Here's something you might enjoy reading or watching, a talk given by Eben Moglen on June 26th to the Scottish Society for Computers and Law, "The Global Software Industry in Transformation: After GPLv3".
You can read it and also get it as streaming video. The theme is the connection between GPLv3, mathematics, and the sharing of human knowledge. As a jumping off point, he asks us to imagine a world in which arithmetic has become property. Even stating that diminishes it, actually. Read it for yourself. You'll enjoy thinking about it.
If you think he's exaggerating about GPLv3 being a product of a broad consensus, here's the statement by the Securities Industry and Financial Markets Association (SIFMA), thanking the Free Software Foundation for considering the views of SIFMA members when drafting GPLv3. Here's a report by Sean Michael Kerner of Internet News on the Richard Stallman remarks, for those who missed the streaming video. And here's attorney Mark Radcliffe:
"The Free Software Foundation listened to people outside its normal
support base. The GPL 3 is better than the GPL 2," said Mark Radcliffe,
an intellectual property attorney with DLA Piper's Silicon Valley
office, during the Linux Foundation Summit.
Also, I've updated Groklaw's GPL Resources page. If it's missing anything, please let me know.
The Global Software Industry in Transformation: After GPLv3
Good evening everybody, and welcome to the Scottish Society for Computers and Law Annual Lecture for 2007.
And it's, I think, a testament to the drawing power of our speaker this evening that we've got such a great turnout and there's an enduring interest in the subject, of course, and so we're very, very pleased to see not only lawyers, but real people as well.
Eben Moglen is a man who, for most of this audience, really requires no introduction.
Eben is someone who, for a very long time, has taken an interest in Free Software, and more broadly in the whole issue of freedom of information, freedom of artistic endeavor, and he brings to his subject a unique challenging and positively exciting perspective.
He's enthusiastic, I hope you'll agree with me by the end of this evening that he's very persuasive in his subject, and so persuasive indeed that as well as being a Professor at Columbia Law School in New York, he is also the founder of the Software Freedom Law Center. Software Freedom Law Center is something, a creature which I think is relatively unknown in these islands; it's a law firm which is also a charity, and it is seriously committed to public education.
He's going to talk to us tonight, I think, about the GPL 3, the General Public License number 3, which he has been instrumental in drafting and preparing, and taking it through a long gestation to its birth, which is expected any day now. So, I think there's little more for me to say, other than just to ask you to welcome Professor Eben Moglen.
Thank you. It's an honor to be here. I want to thank Ian Mitchell and his colleagues at the Society for Computers and Law, and I want to thank all of you for being willing to take a lovely evening to talk about a copyright license on software.
I'll try to make it worthwhile by going slightly beyond the bounds of the discussion of a single copyright license on software, but Ian is right in suggesting that after 18 months of GPL3, I find I'm almost unable to at least begin talking anywhere else.
Nonetheless, I think just so we have a common basis for consideration of what's about to happen, it would be good to take a little bit of a look about where we have been.
What we're talking about are some legal rules that assist in the construction of communities that share knowledge.
The particular knowledge being shared is knowledge about how to make computers do neat and useful things, but that process, that is constructing legal rules to facilitate communities that share, is a process which goes far beyond the sharing of information about how to make computers do neat and useful things. There are reasons that computer software was the first domain in which the process of sharing knowledge inside communities enabled by technology arose.
Communities enabled by technology arose in the neighborhood of the technology that made them possible, that is to say in the neighborhood of digital communications, and digital communications technology is software in action, and so it's perhaps not terribly surprising that the initial explorations of the possibility of sharing knowledge at low friction occurred in the neighborhood of the software that made digital communities possible. But we should unroll that activity a little bit more carefully from its beginning in order to understand the principles that are now transforming the global economy.
Software knowledge, that is knowledge about how computers can be made to interact productively with human beings, can be very closely -- I don't mean from a legal point of view, but from a social point of view -- very closely correlated with the traditional social purposes of mathematics, by which I mean mostly arithmetic, trigonometry, and simple calculus.
Mathematics is primarily a language for ensuring reliable results in human-social interactivity. I say this, notwithstanding the fact that some mathematicians would object that mathematics is primarily a device for creating beauty, which is of course true, pursued by its greatest devotees at its highest level of skill and daring. But the majority of mathematics is not a thing of beauty in itself, or at any rate not a novel thing of beauty in itself, it's a boat. Or a house, or a pyramid, or a tomb, or in fact almost any other activity of human material collaboration, enabled by mathematics in the sense that mathematics as human beings have learned to use it enabled them to achieve results that were reliable, reproducible, and certain.
Economic and safety-regarding activity in the material universe collaborating among people is extraordinarily difficult to achieve without adequate quantities of mathematics, so I ask you to imagine briefly a world in which arithmetic has become property.
Before beginning almost any useful socially collaborative activity in the material world, you are compelled to begin by a stop at the arithmetic store, pretty much the way you are presently compelled to begin with a stop at the petrol station, right?
It becomes a critical input to almost all other meaningful economic activity to have a sufficient quantity of arithmetic to complete the task. Without sufficient quantities of mathematics, bridges do not stand up reliably when the wind blows, railroad schedules are merely toilet paper, and all the other various forms of material-social collaboration which require large numbers of people to predict the behavior of material for forces and groups of humans become largely inoperable.
Once we have reduced arithmetic to property, you'll have only as much arithmetic as you can afford, the consequence of which is that the gateways to material collaboration in the world, successful activity in relation to the physical and constructed environment, will depend very largely upon one's ability to acquire sufficient surplus amounts of mathematics.
