The Hardware Wars and the future of free software
From Wikisource
| ←Wikisource:Speeches | The Hardware Wars and the future of free software by Eben Moglen |
Presentation given at the Free Software Foundation 2006 Associate Member Meeting at Massachusetts Institute of Technology with questions from the attending Associate Members, 1 April, 2006
|
[edit] The Talk
[Eben Moglen:]
Hi! Good afternoon.
Yeah, OK.
Let me just say how glad I am to see everybody, and what a pleasure it is not to be giving a GPL3 talk this afternoon. I will take questions about it; it’s not that I refuse to discuss it, but it is a pleasure to think a little bit about something else.
So, all I will say about that in the main talk:
Vista will be late,
Office will be late,
Virtual Server will be late,
GPL3 will be done on time.
Free Software is better, OK?
But the purpose of the talk this afternoon is to talk about the next thing after.
We’ve gone through some piece of this strategic thinking before, right?
The long term curve here for me, and that’s a personal opinion, the long term curve here for me always was: You free the software executable layer of the net, and the consequence is that the cultural non‐functional layers above are pressed into freedom by the nature of the technology underneath, so proprietary culture becomes much more difficult to operate when the net operates underneath in a free way.
And the net operates underneath in a free way, modulo of the problems of network neutrality and attempting to bias life between the endpoints. The network works in a free way because the software works in the interests of the people who use the net, and they enforce freedom on it by changing it until it works the way freedom requires.
As cultural properties become instead cultural commonses, as we share our way to a different kind of culture based on a net which operates freely, we soon discover that inequalities of bandwidth are, fundamentally, sources of injustice. And we go to work trying to equalize the bandwidth situation in the world around a conception of freedom to share, which assumes that human beings have some rights to expressive technology in the same way that they have rights to the public streets or to drinking water or to other subjects of common rather than propertized development.
So for me there’s a sequence of events, or movements, or turns in the technology which we are trying deliberately to move in the direction of what we think justice in the 21st century requires.
I used to think that the sequence temporally was more or less the same as the sequence analytically. And I went through most of the last ten years believing that the march was pretty much the way I laid it out.
I feel, actually, some hope about the bandwidth questions. That there will be a much more rapid acceleration towards the perception of the possibility of a free bandwidth regime in the world. Because I think Asterisk is going to have an effect on the telecommunications oligopolies that was unforeseeable until Spencer really got himself rolling.
I can see already, in the large industrial players in the IT universe, a determination to abandon the telecommunication system. And their determination isn’t ideological, it’s simply based around the fact that PSTN and circuit‐switched telecoms and paying rents to the telecommunications oligopolists doesn’t do Deutsche Bank any good. They can get for nothing what they are paying for and better, and they will start to get it that way, making use of technology which is Free, in the Stallman sense.
When we built SFLC around free voice over IP, bare metal + Asterisk, for which I’m very grateful to Bradley, that was proof of concept + running code in a major sense, I realized that if a small business like ours could do that, everybody in the world could do that, and the big boys would get out first. Which they are already starting to do.
The consequence for the telecommunications providers is that five, six, seven, eight, nine years from now, they will face a market which is very hostile for them. The problem is that Deutsche Bank will be gone. And the Bank of America, and Goldman Sachs, and everybody else who has technical skill and money to spend not paying rents to telecoms anymore.
The problem is that the people that that leaves attached to wire line telecoms can roughly be described as aunt Sally. And the problem is, aunt Sally has never been a rate‐payer who could support Verizon. On the contrary, Bank of America and Deutsche Bank had to support aunt Sally’s use of wire line telecoms.
If the telcos ever got themselves into a situation where they had to go to state regulators and say “You must allow us to jack up aunt Sally’s rates, very, very substantially. Because Deutsche Bank doesn’t patronize our invention anymore, and the business subsidy to consumer telephony is gone.”, the regulators would say “No”.
“We’re in the business of protecting the rate‐payers against you. Being a public service commission does not, in its essence, require us to protect you against competition, if competition should suddenly develop. We weren’t in the business of protecting you against cell phones, though we did allow you to use the money that you earned from rate‐payers to go out and buy cell phone frequency and buy cell phone carriers in order to protect yourself. You want us to go and tell you you can use your rates to go and buy up VoIP and suppress it?”
Well, there will be two problems with that. The first is, they can’t buy up VoIP because VoIP is free software and it can’t be bought. And the second is that the question of suppressing it is the network neutrality fight now going on. They will have lost it by the time the crunch comes down the road this decade.
“Or”, of course you will say, “they might win the network neutrality wars and build a multi‐tiered Internet that controls everybody and then we have lost.”
So we’d better go back, then, a step.
In between that moment of… software is free, culture is becoming free, folk are beginning to ask about bandwidth. In between was a risky moment of conservative reaction, which we have described in the past, all of us together, as the hardware wars. “Treacherous Computing” is my friend Richard’s usual way of describing it, right? “Trusted Computing” they call it, meaning computers you can’t trust.
That’s the moment at which, like the network neutrality wars, there comes a moment at which gravity might reimpose itself on us.
And as Richard has repeatedly pointed out, in our particular corner of this larger process of making freedom, hardware that resists the freedom of software is a very dangerous form of counterrevolutionary resistance.
And, of course, people have been developing hardware that resists the freedom of free software, no question about that.
Intel’s Trusted Computing activities have been very deep and very significant to the company intellectually as well as, at least prospectively, financially. A lot of effort has gone into the proposition that Trusted Hardware can resolve problems of data privacy and control over individual autonomy against identity theft and personal data mismanipulation.
As well, of course, as the equally prominent promises made to the entertainment companies about how hardware that resists the freedom of free software could help them with their business model difficulties.
And Richard has spent several years warning people that this was a problem that could doom the whole operation. And we’ve talked about it in our occasional face‐to‐face meetings, all of us, for the last several years as a thing we were going to have to deal with.
By and large, when I’ve spoken in venues like this one about how we were going to deal with it, I have said that winning the war for free hardware was essentially a conservative problem. We had to hold on to the hardware we traditionally had been using. I don’t mean we literally had to run those boxes until they fell apart from electron fatigue, but rather that we had to constantly emphasize the consumer’s desire for general‐purpose hardware that worked for the general purpose computing user, and that’s all.
I still think it is correct that this is essentially a conservative activity. That is to say, the principle of hardening machines to resist freedom is a novelty, And in that sense when we fight against it, we’re fighting against a change somebody else wants to make which is bad for us.
