is the platform. And we unfortunately did not really have a lot of time to get into the details of the architecture and the true potential energy of the platform but when we look at, as an example, the Google rant, when we look at something at Sam Ramsey from Microsoft recently talked about, he basically said cloud is impossible without open source. Yes, oh yes, absolutely. Inconceivable that every instance of an operating system be paid for and that you could have anything worth calling a cloud. And I think that what we had 100 years ago the prejudice was the madness of crowds. The prejudice of the past was that you bring together enough people and everything falls to the lowest common denominator and I think that what we have seen is that with the right technologies and with the right social rules of engagement and technological and network interfaces it is in fact possible to build this best of all possible world of real forward progress and the problem with democracy at least in America is that we have this 200 plus year old device which was fantastic for a time and does permit self modification but one of the slides you talked to, the rule that every turn any user can propose a new rule the rules of Congress today do not permit any member of Congress from proposing a new rule because the majority party controls the agenda and their absolute ability to close any constructive commentary from the minority has led to kind of deadlock and gridlock whereas in the world of open source we have the freedom to fork and we have the freedom to build unlikely bridges as you mentioned, adapting Tumblr and Payspin to a major financial protest and I think that as this new digital generation comes forward and recognizes the enormous value that their cognitive surplus represents the payoff of this enormous investment that society itself has made in educating so many people and as your book explains right now, we are in the tenth and hundredth of the decimal point of percentage application of the surplus so we've got a factor of 1,000, a factor of 10,000, a factor of 100,000 more productivity to look for if we can engage this correctly. Are we... I think we're at the end of the hour, yeah. I know people have got the other appointments so Claire, I want to thank you so much for your participation. Thank you, Michael, it's really been a pleasure. I will send you some private email to help improve your history of the GPL. But generally, this is absolutely fantastic and to those who participated and asked questions you can look forward to many conversations about Cognitive Surplus on opensource.com. Great. Thank you. Thank you.