Most people will be compelled to a subsistence-level interaction with mathematics. Other peoples' mathematics will control them, but they will not be able to accumulate a sufficient surplus of mathematics to become enabled, autonomous actors in the creation of social collaboration.
Not with respect to mathematics, but with respect to its very close cousin physics, the sixteenth century European social order brought this subject very firmly to the top, and with the name Galileo Galilei, we associate two of the most important cultural responses to the quandary of possessed physics.
The first is an insistence upon freedom from censorship, that is "e pur si muove" -- determination to prohibit the ownership of physics by an entity rich enough and powerful enough to define its physics as the only permissible physics, the only available physics, for most ordinary people. And second, the first significant attempt in the history of the West to write scientific literature at the state of the art in a vernacular language, accessible to everyone.
Galileo Galilei's decision to publish in Italian is as important as his decision to risk confrontation with the Church, for what it says about the fundamental pillars of free science in the history of the West. Not merely, in other words, an insistence upon the freedom of ideas to work their will in skilled hands, but a determination that the ideas which motivate the world, which explain its behavior and which render it controllable, should be universally accessible to people regardless of their ability to acquire enough social surplus to have Latin.
We have come, at the end of the 20th and the beginning of the 21st centuries, to an equivalently important moment in the history of human civilization, a moment at which the principle of the universalization of free knowledge becomes, for technical reasons, universally fulfillable. Where it becomes, for technical reasons, possible for the first time in the history of human beings to bring all useful and beautiful knowledge to everybody without regard to the ability to pay. We are, to be sure, at a minimum a generation from the achievement of that goal, but we have never in the history of human beings been within one generation of the achievement of that goal before.
The principal social alteration which brings about this epochal change in the nature of human society is the digitization of knowledge. the onset of a frictionless mechanism for storing and forwarding information, for switching it in any direction that it is desired to go, by either of the endpoints in any point-to-point transaction for knowledge. We have now learned, at the end of the 20th and beginning of the 21st centuries, how to use the common property in the electromagnetic spectrum and common physical materials, cheaply integratable into mechanical and electro-mechanical devices, to spread knowledge infinitely wide and infinitely thin.
We can produce anything of value, utility, or beauty that can be represented by a bitstream, which increasingly means all beautiful and useful human activity, anywhere, at any time, to anyone, at no more cost than the fixed cost which created the first copy of the relevant bitstream. That fixed cost may be, under certain circumstances, very substantial. There is no question that it continues to cost money to acquire knowledge and to represent it in beautiful and useful ways.
But what has changed is that the marginal cost of the additional copy of each bitstream has gone to zero, and with that change fundamental economic reordering begins in global society. By the end of the first quarter of the 21st century, almost everything which it has been in the past the purpose of industrial civilization to put into analog representations of information -- music, video, art, useful information concerning the operation of the physical environment, political ideas, comedy, drama -- will all be universally represented in de-physicalized forms that it costs nothing to make, move, and deliver.
The consequence of those changes is the onset of a very powerful moral question. If it is possible, easily possible, to give to each human being who wishes it, anything of utility or beauty in our world of civilization, if it is possible to deliver any such entity anywhere at any time at low cost or at zero cost, why is it ever moral to exclude anyone from anything she wants?
Why is it ever moral to deprive people of that which they could have for nothing and which they wish to have, and you already have made? If you could feed everyone by baking one loaf of bread and pressing a button, what would be the moral case for permitting the price of bread to be higher than the poorest hungry person could pay?
The moral quandary of the validity of exclusion from knowledge has heretofore not been particularly sharp. Those who possessed had a justification prepossessed for them -- "But it costs money to make a book", and they were right. The book is the first and in many ways the most important mass-produced article of western civilization, extraordinarily well-designed, beautifully evolved over hundreds of years of careful thinking and market determination of preferences since the adoption of movable type printing in Europe in the late 15th century. But books are heavy, they cost a great deal of skilled labor and time to print, and they are expensive to move and to deliver.
The knowledge embedded in books, and all other such analog representation systems, therefore came at a necessary social premium. The cost required to produce each marginal copy, which we learned long ago when we began to explore mathematically the theory of the firm, meant that in a competitive market the price of knowledge equaled the marginal cost of its production and distribution.
Since the end of the 20th century, a great deal of effort and enterprise among economists seeking to justify the way we have been doing things has gone into explaining why their prior theory was wrong. The theory of the firm demanded that under 21st century conditions, the price of information goods be equal to their marginal cost -- that is, zero.
A great deal of ingenuity was necessary to explain why that wasn't the actual price in a supposedly free economy. The answer, of course, was not long to seek -- it was the presence of state-created monopolies, eliminating competitive markets in the distribution of information goods. What we came over time to call Intellectual Property rules, rules of copyright and rules of patent, rules of secrecy, which established artificial barriers to the competitive economy's tendency to deliver information goods at their marginal cost -- that is, at zero -- that is, the tendency of the market, if left free to itself, to extirpate ignorance and cultural deprivation.
Because that is the stake involved in the moral question now posed to 20th century society -- ignorance and cultural deprivation are now preventable. What is the moral case for their continuance?
It is in this context that we spend our time worrying about the minutiae of a license for the distribution of computer software.
The goal of my client, my friend, and my more-or-less-constant interlocutor for the past 18 months, Mr. Richard Stallman, was to make a small point about the morality of computer program ownership which, when conceived in 1982 was a question a little bit ahead of its time, but it is a clear species in the genus of the larger moral issue I've just laid before you -- "Why is software property?" Stallman said. "It should be knowledge to be shared, like math, like physics. It's unethical," he said, "to deprive people of information evidently available to them about the artefacts of digital society with which they are daily in contact -- it's evidently immoral to deprive them of knowledge. You've given the knowledge to the computer sitting next to them. They're using it -- the knowledge is playing a potentially determinative role in their lives. You've already delivered it to them -- all you haven't done is to deliver them the ability to know."