But I have begun to believe that we are not faced with a problem that can only be solved by organizing consumers to resist “Trusted” hardware.
I’m glad that I am beginning to believe there are other ways of going to work because I was afraid of having to go to work that way. Organizing consumers is hard. It’s expensive. It’s lengthy. And it is always vulnerable to counter‐advertising. Not to mention vulnerable to laziness and inertia.
So the idea that we might have to mobilize all the consumers to resist hardware that resists the freedom of free software was a problem that I was not eager for us all to start having.
Nonetheless, it was clearly a fight we were going to have to have, and that we were going to have to win.
And at the moment, whether you take Richard’s generally more pessimistic or my more generally optimistic views of the situation, At the moment it is clear that that confrontation is getting much, much closer.
Anybody who has been to an electronics store in Japan, for example, in the last 15, 16 months, has noticed that there are now linear feet, indeed almost linear miles, of hardware sitting on little shelves waiting for consumers to come and get it, all of it prominently marked with the “Includes Windows DRM” little thumbprint down in the corner of the label.
Everybody is, of course, aware that Mr. Jobs has moved into our neighborhood, bringing a TPM in the moving truck, and we have been wondering what it was that all that Trust inside the MacBook was supposed to do.
And then of course we said a couple of pretty mild‐mannered, straightforward things in January about DRM in the GPL, and everybody went nuts.
Which, trust me, they really did do. I’m not talking about Linus. Linus has not been, by any means, nuts, compared to what else there has been.
You will have seen on the surface a little bit of all the tizzy that that quite reasonable language has put people in.
In other words, everybody now knows it’s at stake. We know it, they know it, we’ve taken some moves, they’ve taken some moves, we’re getting closer. And from here you can almost sort of see how the battle might go. And as I say, I recognize the threat and I appreciate the extent of the danger, but I don’t think the strategic news is terrible at all.
Here’s the thing.
Mr. Jobs, with the kind of good fortune that we could claim was his birthright if we didn’t also know he was suffering from some serious downsides in life, Mr. Jobs wins the prize of as the last manufacturer of the Gillette razor. Right?
The 20th century had many business models like this. You make a cheap razor and you sell expensive blades.
I thought that Curley was going to go out as the last razor manufacturer, that the inkjet printer with the cartridges that cost an arm and a leg was going to be the last such business model in our field, but he snuck that iPod in under our nose and won. At the end of century, he won the prize; it was small, it was cute, it was cheap, compared to, I don’t know what, but I guess compared to Apple hardware generally. Right?
And it was the songs that were expensive. Expensive in all sorts of ways: Morally expensive, physically expensive, expensive to your belief that you could have music that would last and that you could carry with you through life, from place to place and time to time.
And it has, to be fair, exceeded at least my expectations about how cool unfreedom could be made. Right? I mean, he has turned unfree into hip. And you’ve got to give the man credit, that’s an achievement. But when you look at all those little boxes out there in the world you realize that it was a freak. It wasn’t even what he meant to do.
In North America they’ve been reasonably responsive to people whose batteries died. Although I have friends in western Europe who are still complaining very vociferously that, dead of North American consumer protection law, Apple’s basic position is “When your battery dies, throw away the iPod and buy another one”.
Engineers didn’t design that as a product to obsolesce early, under the assumption that people are going to be carrying them around for 20 years.
They thought it was the first version of the razor handle, and that the iTunes Music Store was the razor blades and everybody was going to pick it up, and they were going to win, and this little loss‐leader device was going to be the throwaway.
Then the loss‐leader device turned out to be the profit center, They couldn’t redesign the product to last forever; they gave us a couple of versions of things that were a little more durable, but now they’re up against a problem.
The thing looks small, but it takes up way too much space in your pocket. Because there’s really only room in your pocket for two things, one on each side.
And, a thing that only plays music, or even a little grainy video along with a little bit of music, or even a little porn along with music, it’s not enough. Right?
So the cell phone guys think “OK, we’ll get you. We’ll make it also a telephone. And then we’ll really win because we can give away the phone and the music, in return for selling the bandwidth over which it all travels. If we can just convince the music companies that we’re locked down enough to trust.” So they, too, go into the business of making a device that does a couple of things.
Upstream from them, Nokia is thinking to itself: “Boy, phew, we’re really glad we’re not a cell phone company anymore. We were the world’s leading manufacturer of cell phones a nanosecond ago. That’s why we’d not only like to get out of the cell phone business, we’d like to kill the cell phone business off altogether.”
Because under 21st century conditions, when your’re the world’s leading manufacturer of something it’s time to get out of the business, and if possible destroy the business behind you. Because right behind you come a thousand Taiwanese design houses and ten thousand factories in Malaysia and the People’s Republic of China, and tomorrow you’re going to be competing at a disadvantage in the business you broke the ground for.
Take your winnings and get out. Do something else.
In Nokia’s case, what do you do? You become a Internet tablet company instead of a cell phone company.
You look at Wi‐Fi‐enabled, peer‐to‐peer, voice over IP, converts to GSM or G3 when it has to, only if, maybe not, and by the way, also crunches your numbers, makes your dinner, shows your pr0n, does your music, whatever it is. Right?
“And whatever you need tomorrow we’ll make that too, inside the same box. Because if we don’t put it inside the same box we might lose the war for cubic centimeters in the pocket, which is the biggest war of the next ten years for us as device manufacturers.”
Alright, so you sit back and you appreciate where the device manufacturers are now living. What the business is that they are now in, and the challenge.
Their problem is to be agile manufacturers of everything you need in pocket size. And it’s all got to work in the same shell. And it probably does have to be interoperable with a lot of other stuff. “UWB Bluetooth, I take it?”
Alright, so what do you need in order to be a manufacturer of devices in the 21st century? What do you need in order to succeed at being Sony, Nokia, Siemens?
You need an omnicompetent platform that can do anything.
It needs to be robust.
It needs to be fully debugged at doing everything you want to do.
It needs to be extremely well understood and very secure.
And it has to cost zero dollars per unit.
Because if it costs anything more than $0 per unit, the guy who makes it is going to wind up eating your lunch.
Now, oddly enough, there’s exactly one place in the world you can go for a platform that meets those needs. There’s a lot of stuff you could almost use. But there’s only one thing you can really use.