And out of that fundamental ethical insight, there followed a determination that there ought to be some freedoms in connection with peoples' use of computer software -- they ought to have a right to use; they shouldn't be required to get permission. They ought to have a right to understand; they shouldn't be required to pay extra in order to know.
They ought to have a right to modify and experiment.
Though Stallman didn't give a traditional reason for this, didn't ground his position in any particular philosophical or ethical theory, he was in fact a Deweyite. He was in fact saying what John Dewey had said at the beginning of the 20th century, that the education and expansion of the human mind depends upon the opportunity to experiment with the world, that it is the ability to join forces with the material and immaterial knowledge surrounding us in the planet which makes our minds grow and develop.
You ought to have, Stallman said, the right to share, that is, the right to offer to anybody else the benefit of what you've learned yourself -- and with a series of homely metaphors in cooking, and other human activities of collaboration for shared psychological and social advantage, he made clearly enough the point that there was an ethical issue lurking in the determination to proprietize knowledge which is then delivered to people in forms that exercise power over their lives.
The result is the birth of the Free Software Movement. It has a technical objective, because its founder is a technical person. Its technical objective is to reproduce in freedom all of the best that existing computer technology not subject to production in freedom knows how to do, to recreate all the knowledge not uniformly and universally available in a form which will be guaranteed to remain free.
It is an attempt to use the creative power of a community of human beings to free everybody else's ability to know in the area of their human specialization. It is the motive which, transmuted to a different area, keeps people making Wikipedia entries and updating them and improving them and sharing knowledge with one another every day, in every language around the net, with which you are now utterly familiar and upon which most of us are to one degree or another intellectually dependent. This is its beginning, historically speaking.
The principle that this is limited to computer software is a useful opening limitation. At a time when neither the moral nor the economic consequences of this idea were fully grasped by anybody, even by Stallman, the limitation to concern with computer software served an extraordinarily useful purpose: it permitted work to get done. Two kinds of work got done -- software got made. A man who had never taken a course in compiler design, who had never studied the design of a compiler, who had no particular experience with the compilation of C in particular, but who was a very gifted technological designer and an indefatigable implementer, sat down and wrote from the light of natural reason a C compiler meant to be portable.
Stallman wrote the entire first version of gcc by himself. There are now more than four dozen people at IBM and some several dozen people at Hewlett Packard working full-time on gcc, and that's just two of the many places around the world where the compiler is constantly worked on. A community of people now maintains one man's shared idea. I haven't mentioned emacs, gdb, glibc, or any of the rest of Project GNU, but you all know that it is lurking there on the same principles.
In addition to the software which got made some early versions of the rules for sharing got made too. A first version of the GNU General Public License in 1985 and a second in 1992. At the moment that these licenses were written, the rules regarding the sharing, or for that matter the ownership, of computer software in the world's legal systems were indistinct at best.
The United States government, where Stallman lived and worked, and where almost all the users or writers of the freely shareable software then existed, the United States had great uncertainty about the legal treatment of software through the end of the 1970s. It was generally conceded, given the words of the United States Supreme Court, that it could not be patented. It was doubtful in many minds whether, the misunderstanding went, software being purely functional, it could be copyrighted. Though many thought some sui generis protection in law would be a good idea, nobody knew what it ought to be and no legislature had taken any step in that direction, and to leave it unprotected, save only to the rules of ordinary commercial trade secrecy -- though it was the course actively pursued before 1976 -- seemed to many businesses inadequate.
By 1979, a compromise consensus had seemed to have been reached in the United States, and on the basis of a Blue Ribbon Commission, the [indistinct] Commission came the general tendency, trend, or step in American legal thinking to assume that software was protected, if that's the correct word, by the law of copyright.
Thus, the initial rules for sharing undertaken by the Free Software community, Mr. Stallman and his allies, the initial rules for sharing assumed that only the law of copyright need fundamentally to be considered. And what was achieved was, within the vocabulary of the community, a very pretty hack --
a hack in the sense that the word is ordinarily employed in our, if I may call it, our community, an unexpected result achieved by creative deployment of existing parts in an unexpected or unusual configuration. The hack to copyright law was the recognition that the purposes of Free Software could be achieved by subtracting from the rights exclusively given to the author by the law of copyright as it applied to computer software.
What the Free Software author wanted was actually simply to remove a few pieces from the existing copyright machine. He didn't need to add anything to it -- no additional obligations needed to be placed on any user of the software, no additional agreements needed to be gotten from anybody who had a copy of the software -- all that was necessary was to remove some restrictions by sharing.
The copyright statute accorded to each author of a computer program, the exclusive right to control copying, modification, and initial distribution of copies. Under US copyright law, that's all there was exclusively vested in the author. What the author then wanted was to give the power to copy and modify away, to remove exclusivity, and to provide to others what the statute gave exclusively to him or her. With respect to distribution, the only principle necessary in order to protect sharing was to say, "If you redistribute, whether modified or unmodified, use these permissions and no other."
The result of which was to ensure that downstream, if we may begin to adopt the riparian analogy, downstream from each maker, everything would float freely available for further copying and modification, and each party redistributing to the stream was compelled to redistribute under those and no other rules. By honeycombing copyright in other words, by returning to the user some, but not absolutely all, of the exclusive rights vested in the author under copyright law, the social artefacts desired -- the freedom to copy, modify, and share -- could be ensured at the full strength that copyright law ever ensures the author's rights. No further contractualization, no further compulsion, no further form of legal coercion is necessary but a determination to enforce copyright for the benefit of sharing.
This structure, the slightly honeycombed copyright law, serving the purpose of enabling sharing, came to be called Copyleft. All rights reversed, Stallman said in the beginning. Later, my colleague Larry Lessig amended it to Some Rights Reserved.
The consequence was the creation of a social regime of sharing which unlocked in a very short period of time an explosive alteration in the nature of software in the world. The quality of software between 1980 and 1990 declined precipitously. There are few of us who worked in the industry on either side of that decade who don't know exactly in what ways it was massively dis-improved.
But the situation is straightforward, the theory of the firm and the economic theory of the 1890s are sufficient to the explanation. Monopolies, as we all know and were taught, produce inferior goods at high prices and stifle innovation. The wealthiest and most deeply-funded monopoly in the history of the world spent literally billions of dollars over the last two decades trying to convince you and everybody else in the world that the laws of our economics had been repealed for the special benefit of Mr. Gates and Mr. Ballmer.
This was, of course, false -- like all other monopolies, they obeyed the laws of the free market. They produced lousy goods at very high prices and they stifled innovation -- as they became more and more dominant in the world of software, quality enhancement in the world of software first ceased and then began to reverse itself.
The situation in the 1990s was decisively affected by the new economy of sharing. As the two largest governments on earth were first seeking to and then largely abandoned any attempt to restore competition to the global software market, as most of the commercial competitors who had attempted a run at Microsoft in any comprehensive way gave up and made peace, a small and largely disorganized community of people engaged in making software for sharing transformed the terms of the debate, produced excellent goods at zero cost, and began the process of dismantling the monopoly, which as you see is now beginning to take full speed ahead.
But my purpose is not to speak primarily about what was accomplished by Free Software in the first 15 to 20 years of its existence. That's then. This is now. My purpose is instead to talk about what has recently happened in the legal evolution of the principles of Free Software, and how that bears on the larger question of the political economy of the 21st century.
We have been negotiating the third version of GPL for the last 18 months. That is, for the first time since 1991, sixteen years ago, the fundamental legal mechanisms which established copylefted sharing and which produced this outpouring of free and competitively superior software are now under revision. Unlike the definition of GPL2 in 1991, or the definition of GPL1 in 1985, this was not an activity undertaken by Mr. Stallman and his lawyers in private. On the contrary, this was a vast and global negotiation, vast, at any rate, to me -- I put in more than 200,000 miles on it.
The process of negotiating the content of the third version of the GPL was the process of assembling a community, physically and technologically -- not a community which had not previously existed, but a community which had not previously engaged itself in an act of common legislation.
What we learned during the first 20 years of the Free Software movement's existence is a style of construction -- a style of making, which can be defined, roughly speaking, in terms of three essential components of any social and economic movement in the 21st century economy -- by movement I mean here including a business, a firm, or a government -- the three required components of 21st century economic activity are proof of concept, running code, and presence of community.
That's all that defines 21st century economic activity, in deep contradistinction to the industrial economies that preceded it. What the Free Software movement showed was that by proving a concept within reach and offering some working model, no matter how defective, partial or bad, in the presence of a community sharing the objective defined by proof of concept, the achievement of the outcome is simply a matter of allowing people to work freely with one another.
In the world of software, which is low capital intensive in every sense, that's all that's taken -- that's all that's required for the output to take form. We built tens, and then hundreds of billions of dollars, in valuable software using almost no venture capital inputs.
Capital wasn't what the system required to operate. What it required to operate was proof of concept, and an itch could be scratched. Running code -- something that began scratching the itch, however badly, and a community of people who wanted to see it through.
So what GPL3 meant was proof of concept + running code + presence of community. The concept had been proven already by GPL2. That is, the substantive concept -- we could make a set of rules for sharing that would make it possible to produce software all around the world that would be of ultimately high value but could be offered to anyone free of charge -- and provided with immense freedom to study, modify, and share.
There was running code -- we worked very hard for almost 2 years to produce a first discussion draft of GPL3, which we unveiled on January 18th of 2006 at MIT. And there was a community -- many communities, in fact. But their convocation for the purpose of legislation was a unique event. Every other week for the past 18 months, we've convened a conference call of 21 of the largest IT vendors in the world. Those companies, whose names are household-familiar in every household and business-familiar in every business.
Working in teams that varied from one person from some of the companies to five or six in others, carefully studying every single word, commenting as though their lives depended upon it -- as in some of the businesses they did, on every detail of the license's functioning in the global IT economy. We also convened, every other week, a conversation among 24 of the largest users of software in the world -- banks and brokerages, government agencies, and the lawyers who acquire software on their behalf.
We consulted every single week with the leadership of large free software projects around the world, some of whom use GPL and some of whom only interact with GPLd code. We spoke to hackers of enormous influence in the community, influence they have gained by their skill in programming and by their willingness to share, by their selflessness in helping others learn, and by the extraordinary wit and intellect whereby they have produced miracles out of thin air for all of us to use for years.
We conducted public meetings on every continent save Antarctica. We negotiated ceaselessly with people over what they needed, what they wanted, what they doubted, what they feared, what their concerns were, and in the end -- that's now I'm speaking of, this week, between now and Friday, the license gets itself finished and comes out the door as a final product -- in the end, we got agreement, we got consensus. Those who predicted at the beginning of this process that it would dissolve into flame wars or bad netiquette or some screeching meltdown benefiting only the monopolists were wrong.