If you want to do it Mr. Jobs’ way, they realize, you could go FreeBSD and wrap some stuff around it, but the reason you can do it if you’re Mr. Jobs… Doing it that way is…
You have a very narrow range of hardware around it. And you control it totally. And you don’t have to work with or about everything else. And you get a chance to figure out which feature set to add when, in a sequence that works for you.
You want to do it the other way, in real, total chaos, in the actual rough‐and‐tumble of the market? You need us. GPL’d free software, a kernel, and a bunch of other things. And they’re the only choices you really have right now.
Moreover, it doesn’t do you any good to develop a new one. You’ve got to debug that one, too. You’ve got to maintain that one. You’ve got to make that one work everywhere on earth. Which puts you back in the position of having an unacceptable cost structure.
So as we went from 2002 into 2003 into 2004, as Mr. Jobs accelerated the race for cubic centimeters in the pocket, as the cell phone manufacturers realized that only agility in device manufacture would do, as the time to market that you could allow yourself before you begin to compete in the embedded market at a disadvantage shortened, and as tens of millions of units of hardware with our stuff in it started appearing all over the world, I realized, we are becoming an essential raw material.
To agile electronics manufacturing in the 21st century, we’re not, at the moment, replaceable with anything else.
So the moment at which you’re an essential raw material of manufacture, is the moment at which you have some leverage.
Not, I admit, all the leverage in the world. In the traditional way that the game called industrialism was played, when you become an essential raw material of manufacture, the United States marines show up off shore, and, right, you know?
I mean, there are risks, being a raw materials producer wasn’t a great thing in 20th century political economy. But if the raw material you produced was petroleum, you did OK.
A dicey, careful, dangerous neighborhood, all that stuff, liars, cheats, warriors, generals, secretaries of defense; it’s not the pleasantest bunch to work with.
But as an essential raw materials producer we have some power and the time has come to use it.
That’s the logic for me of what’s going on in GPL3. And it’s also the logic of what I hope will be going on in a social sense in the free software movement for the next 18 or 20 months.
This is the moment where we have to press the door and try and make it turn.
What are our advantages in that process?
First, the virtualization thing cuts very much in our favor, oddly enough.
Everybody has begun to realize that we now make rocket ship general purpose hardware in tower and desktop cases. And that it’s really a good idea to jam it as full of processes as you can. So virtualization is becoming a terribly, terribly important idea in all the heavy hardware manufacturers’ views of the world.
When you have working virtualization, you begin to recognize that the idea of trusting an operating system kernel, a whole kernel, is a terrible idea. Kernels change too quickly. And they are too big.
If you want to build hardware that’s going to attest the remote state of software, da, dada, dada, dada… The last thing you want is to be trying to attest to the actual moment‐by‐moment maintenance condition of half a million lines of code.
What are you going to do, have hashes lying in some Trojan point out in Earth orbit, where everything goes and tests for each point release? “Oh no, I was running the -mm one, and not the…”, right? No, it’s…
So the very guys who think in terms of trust, are beginning to ask themselves whether they don’t mean, actually, a much narrower range of software in trust with hardware. A virtualization layer. Trusted Xen. You know, eighty, seventy thousand lines of code, that doesn’t change very often, that sits between the metal and everything on top of it. Trust that, and let the rest go.
And as I go around talking to the engineers around the world working on trust, that’s more and more of what I’m hearing about. Which is nice for us, because it is reducing expectations about what the technology can really do, to a point at which the engineers can satisfy themselves that they have used the technology wisely, without inevitably coming into conflict with what we need and want.
From our point of view, if you can install and run a modified version of some code that you received, and if that modified version will execute with respect to all the data that the version you started from was capable of reading, writing, playing back, or doing whatever it did with, then you can continue to build free software.
If underneath there’s some trust layer of Xen or some other virtualizer that’s interfering with what happens at the bare metal, well, some human beings will take it out.
Similarly, if a guy builds a DRM application in user space to play movies in a disempowering way, if he gives you a music player that won’t let you do everything with music that you would like to do, but it’s only sitting up there in user space and it depends on kernel services underneath that it can’t control and can’t get any attestation about the state of, then it’s not very dangerous to us.
If it’s proprietary code we’ll replace it with free code, if it’s free code we’ll take out the features we don’t like, if, for one reason or another, we have to live with it, we’ll wrap it up in such a way that it isn’t harmful to us, or we’ll modify the kernel underneath so it does what we want.
In other words, as the theorists of trust move down towards thinking about virtualization, Or up, to think about application layer DRM, they’re doing things which makes the collision between their activities and freedom of free software less intense.
That’s the first thing.
The technological drift is not engineers wanting to extinguish freedom. Although there may be people who want our software to stop working, they’re not primarily the engineers working on trust, anymore.
Next.
The music industry is not having a good year, either.
The rootkit thing at Sony was a very important turning point. There are two reasons it was an important turning point, the first thing was it struck with lightning the most schizophrenic of the content companies.
Content Sony and hardware Sony had very hard time living together in the last six months. Not least because there were bad actors like us coming along and trying to fan the flames of the civil war. Right? And we spent a fair amount of time saying to guys who are box builders in Japan: “You need us.”
“You just let some two‐bit German executive go and steal our stuff and put it in your box and create malware and make trouble, and all of that is ruining your rep with the community, and you’ve got billions of dollars in new, smart, agile devices with our stuff in it coming to market and you piss on us now?”
And as you see, it was not possible for content Sony to keep some heads from rolling. But it was way worse than that. The Sony rootkit thing also produces the following bad effect:
Guy goes to work at Bank of America, with music CD in pocket. Puts into slot in Bank of America computer. Causes problem.
Bank of America says “Dell, what do you say?”, Dell says “Not our problem.” “What do you say, Mr. Microsoft?” “Well, not our problem either, we had no idea.” “You had no idea?” “Well, we’ll get back to you on that.”
Which means that a whole lot of businesses had a lot of “Oh my god, computers we can’t trust” experiences, based on behavior—bringing a music CD to work—that they were not in a position to tell their employees they could’t do.
It’s all very well to say “We’re going to have massive new wonderful investments in computers on your desk at work that won’t take your USB key and won’t install your software, won’t do this and won’t do that and won’t do the other thing.” But when it gets to “won’t play your music”, it goes too far.
So, business now learned that malware is a thing that happens because entertainment companies are trying to protect music. That means that everybody starts thinking “Wait a minute, wait, what are you going to do to our network?”