I admit that there were days when I feared that they might be right -- it was no cakewalk -- but everyone who engages in legislation knows that it's never a cakewalk and almost never pretty. What is interesting about the legislative experience we've just gone through is how little of it, however, had the ultimate ugliness of legislation as we know it in the public sphere.
There was very little by way of campaign donation in this process, very little by way of buying results, very little by way of corrupting decision-makers. I don't think anybody actually thought there was any point in offering Stallman money. (laughter) And I don't think there was anyone who thought there was any point in trying to scare him. I think people began by pretty much assuming that the conversation was being conducted for the benefit of a decision-maker who could not be fooled, who would not be bought, and who could not be intimidated, and we skipped a lot of bad legislative behavior on that basis.
Technology came to our defense. I began the process saying that we would not use any technology not off the shelf, and I was wrong. We needed, it turned out, a web tool for allowing large numbers of people to mark up a document held in common in such a way that they could see and judge the intensity of one another's comments and participate not adding one more to a thread, but by seeing in a localized way what it was that others had already said about language that concerned them, and we required those making comments to anchor themselves in the text.
The consequence was a much lower volume of public commentary than I had originally planned for, by a factor of between 5 and 8. But the quality of the commentary was extraordinarily high. Because parties a) had to anchor their public commentary in the text, had to highlight a piece and say "here, this is what bothers me, and here's why." Because the tool made it inevitable that they would see what others had said before them on the same subject, because they could quickly visually identify the parts of the text that were already hot with commentary from others, we arrived at a mechanism for controlling the flow of commentary, which made it very easy to route comments from one place to another within the process, to give them the analytical treatment that they deserved, and to track everybody's comment and give them back an opportunity to know exactly what was made of the idea that they presented.
Every other place in the process where a particular comment was discussed, everything that was said about it, and what the final determination in the changing language of the license was -- we created a legislative process in which ordinarily people could vernacularly participate on more or less equal terms with the best informed lawyers and technologists on earth, and many people did so. Thousands of people did so.
In the end, the license bears in my judgment many of the marks of legislation. It is a little too long; it is a little too complex. It divides cases where they might with some analytical clarity have been merged, and it merges cases that might with some analytical clarity have been divided. It isn't one man's work of art -- it's a community's work of self-definition. And in that process, it replicates an early version of a 21st century reality which is that if in the 21st century what is produced is produced by communities, not by individuals and not by factories, then under 21st century conditions, what produces law is communities, not individuals and not the factories we call legislatures.
One of the great legal innovations, as the Americans saw it, of the 20th century development of law in the United States was the 20th century design and implementation of a Uniform Commercial Code from scratch. Americans being as they are, a hasty people, full of a desire for self-reinvention and little sense of history, they were spared the blessing of a commercial code based primarily around the maritime commerce of the Mediterranean in the 1st Century AD. They were prevented from having to adopt a whole series of mind games with which to bend the rules of particular commodity-based exchange in Rome into a structure capable of manipulating a 20th Century commercial economy, and thanks to Carl Llewellyn and his colleagues in the construction of the UCC a basic principle was always and everywhere, unprincipally observed -- "When in doubt, punt, and say whatever is commercially reasonable is what's the law."
The UCC's determination upon commercial reasonableness as the touchstone of activity is disgracefully imprecise, from the point of view of the Roman lawyer, and we glory in the fact. The Worth Street rules explain how to buy and sell cotton -- after all, they're doing it every day -- do you really want to know what Papinian thought about buying and selling cotton in Lower Manhattan? Well, much can be said about why this is a terrible idea, and I wouldn't be offering such a cartoon of it in Edinburgh if I didn't expect that there were people in the audience much better at playing the other side of the game than I can possibly be.
But I would present to you the possibility that the UCC and the GPL 3 are in themselves a pair -- a pair, organizing an idea about the method of the creation of 21st century law. 21st century law is born in the street in the same way 21st century television is born in the street, not sent to you from the top of a broadcast tower, but upward from the cellphone and the portable camera put through YouTube.
21st century law is like 21st century music -- not made in an expensive recording studio or legislature, rented by the hour by people with the power to rent studios and legislatures, but made in every laptop in every den in every corner living room, in every garage, where a musician and a computer are, which is pretty much everywhere a musician is.
Flickr, YouTube, and the other great distribution breakthroughs in our time are not actually mechanisms for the distribution of photographs or video. Flickr and YouTube and their equivalents are places for human beings to create communities through their shared interaction with images or moving pictures, as Wikipedia is a place for people to create communities around pretty much all the general knowledge that all the people in any particular community possess -- a place, it's true, for arguments and handwaving and pump-type disputation, as well as learned discourse, but a place primarily where the whole point is that we're all doing it together.
Of course, the idea that law might be something we all do together, it's got a long history stretching back far beyond the Free Software movement, stretching, in fact, back to beyond democracy. One of the characteristics that the continental Europeans noted of the English speakers, North Britains and South Britains, similarly in the course of the 16th, 17th, and 18th centuries, was that English-speaking people had an almost personal relationship to the common law. A man might be an artisan or a yeoman farmer, but he believed the law to be in some sense his own. He was familiar with the courts, he served on juries, the language of the law was in his mouth. Even beyond the language of literature and religion, it was in his mouth. It was, if not folk law in some forests-of-Germany sense, community law. The law of us, and to be not of this law was to be not of us in some fundamental way.
One of the problems faced by the monopoly, as its leadership now well understands, is that any community that it can buy is weaker than the community that we have built.