That’s our second great opportunity.
Third thing that’s happened, is that people have begun to realize that there isn’t some overwhelming juggernaut of “one way to do it”. The entertainment companies and the box builders are having another VHS/Beta war that everybody knows is stupid, but nobody can figure out a way to stop.
They recognize that consumers are not actually happy that music is being sold in ways that won’t move from here to there and that there’s some kind of new rule that you’re going to have to throw out all your music and buy it again every five years.
They recognize that they’re being pushed to make and deliver products to consumers that are neither the product they wanted to make nor the product the consumer wanted to receive.
So, each of these elements, taken from around the edge of where we are, can be wrapped up together to make a frame for us, which if we push on right now, while we are also a indispensable supplier of raw material to device manufacturers, might actually change the state of play.
What would we have to be successful in saying? I think there are three things we’d have to be successful in getting across to the world. And different parts of the world, moreover.
First, with respect to GPL3, and with respect to the attempts of the free software movement overall, we have a message to get across to people in our industry:
“No, we are not primarily fighting with you about movies. We are primarily fighting to protect our way of making software. Our efforts are designed to protect our right to modify and reinstall, improve, and share software. Don’t tell us that we’re off on some quixotic, romantic engagement with reshaping the movie business. We’re hackers, we make software. Our concern is protecting the integrity of the process that makes software. And all the legal devices that we recommend, and all the trouble that we are making, is to protect our homeland against somebody else, who wants to come in and prevent us from doing science and technology the way science and technology should be done according to us, the scientists and technologists.” That’s point number one.
Point number two:
“What we are playing for is the same thing always: rights of users. But users’ rights are about to take a very different form than they have taken in the past. In 2006”, we say, “a home is some real estate containing appliances. In 2016, however, a home will be a digital entertainment and data processing network with some real estate wrapped around it. Now, who do you want to have the keys to your home? You? Or the people who deliver movies and pizza? This is really about who’s going to have the keys to the home next decade”, we say. “And what’s the other side of this one?” we say.
With the executives who’ve been calling up and complaining about GPL3, I have been asking a question, I say: “Gentlemen, at what point do you want to admit onto your data processing networks, in your buildings, computers which run software you can’t see, can’t understand, can’t control, and which reports to other people what is going on in your network without your ability to interrupt or do anything?”
“Well, not at all” they say.
I say “Right”. I say “At what point do you want to bring a box like that home and put it down on the desk next to the computer your child needs to do his homework?”
“Not at all.”
“Right.
That’s the situation we’re trying to prevent. Those boxes, general purpose computers running software you can’t see, can’t control, which are in touch with other people on the network in ways you can’t understand, those boxes are called intruders. Do you want to have an intruder in every room of your house in ten years or don’t you?”
“We don’t.”
“OK, then you know what we have to do.” That’s point two.
Third point.
Which, again, unlike the one in the middle, is primarily addressed to our friends in the industry.
“Well”, we say, “fifteen years ago we told you patents were a problem.”
“And the problem we told you patents were, was a problem of third party control, of your destiny as makers and our destiny as users. We told you ‘If you don’t do something about patents on software inventions, you’re going to wind up living in a world in which you don’t control your destiny, because you can’t be sure that you’re allowed to make the boxes that your consumers want to buy from you.’”
“That happened to you, just as we said it would. And everyone of you now knows that you have lost ability to control your destiny which it’s expensive to get back. What do you think is going to happen if you let a few third party industries determine for you, for the next ten years, what machinery you’re allowed to make, and what your consumers are allowed to do with it? You’re going to lose control exactly the way you lost control over your destiny over patents. And it’s going to be an even bigger problem. You’re just empowering third parties to control your destiny, which as businesses is a thing you hate to do.”
If we make all three of those points clearly, firmly, sharply, in their appropriate context, and we refuse to play ball with conditions for the deployment of our software which prohibit the freedoms that our software is built on, we can win.
I don’t say we do for sure, not by any means. The pushback is going to be extraordinary no matter what. But those arguments are now addressed to people who can understand what we’re saying.
The technical conditions, the market conditions, and their own sense of the risks to their business models mean that they can hear that the argument we are making is a solid one.
They don’t necessarily believe as we do that it’s for sure right, and they are terrified of saying that the other guy is wrong. Because the other guy is a very good customer, because the other guy is a distinguished ally and joint venture partner. Because the whole system of the ownership of culture is a system from which the IT industries have made a lot of money over the years, and they expect to make a lot more. But it is possible to explain to each of those businessmen individually, and to their businesses collectively, why they benefit with us from a guarantee of operability of hardware in the free software mode.
So that’s what I think we’re now about to try. I think Richard is entirely correct, if we don’t do this at the right moment we are lost. I often find myself listening to Richard wondering if he is right about the moment. This time I think he’s right about the moment.
This is it.
We’re not going to get a much better condition for this fight than we have right now. We might have an infinite number of worse conditions. So I think we’re now probably in the unusual condition of wanting to speed things up a little bit, on this subject.
We can’t move the hardware technology, but we don’t have to. Because for the first time in 15 years, we have the license open.
Now, there has been a great deal of pressure on us to remove the DRM components of the new license draft from conversation altogether. And that, plainly, would not be a good idea.
How, exactly, we choose to play out the diplomacy around the license, and how that fits into this larger politics, is for me the big question of the next several months.
We got heaps and heaps of stuff to do. We’ve got the whole GPL3 process to run, and we’ve got some more stuff to do around other licenses.
We’ve got a lot of politics going on about patents, a lot about source code, a lot about license compatibility, you’re all following the comment stream, you know what’s being discussed.
But for me, the big question now is, how to drive the available resources strategically at the goal, which is to split the coalition in favor of DRM.
To begin dividing the device manufacturers directly from the entertainment population. To begin to get the agile device manufacturer business model drivers to recognize that it threatens their own business model to disable the freedom of free software in their devices for the pleasing of the music and movie companies.
We get that point correctly across and we set the terms of the industry in the GPL3 and other licenses, we really will win.
There will be plenty of DRM around, and everybody except children, and hackers, will be in some danger of occasionally bouncing off it. But hackers and children will go right through it like grass through a goose.
There’ll be lots of userland DRM that we’ll be going to get around in a million different ways, There’ll be lots of trusted virtualization “brought to you by” the People’s Republic of thus and so, and people’ll get around that too.
But where the hardware gives way to software in a real sense, and all the way up to the point at which there’s always a layer underneath you can fool with, we will have protected ourselves against business models that disable our forms of development and tinkering. Which is, I think, a practical definition of success in the hardware wars.
So I truly think that it’s probably a good moment. I see just enough to work with and I don’t believe we’re ever going to get much more. I see us striking while the iron is hot. As I told The Guardian earlier this week, sooner than this and we wouldn’t have had enough support, later than this and it would have been too late. I think this is when it has to happen.
When Stallman is standing here, he’ll say: “If we don’t do this, we’re doomed!” I don’t know if we’re doomed or not. I would like to believe that we could even survive a reverse on this one.
But there’s no question, it’s a big deal. This is an issue on which our loss is a permanent loss to freedom. So we have to do it right if we can. And judgements about timing are really hard.
But then, Vista will be late,
Office will be late,
Virtual Server will be late.
That’s to say, the other guy is not looking good.
If you want to build agile devices, that are going to do everything, that are going to fit in the pocket… Have you been in an airport recently? Everywhere I go, it says “Is that Microsoft Office in your pocket?” I think to myself, there’s an infinite number of dirty jokes there, risqué…
They can’t get anything onto the desktop this year. They’re not going to win a race to the pocket with us either. They’re not about to become the omnicompetent, totally debugged, completely safe, utterly secure, fully robust, “build anything and it will just work” platform for stuff you fit in your pocket. They’re not going to endanger our current leverage with the market.
And all this “DRM is inevitable” talk only makes sense when you’ve got a technical backup for this claim of inevitability; I see inevitability a very hard claim for the market to make right now, so let’s bite the jugular while it’s exposed.
Which brings us to one last question, which is “Could somebody give us a kind of DRM we could live with?”
And here I have to admit that I don’t understand what’s been going on the last couple of weeks. I don’t think there’s a nice, sweet, “let’s all just get along” answer to this one. It’s not that I don’t think Sun has it, or that Larry doesn’t have it, or that so and so… It’s that I don’t think it exists.
I don’t think there’s a nice, easy “we can all just get along” thing here. It’s true, if we free our layer of software, if we mind our knitting and free software, movies and music are going to change their business models. They are.
And they’re going to resist. But they’ll do it in the end.
And yeah, the hardware business will change, and yes, conceptions about how to do data privacy and other subjects will get real, instead of believing that locking up all the leaf nodes in the network is somehow going to solve major social technology problems.
So, I feel reluctant, I don’t want to be the guy who says “I shot Archduke Franz Ferdinand this morning, you all have to fight the first world war.”
But it’s kind of like that.
Right? I mean, it is kind of like that.
You kind of trusted us to tell you when was the time. So now’s the time.
We’ve got to teach people about this. We’ve got to make ordinary people understand that their home is where this happens. That their living room, and their bedroom is where this happens, that the family room is the place whose key ring is up for sale.
We’ve got to make them understand why the stuff we do is about making sure that they have the keys to their house in 2016, not Mr. Eisner or Mr. Murdoch.
We’ve got to help get across, to the businesses we do business with, that we understand our role in their ecology – we make stuff for them, and we give it to them gladly. But we want to have preserved beyond any peradventure of doubt, our right to make software like that.
“That’s all we’re asking for.” we need to keep saying. “Don’t get confused by FUD you hear that we’re trying to get you to free our movies for us, that’s not your problem. We demand of you, who redistribute our software, that you protect the mode by which we make that software. That’s all we ask of you, but it’s perfectly reasonable, and you have no justification for saying no.”
We all need to say that, because if they get successful in characterizing it as “something Stallman thinks”… “One man’s opinion”, a lawyer said to me in one of the corporate meetings on GPL3 in committee B a few weeks ago, “One man’s opinion, you can’t put that in a software license.”
So we have to be able to say “No, it’s not one man’s opinion. It’s a political point of view, with a party behind it, that comes loaded for bear, that has assets you need, with whom you must bargain. This is a coalition and we’re part of it. We make billions of dollars for you and you have to respect our part of this system.” That message we also have to get across.
And then, last, we have to say “Look, we all understand, some of us are in business and some of us aren’t, but that doesn’t mean we don’t understand the problems of business people. We think that business people have a real problem when somebody else is allowed to come in and order them to do business the wrong way. We just want to point out that that that’s not us. We didn’t go to the United States’ senate and say the business models of IBM and Hewlett‐Packard were piracy. That was Mr. Eisner. We’re not the ones who beat you up all the time and humiliate you and pistol‐whip you in public over whether your boxes work the right way or not, that’s them. We are with you about this, your destiny should be freely in your hands, as our destiny has to be freely in ours. We’re for your freedom too.” has to be our statement.
Now, some days that’s going to be a little hard to get out of the craw. Because we’re mad at them about stuff. Because their freedom just isn’t as valuable to us as ours is to us. But that’s the problem of coalition building.
If we can make a coalition with one half of Sony, three quarters of Hewlett‐Packard, 95% of IBM, et cetera, et cetera, et cetera. We win. This problem’s over.
There will be trusted boxes, there will be Microsoft Windows DRM, there will be DRM in Linux userland, there will be all kinds of offenses before God.
But the freedom of free software will be protected and the underlying mechanism of how to free culture, and bandwidth, and move towards technological liberty in the 21st century will still be viable. That’s what we’ve got to come out of here with.
So, my sense then is: We begin from here figuring out how to make that campaign work, we are overmatched as usual, we are small, we are poor, we are humble, we are this, we are that… But we have right on our side. We have some leverage to exploit, we have some resources at our disposal, and I truly believe good timing is 90% of the chore.
What if, contrary to all the “DRM is inevitable” stuff that you are hearing now, what if two years from now, there is still no fundamental attempt to lock down our boxes on us? How much more delay becomes our victory?
Here’s the last piece:
After the MacBook was announced and Mr. Jobs’ latest razor came to market, Richard was pushing me, pretty hard.
“There’s a TPM in there, this is awful, they will prevent GNU/Linux from running on Intel‐based MacBooks, you have to do something about this, how do we fight them?”
I said, “Well, let’s examine the premise just for a minute, Richard. TPM in there can point two ways. One thing it could be there to do is to prevent our stuff from running on their hardware. But another thing it could be there to do is to prevent their software from running on other people’s hardware. Which do you think it really is?”