In any given confrontation between empires, there will be one side whose spies are paid, and one side whose spies act out of ideological conviction. The flow of the Cold War is in some sense the flow from one hand to the other of that painful proof, and I ask you this: Which side in the confrontation between the free world and the monopoly do you think pays its spies?
So the community that has grown up out of our acts of community legislation, the community which has grown up out of our ability to define what it means to share, the community which has used those rules of sharing to drive its economics deep into the heart of the global software industry, is now also beginning to model what the future of legal production is really like.
We have learned that the Flashmob and the MoveOn and the peoples' revolution in the street facilitated by the Internet are powerful forces in the politics of governmental change, new ways for people to bring to bear the sheer political weight of their approval or disapproval with those who govern. And the lessons on that subject taught so far in the history of the net are just the beginning.
What the net can do to politics in the YouTube, wifi, MoveOn, Facebook, MySpace Flashmob era remains to be written out in full, but I am less concerned with political process than with political substance. What the GPL did to the law of copyright was not an act of subversion -- it was an act of evolutionary improvement. It was a way to take the specialized law of the legislature -- let us call it the law of the publishers (which is what, to be fair about it, it really was) -- and to make it the law of creators by the joint unified act of creative people.
What we've just finished is the demonstration that that can be achieved even where the complexities of political, economic, and legal life embrace such additional concerns as the patent system, DRM on music and video entertainment, and many other controversial propositions which the community was required to discuss and reach consensus on in order for the license we have just finished making to come into broadly accepted, uniform effect.
The coming weeks and months will show whether I am overly optimistic in believing that goal to have been attained. I will say that no less than the fate of the monopoly depends upon whether we are right that what we have done will be successful, and the monopoly knows it too, as you have seen in recent statements. There is nothing left with which to threaten the monopoly's control of global technology but one slight chance that the European Commission will remain in the game and GPL3.
But that's sufficient, I believe -- we shall do their business as we have meant for the longest time to do, and the world will be a better place when we have done. That's just the beginning. That's only a matter of clearing brush away. The monopoly isn't in any intellectual sense interesting, it isn't in any ethical sense tolerable, it isn't in any economic sense necessary, it's simply a thing that happened to happen and that we will soon be finished making no longer there.
What is really important about what we are doing is that we are modeling other things that people can do for themselves. We are not creating something that you have to take from us, you either like or dislike, you either approve of or you disapprove of, you either wish for or hate. We are only establishing proof of concept plus running code plus community equals freedom. And that proposition applies far beyond the domain of computer software, applies far beyond the domain of freeing music from its owners, applies far beyond the domain of making the children of the world the programmers and the videographers and the producers and the directors, not the *consumers* of culture. Those things this is about, no question, and the mighty will fall in many directions, as communities begin to out-produce what capitalism's deepest and intensest collections of power can presently do on their own terms.
But this is the least of it, really, the least of it. The fundamental improvement being reached here is an improvement in the technology of self-government. An improvement in the technology of human freedom through the substitution of words and ideas for force. An improvement of the substitution of community values for the single minded pursuit of individual or aggregate profit at the expense of any other competitor that might get in the way. We have refashioned what competition and cooperation mean, as we have refashioned how production occurs in the digital economy. This, though large, though some will say too grand to be real, is the actual veilance of what we have just lived through.
"Seen backward through the end of the 21st century, our achievements will seem very primitive. They thought that it was something that they got a few tens of thousands of otherwise hierarchically disorganized people around the world to cooperate on a single act of limited purpose legislation, regulating the share of software," the 22nd century will say. "How quaint."
But it was the beginning of a joining-together of communities of affect in the global organization of power, the beginning of affiliation rather than territorial location or political domination, as the source of legitimacy for legislation. It was the beginning of the idea that cooperative private agreement can substantially oust public law institutions without challenging the legitimacy of the governments that participated in making the public law.
And it provides an escape from the moral dilemma presented by the myth of endlessly acquisitive homo economicus, the little homunculus of economic dream, the independent entity with the exogenously derived preference schedule, competing with sharp elbows in the market against every other homunculus economicus seeking only the same narrow benefit off the same asocial schedule of what I need today.
A bad myth about human nature, dying now the death it has long deserved since the middle of the 19th Century. A bad myth about the nature of technology. A bad myth about the nature of the social production of knowledge. And a doing-in of the primary obstacle to universal information and the end of ignorance. It's just as small and just as large as that. Just as tiny a step in just as epochal a direction. Just as small a project with just as large a set of consequences as those. We were lucky to be there.
Generations of people have participated in the struggle for liberty of thought. Crucial rounds in that struggle were fought here in this city, as they were fought in Rome and in Paris and in Berlin, and even in New York and Los Angeles.
The difference between us and all of those who've struggled for the freedom of thought in the past is of no particular credit to us. We are not smarter, we are not stronger, we are not more indefatigable. We are merely lucky. We are lucky because along that long scale of the struggle to know and to share and to improve humanity by the control of nature for the benefit of all, in that long sequence of people, many of whom died unlamented or unwished because of their adherence to our goals -- in all that long period what distinguishes us is merely a contingent fact of our role. This time, we win.
I'm happy to take your questions.
The role of Stallman almost looks like a king with a very large council of advisors. Do you think the process would have worked without him, or do you think the process would have worked without him or a similar candidate?
That's a very important question in my judgment -- I think the answers lie in two directions; first, Mr. Stallman's presence as an ultimate decision-maker provided the necessary political will to get things finished. The difference between this negotiation and many standards-like negotiations in which I have participated over the years is that the parties who usually find it desirable to extend time indefinitely could not do so.