“Well, it’s the worst, obviously, got to take everything to be at its worst.” he says, being Richard. “It’s hardware that’s there to prevent GNU/Linux from running on MacBooks.”
“Well”, I said, “What if it isn’t like that? What if it’s hardware designed to prevent OSX from running on standard Intel whiteboxes?”
“Oh, well, we don’t care about that.”
“Right.” I said. “So, let’s negotiate before we start fighting. If we get out of them a promise that they will not use their hardware to keep our software off their boxes, or to limit the rights under GPL on those boxes, or to limit the rights under GPL on those boxes once that software is there, how do you feel about that?”
“That’s OK” he says.
“Is that OK? I think that’s doable.”
I still think that’s doable, I’m not done yet, but I think it’s doable. Because they want us, too.
When you’ve got 4, 5 percent of the hardware market on a good day, squinting, into the sun? Throwing away folks like us is not a good idea. We buy hardware. Why should they not sell us hardware?
Merely because we run our own software on it? That’s not their problem. De‐branding the OSX so that is runs on every whitebox, cheap system in the world, it’s in just another commodity, that Mr. Jobs doesn’t want to do.
In other words, there are ways we can divide up this market and conquer pieces of it as we go. Provided that you’re pushing public opinion in the right direction.
Mix and match. I’ve stated there are three points, any good campaign should have not more than three or four, but maybe I’ve got’em wrong. But mix’em up. You’ve got one op‐ed per person per issue per local newspaper per year.
We got one radio call‐in or one Q & A show‐up at a technology day at a public library, or one operation at a high school or one place where a PTA can be addressed, there’s some place where everybody can help to get across at least the message “This is your living rooms’ future we’re talking about.” If only that, we’ll get enough out of it.
And we will bring in our friends, I mean, Mark Cooper at the Consumer Federation of America is still there. We’ll do some work together, we’ll reactivate some portions of the free digital rights programs that we were working on before. We’ll invest your dues wisely.
But I think this is now mostly about energizing the bits of public opinion that we do possess to spread the memes a little further. Because I think the tipping point is not out of reach. And if we win the hardware wars it’s very clear sailing after that.
OK, that’s all I want to say, but I’m happy to take some questions.
[Applause]
[Unknown:]
I have a dull question, I think i have a more interesting question…
[Eben Moglen:]
Why can’t you ask your more interesting question first?
[Unknown:]
So the dull question is, you mentioned a timeline of 18 to 20 months. And how… What is that a timeline, concerning…
[Eben Moglen:]
Well,
[Unknown:]
And how does that relate to a 12 months timeline given a moderate number of comments on the GPLv3 draft?
[Eben Moglen:]
Because it’s not all about the license.
One part of it’s about the license, another part’s about what happens after Vista hits the street and Microsoft starts talking about more DRM next time.
And another part’s about the product cycle for that first round of post‐MP3‐playing cell phones. It’s about what happens as the Japanese embedding manufacturers start to move a very, very large number of consumer functioning free software devices towards market.
We need them all to think as they design the follow‐on to their first round of products, “No, it is not worthwhile” to put trusted computing hardware in their boxes.
None of them wants to do it if they don’t have to because it costs them money. So the question becomes “Are we we going to be forced to do this by Disney?”
And our answer should be “No, actually, you’re not. Actually, you’re going to be forced to do the opposite by us. You’re going to have to go home and tell Disney ‘We don’t know how to build these boxes, except using GNU/Linux, and we’ve got to play by the rules. We can give you userland DRM, will you take that?’”
Disney will say: “No, we can’t take that, we are against userland DRM, you’ve got to harden it down completely!”
“Well, we can’t, I’m sorry, we’ll give you userland DRM, will you take it?”
That’s the bargaining that’s going to have to happen.
And where’s the other part of that 18‐to‐20? It’s the end of the Blu‐ray/HD DVD wars. It’s when they all begin to figure out what their formats are for the next round of video and music and they start figuring out how much protection they can demand or expect, and we need to have some leverage by then.
[Unknown:]
So you’re still envisioning the code freeze and final adoption of a GPLv3 circa 12 months, rather than…
[Eben Moglen:]
Oh yeah. I told you, we’ll be done on time. I think we can see new licenses in January of 2007. Maybe Richard thinks we’re going to see them earlier than that, but we’re going to see them in January of 2007.
The other thing I would say about this is: There are a bunch of guys who’ve been yelling and screaming and carrying on with us about DRM in GPLv3, and they’ve been saying to themselves under their breath: “Hahahahaha, we’ll just switch to LGPL.”
But they haven’t seen the first draft of the LGPL3, yet, either. I mean, we’re going to be clear about this, we are not playing a game where we’re going to leave exits open. We know that this is important and we’re going to play it to the end, so no, I just mean by saying “18 to 20 months” that there’s life after the license.
We can do the license, you’re going to have to do some of the other lifting.
[Unknown:]
So, just to conclude that, you do believe that it’s important to adopt an LGPLv2 license coincident with the GPLv3?
[Eben Moglen:]
We’re going to have a GPL and an LGPL that are fully interoperable in the sense of being paired as the current pair are paired.
And they will both speak importantly to the question of how to protect the freedom of free software.
[Unknown:]
And with to respect to the problem that you just pointed out, they need to be released simultaneously?
[Eben Moglen:]
Correct.
[Unknown:]
OK. So, I’ll get back to the last question, I still have the interesting one.
[Unknown:]
First of all, thanks for your talks.
So, the question’s this: Do we need, and, technically, can we get, the kernel to upgrade its license?
[Eben Moglen:]
Technically can the kernel upgrade its license? Yes.
That’s not, in my judgement, a subject on which any lawyer ought to speak authoritatively, except the lawyer representing kernel developers, which I don’t. But, yeah, I think so.
Do we need? Look, I… You know, it was never my idea that we would make a GPL version 3 that was almost good enough.
I have always thought we would make a really better license. So far I think we’re on a path to make a really better license. I don’t intend for any project anywhere in the world using the GPL to have no good reason to change to version 3.
Do we need for the kernel to change?
As a practical matter, speaking as one individual, not on the behalf of the Free Software Foundation, not putting words in Richard’s mouth, not making any announcement beyond that,
[Unknown:]
I mean for the context of the tran [Unintelligible]
[Eben Moglen:]
as one observer: I don’t think it would make a lot of sense to go to all this trouble only to wind up with a license that the Linux kernel developers couldn’t use. In their own judgement.
[Unknown:]
OK.
[Unknown:]
Thank you very much for an inspiring talk, I’m ready to march off.
I come out of the political world, I’m not a techie. When I come into this world, my sense is: one of the greatest vulnerabilities is the accusation that free software is unamerican.
And given the dust‐up over the Dubai Ports and a lot of other things in the air right now, and you could pull some quotes out of context from various people and make a nice 30 second ad and run it on Lou Dobbs for six months and you could, you know, pound, into millions of people’s heads, some ideas that would be very awkward for us to overcome.
And, I’m wondering, do you see that as something that needs attention, and, you know, is there someone out there doing that?
[Eben Moglen:]
Well, I’ll say three things.
One: It is true, that as a political and social movement, this one, the free software movement, has a whole range of risks that it inevitably takes on, with respect to an overstrong identification between exclusionary property rights and “The American Way”.
Right? That’s always been there, it never goes away, we can’t do much about it, and it is not my job, as a lawyer, to discipline the political or social statements of my clients, that’s not a dignified practice within the professional obligations of lawyers to control what your clients think about politics.
So, I’ve never actually devoted myself to strategies that would require me to muzzle my clients, they’re just not thinkable any more than cheating my clients is thinkable.
So I live in a world where there’s a certain amount of this risk that has to be run. And I can’t answer you by saying “Oh yeah, I’ve got a little thing in the back room that’s going to make everybody speak politically as though they were country club republicans,” because that wouldn’t be right.
On the other hand, there are some very clear ecological forces that militate against doing that nowadays. If Mr. Ballmer came back out and said “Linux is a cancer on the American way of life and the GPL is Satan.”
Then on the phone the next morning would be a bunch of executives from companies that he can not ignore saying “That’s our business you’re talking about, Mr. Gates, you want to know what we think about yours? Read The Wall Street Journal, we’re quoted there this morning.” Right?
My friends at Microsoft had an enormously successful week in one respect only: They got the United States’ government to send officials to Brussels to complain that the European Union was not being fair enough to Microsoft. Right? All of that is based on good behavior, both sides.
If the guys on the other side start demonizing us as unamerican, then they’re harming IBM’s business, and they’re harming Hewlett‐Packard’s business, and they’re harming Dell’s business, and they’re harming Sun’s business, and there’s going to be a lot of complaining and squawking, and some harsh name‐calling is going to go on on other sides, too, some of it under the breath and some of it in public.
There are ecological limitations to how far you can go on that. The ODF wars in Massachusetts have begun to explore some of those outer reaches, right?
And Microsoft haven’t done itself too many favors there either. Mr. Quinn’s an extraordinarily dignified and thoughtful man, and to have him going around saying, you know, “You walk into a buzzsaw when you deal with those guys.”
That hasn’t been helpful to them. They don’t actually want to be seen as beating up on democracy too much.
So, yeah, we have to be a little careful about it, There are days when I read things that my colleagues and friends in the free software movement do and I clutch my temples briefly, and I… Right? And I know that there are people who do that about stuff I say, heaven knows, right?
But we’ll get through that. It’s a risk. But I think there’s much more risk in our simply failing to get our points across than there is in having “Communist!” called at us down the street, or something.
[Unknown:]
I wanted to address the future of the free software movement.
I’ve been following your work, I’ve been following your writings, Stallman’s writings, And, to be brief, nothing short of brilliant.
And I want to know, what happens after you two and some other members of the FSF retire?
[Eben Moglen:]
Have you met Mako Hill?
[Unknown:]
I am looking f…
[Eben Moglen:]
[Unknown:]
I am… That’s my answer, and that’s… I'm very…
[Eben Moglen:]
That’s one part.
[Unknown:]
Yeah.
[Eben Moglen:]
That’s one part. Right? I mean, you see around you that that’s not going to be a problem when it comes to hackers, and political workers and organizers.
My job is training the lawyers. That’s what the Software Freedom Law Center is. That’s the place where we make new, young, enthusiastic lawyers, and we send them out all over the world to do this work.
I am glad that, about the time that started really to bother me, I found myself in a place where I could go and conscript several millions of dollars a year to do that training with.
If I didn’t have the Center, I’d be getting antsy about that too, right now. But I’ve got a machine for training lawyers to do this work.
I’ve got a machine for training lawyers to do this work that allows me to pay them what they need to be paid in order to pay their loans back after school, which means I’m not going to lose them to the law firms where they’re going to do horrible work that they don’t want to do but they have to pay back their educations again.
I’ve got a deal for people outside the United States that is the best deal in the world that I know of for a foreign lawyer wanting to work on a hot topic in the United States right now. I can give them an LL.M at Columbia Law School, I can give them training, I can give them a job, I can give them a Rolodex, I can give them the right to practice in the United States and a heap of cash.
Right now, I have an ability to get lawyers produced. And I’m going to use it.
So, between the people you see in this room and the people I know and work with from day to day I think we’ll have a cadre way better than the one you have right now.
I appreciate the praise, but trust me, it can be done a lot better than he and I have done it, that’s for sure. You know him, I know me. It’s, uh… Right, so we can be improved upon, and we will be.
[Unknown:]
Well, thank you very much. Thank you.
[Unknown:]
You touched on this briefly last year; a major, you know, component of your talk was about the way the software patent battlefield seemed to…
You know, you had several of major industry players, some friendly to free software, some not, that were amassing enormous libraries of patents.
[Eben Moglen:]
Yup.
[Unknown:]
And I’m just curious to see how you see it.
Has it changed at all, since last year?
[Eben Moglen:]
Well, so let’s see where we stand on the patent side.
The European situation is now a permanent political standoff, right? We’re at stalemate and stalemate suits us better in some ways and worse in others, as I pointed out last year.
Continental European programmers are mostly still protected against the worst of what can befall, and programmers in the UK are in the soup, and they didn’t get any help out of the European system, which I was hoping that they might.
On the other hand, for as long as stalemate persists, the possibility that we achieve Stallman’s wartime goal, which is to carve software out of the patent system altogether, remains open and on the table.
Second. The patent promising and related, moderately good to totally awful, habit of saying you’re going to be very virtuous, is catching on a little bit.
And so we’ve got a lot more promises lying around than we used to have. They vary in worth from about “worthless entirely” to “not worth all that much”.
I found myself, not very long ago, in a telephone conversation with a character called Craig Mundie. And this was a surprise to me, because he usually tries to avoid talking to me.