We were very careful to publish deadlines and structured guidelines to process before we began for the purpose of making sure that all participants knew on what date things would be finished, when the transitions would be from draft to draft, so that it was possible to prevent the dilatory practice that would otherwise undoubtedly have run out any clock ever made.
I said by the second month of the making of GPL 3 that the first consequence of the publication of a draft of GPL3 was that GPL2 became "perfect". I myself had some doubt that the license was perfect; it appeared to me to have some flaws, it was a parochial American document, it had little to say about the patent problem. It had done nothing that it could have done about problems it had not foreseen like lockdown in aid of DRM and the growth of the web services controversy. It seemed to me plainly a license in need of some improvement, but the minute we published a draft, everybody loved it. Loved it to death, loved it far more than that -- human beings are afraid of change.
In this respect, and probably in no other, Mr. Stallman's presence was essential to what happened, and the next time -- that is, when one repeats the experiment without this, I think you're quite right, 'whiff of enlightened despotism' to it -- one has to find a different mechanism for ensuring that parties will face a similar necessity to get things done. The concentration of the mind does not depend upon the imminence of being hanged -- Dr. Johnson only meant that it was a sufficient condition, not a necessary one. (laughter) In the same sense, I think Stallman's participation and the uniqueness of his role is sufficient to achieve certain consequences but not necessary. The process of democratizing, in the fullest sense, collaborative legislation is a difficult process. The history of Wikipedia shows just how difficult true unhierarchical government of a collaborative structure for production is. And solutions like rotation of authority or debate over merit or even just straight election will not do.
I don't mean to suggest that I know exactly how to replace Richard Stallman. I don't know how to replace Richard Stallman. That's why if you'll permit me to say so -- I've just been the lawyer for the last 13 years. If we'd known how to replace him I suppose there would've been an overwhelming public cry to do so. (laughter) But indispensable is a thing that societies sometimes recognize, and Mr. Stallman has been in the literalist sense indispensable to what it was that we were trying to achieve. He was the party everybody trusted to be an honest broker, there was more to say beyond honest brokerage but that's sufficient for the moment.
What I think that structurally means it that you will see in the course of the 21st Century the evolution of strength through poverty again. And I say again because if you think about the history of the West, there have been times when moral and political leadership was very strongly vested in organizations whose primary claim was poverty, humility, and sanctity. It doesn't surprise me that we found ourselves back in relation to a system not entirely unlike that at this moment, it's an evolutionary stage in the structuring of politics. Powerfully poor, powerfully honest, powerfully sad non-governmental organizations will play a very significant role in the voluntary construction of law in the 21st century. Look at it this way: who do you want making the international law of criminal rendition at the moment -- the United States Supreme Court? The European Court of Human Rights? or Amnesty International?
I think that the fundamental truth presented by Stallman's role in the GPL3 process is that human beings recognize that there is a need for intransigence in the pursuit of freedom. Which is more generally construable as when you have a community which thinks of itself as distinguished by its values, those values need a personality to cluster around. That that leads them to be an ultimate decision-maker with complete power, such that everything else is advisory I don't think follows. And my guess is that if you want to imagine the flow of the evolution of those institutions over time, you'll see ones led rhetorically and intellectually, by people of unimpeachable honesty and poverty and humility, maybe something even approaching sanctity, but without the power of ultimate decision.
One of the most controversial parts of the GPL version 3 was the inclusion of technological protection measures, and seeing as it may be happening that technological protection measures of DRMs may be killing themselves by their own, let's say, unworkability, and ineffectiveness of the whole concept, is it a wise decision to include it in the license, seeing that it generates so much discussion and so many problems?
No, but, you needn't think of the discussion in and of itself, or even the disagreement in and of itself, as somehow a drawback in a legislative process. The task I believe that I was set as the lead negotiator in this area was a diplomatic task, to separate the IT industry on Planet Earth from the entertainment industry on Planet Earth. The entertainment industry on Planet Earth had decided that in order to acquire Layer 7 Data Security, it was necessary to lock up layers 2, 3, 4, 5, and 6 so that no technological progress could occur without their permission.
This was known by the IT industry and the consumer electronics industry on the planet to be offensive nonsense, but there was no counterweight to it, and there was no organized consumer dissent sufficient to require them to stand up for technical merit and their own right to run their own businesses without dictation from companies a tenth their size.
Not surprisingly, since it is part of the role we play in this political power concentrated in poverty, humility, and sanctity, we bought them to a consensus they were unable to bring themselves to -- which is represented in the license by a rule which fundamentally says, "If you want to experiment with locking down layers below 7 in the pursuit of data networks inside businesses that keep the business's data at home, you may do so freely. We have no objection -- not only do we have no objection to you doing it, we've no objection to your using our parts to do it with. But when you use our parts to build machines which control peoples' daily lives -- which provide them with education and culture, build devices which are modifiable by them to the same extent they're modifiable by you. That's all we want. If you can modify the device after you give it to them, then they must be able to modify the device after you give it to them -- that's a price for using our parts. That's a deal which has been accepted."
To think of this somehow as tsoris we shouldn't have had is to miss the importance of the social agreement that results from it. And if you want to say, and I totally agree with you, that the possibility of Layer 7 data security locked down by layer-crossing technology of superlative badness and stupidity is going out... yeah, that's right, it's going out. It's going out because our friends have found their backbones. They don't wanna make the stuff, they don't wanna sell the stuff, because they don't want to make and sell the stuff to people who don't want to buy and use the stuff. So we're gonna win. You're quite right. And that's very good, and the people who don't like it can just plain lump it. As they should. They were asking for something which, whether you think of it as technical architecture or business model or social competition policy, or a way of educating human beings in their culture was defective, bad, and wrong. And stopping them is for the public good, and we have done it. But it isn't right to say that it all would've happened anyway, and therefore we shouldn't have had the conversation. The conversation we had is part of the reason it isn't going to happen any more.
I think we've time for probably about one more question.
The GPLv2 is frequently, or sometimes used in a way that allows it to be succeeded by the GPLv3. Is that used regularly enough, in your opinion, or.. ?
Is it used.. ?
That clause that allows the overriding by the more recent versions.. .
When a program is labeled "GPLv2 or any later version", or after next week, when a program is labeled "GPLv3 or any later version", the author is delegating to the users a part of the authority to relicense.
The decision to delegate to users a partial authority to relicense is a policy decision by an author. Some will make it, some will not. Linus Torvalds is a good example of a programmer who decided from the very beginning of the life of his project that he wanted to retain the power to determine what license would be used, not the users. Mr. Stallman, and many other users of GPL 2 in general chose the other answer -- and delegated to users, for purposes they thought good and sufficient, partial power to relicense the code. GPL 3, like GPL 2, will retain that flexibility. An author may decide to keep all the power to relicense in subsequent license revision to herself, or she can decide to share it with her users. GPL 3 adds a third possibility -- she can decide to designate a proxy to make that decision at the time when a future license is issued. What we might call putting the re-licensing decision "in trust", to trustees.
All three choices seem to me beneficial choices to have -- I don't think there's any benefit to exercising any one of those choices in the architecture of the license. It's compatible with the license's purposes to afford a rather broad range of delegation options, from none to complete. At the present time, my guess is that the desirability of using GPLv3 or any later version is largely seen in the trade and in the industry as a question of "Do you trust the Free Software Foundation?" And parties either will or won't use it under GPL 3 according as they do or don't trust FSF. I actually think it goes a little deeper than that. I think the question is, "Do you trust the community?" Which is not just FSF -- but that whole GPL society we just convoked.
If I were myself working primarily on the construction of code these days instead of law, I would label my code GPLv3 or any later version, because I trust the wisdom of the society I've just been working with. Maybe I have a better view of that than anybody else on earth right now, because I've faced more parts of that society in more different contexts over the last 18 months than anybody else. But I think this, as so many other issues, comes down to the value of trust. Trust is the hidden feature in the economy, as Henry George thought land was the hidden real single factor in the 19th century economy. When I talk about Proof of Concept plus Running Code plus Presence of Community equals Freedom, you could do a parameterized substitution and say Presence of Trust equals Freedom. Because what we're really saying is, communities that make on the basis of pure reciprocity are communities held together by the Power of Trust.
The decision to license code for downstream relicensing is a decision to trust the rest of the community with everything you've done. It is remarkable how often that decision has been made in the past 20 years, and it is remarkable how high the returns on that investment have been, in my judgment. Thank you all very much.
Wow. I'd like to call on Paul Motion, my predecessor as the Chairman of the Scottish Society for Computers and Law and presently the Convenor of the Law Society of Scotland Technology Committee, to give the Vote of Thanks.
Thanks, Ian. Well, where to begin? How to follow that act? The vote of thanks, traditionally, has been three minutes of self-conscious groveling, and I'm going to try and avoid that if at all possible. Just to introduce a wee bit of audience participation, there are 142 of you here tonight, which is an astonishing achievement for the Society -- I think it's one of the biggest turnouts we've ever had, and a tribute to the speaker. Can I ask how many of you are not Lawyers, can we just have a show of hands?
(many hands go up - laughter)
There you are, Eben, there are plenty of real people here tonight, that's great. Thanks are due to a number of people in particular, to Ian Mitchell the Chairman of SSCL for organizing tonight, to Rosie Saunders of the Faculty of Advocates Training Organisation for helping to muster the numbers and deal with the applications, to Valley Technology behind the cameras and on the cans that are recording the event for posterity -- I'm sure you'll all be downloading it festidiously. And to my competitors, MacLure & Naysmith, for sponsoring the event tonight. Nice to see you guys. And to Script, of the University of Edinburgh, for helping with the organization generally.
Open Source Software is becoming more relevant to unreal people, lawyers like me; I'm currently having to get to grips with GNU Version 2, and the Creative Commons license in relation to copyright, and no doubt version 3 of GNU will be on my agenda very soon. In the course of the speech tonight, we heard a number of very novel and interesting concepts. We heard about the Software Freedom Law Center, which is a law firm that is also a charity -- foolish notion, but (laughter) I'm sure it'll work. We also heard that there was criticism of the morality of depriving people of that which they want and which has already been made -- admirable sentiments. But I think what caught my attention most of all was the remark that what the GPL did to copyright was not an act of subversion but an act of improvement, and that was carried out in the context of the substitution of community values for the aggregation of profit. I think tonight we've heard more about community values than anything else.
And you can see that the community is beginning to assert its will, not only in Version 3, but it has been reported today for example that the Open Source Consortium is about to report the BBC to the European Commission in relation to the most recent version of the BBC's iPlayer, which only works with Windows Media Player and DRM, oddly enough. So we wish them success with that, and the OSC have also challenged the UK Education Authority, BECTA, which in their opinion is not promoting Open Source software within the education system adequately, so I think the Open Source bandwagon has moved through to the realm of legitimacy. We wish Version 3 the very best for the future, and I would just like to ask you to join with me in thanking Eben once again for his fantastic presentation.