On this particular occasion he had to talk to me and so the telephone called. And in his very jovial way, he said:
“Eben, I don’t remember the last time I talked to you!”, I said “Yeah, sure. It’s nice to talk to you now, Craig.”
He said “So what’ve you been doing?”. I said “I have been kicking the tires on patent promises.”
He said “Ah, they’re terrible, aren’t they?”. I said “Well, they vary.”.
There’s a real awareness on all sides, including with the adversaries of free software, that patent promises are worth squat, but you get them because people are aware of the problem.
So if nothing else, they serve as information that people know that the problem is there.
The Open Invention Network move of IBM, Philips, Nokia, and others is a very interesting move. As much for the parties who are in as for what they are, so far, in with.
What’s most interesting about OIN from my point of view is that is doesn’t look so much like a classic patent pool, it looks much more like a bank for buying patents.
And this means, I think, that the parties comprising OIN have decided, among other things, that they need to be awarier of the trolls, they need to be careful about the possibility that there will be patent auctions going on concerning patent claims that are strategically important to us, and which nobody from our side of the world has any money to buy, and they are becoming afraid of Nathan Myhrvold’s Intellectual Ventures.
I think “Open Invention Network” stands for “Not Myhrvold’s Intellectual Ventures” as clearly as it stands for anything else.
Whether to be afraid of Myhrvold, and how much to be afraid of Myhrvold is one of the most difficult questions for us to pursue as strategists, because he’s very dark.
“Obscure”, I mean, not “evil”.
The theory that he’s a harmless crank, overfed with Microsoft winnings, and not particularly dangerous to us, has some very important adherents. There are others who apply the “Prince of Darkness” theory and come up with a rather different set of views. It’s that nobody knows, I think, firmly, convincingly knows, that is causing a lot of unease.
So, my sense is that in the actions, we’ve seen a year in which a fair number of people continue to show us that they knew it was a problem, and to be very attentive to possible large‐scale systemic activity.
At the nuts level, at the rubber‐meets‐the‐road level, there’s heaps of worrying going on. Companies with lots of money to spend are spending millions of dollars inventing around.
Very significant attempts to figure out how to conduct patent defense in a meaningful way are going on.
On the other hand, the Public Patent Foundation, my other little venture with Daniel Ravicher is still a “don’t be seen in the photograph with” class entity for all of these people.
We still have a very strong Chinese wall concerning the funding of PUBPAT over here as against Software Freedom Law Center over there.
And, this is an “and”, it’s an inevitable “and”, the very parties who won’t be seen with us and want to be sure that not a dime of their money is going to PUBPAT are constantly sending us little messages by underground communication, asking us if we would please go and take care of this, that, or the other problem for them.
Which is a reminder of the last piece of how I see this.
We look at the patent problem and it looks pretty simple. Partly because Stallman has just been saying “Just say no” for so long, that we think we have a simple answer.
They don’t think anything about it is simple, because they live in a different world than we do, when you have many tens of thousands of patents of your own, the world looks like a different place.
Some of the things we think can be done with a stroke of the pen, they can’t do at all, some of the things that we can do with a stroke of the pen, they can’t do, and don’t want to encourage us to do except implicitly.
About 11 months ago, a lawyer called me up from one of these companies, and he left a telephone message, he said: “I’ve got that phone number you were looking for.”, he said, and he read me seven digits. Click!
And I hadn’t called him up and I wasn’t looking for a phone number. But I looked at it and I thought “Yeah, that’s not a telephone number.”
So I pulled the patent, I looked at it, I thought “Oh! Yeah, isn’t that interesting?”
It was a piece of Microsoft skullduggery. A bad piece of Microsoft skullduggery. An intricate piece of skullduggery.
And I thought “Yeah, OK, well, hmm. It’s also a crummy patent, it’s never going to be enforceable. Maybe we don’t have to do anything about this.”
But I said, “Yeah, thanks, got your message. Bye.”
And we let it sit for a while, and then after a while people came round and said “Hey, did you ever do anything with that phone number we sent you?”
So, “Why do you ask? Wasn’t it a deep secret? You’re not paying us, so it’s… You want me to tell you what I do with the stuff you don’t admit being seen telling me?”
But we did a little stuff, and then after a while, in a certain place or other, Microsoft came and wrote a letter that said “We would never, ever, ever, no, we certainly didn’t mean, no, we wouldn’t be want to be thought to have, no, absolutely not, no, no, no, absolutely not.”, signed Microsoft.
And these guys call us up and they’re very fulsome in their “Thank you, we’re really grateful to you for doing all that wonderful work.”
Now the truth of it is they really couldn’t do it themselves. They really couldn’t. I mean, we think “Well, what’s the big deal, you make a phone call. You call up the so‐and‐so people and you say ‘Do you know about this such‐and‐such piece of skullduggery?’ and you let the nature take its course.”
But you can’t make that phone call if you’re a guy with a zillion patents. Because then you start worrying somebody is going to come and do it to you. And you don’t want to be seen talking to PUBPAT, because that would imply that you don’t believe in the sacred religion of all the patents that you have.
How can you be seen with terrorists? You know, I mean, those guys blow up real estate, you know?
So what we’ve got is a situation in which nobody thought the patent system was less of a problem to software at the end of this year than they thought at the end of last year. Everybody knows the problem only got worse, OK?
And they want it solved. But they’re not yet ready to say “Let’s just carve ourselves out and be done with it.”.
And even more importantly, PhRMA, the real owner of the world patent system, does not yet think that carving out IT is better than keeping them in its soup.
My personal belief about the long‐term strategy of this, and again, this is one person speaking, my personal belief is: If we keep on this line another four, five, six years, PhRMA’s going to reach a tipping point. At which it’s going to conclude that it is safer to have IT out of the boat than to continue taking risks with the stability of the world patent system overall, that come from having IT executives in the boat who no longer firmly believe.
And at that point, with a magic of suddenness, reform will start to move. OK, but there’s an important Supreme Court opinion out there that they haven’t written yet and we don’t know what it says.
And until we find out what goes on in eBay and MercExchange we don’t know one piece of information.
We don’t know exactly how many United States senators really thought their BlackBerrys were going to stop working. Right?
We don’t know exactly how the fallout of that thing happens yet.
We don’t really know what Microsoft might do next year after Vista is out, and how the market might respond to misbehavior on their part about their patents,
There’s a lot of uncertainty. I think we made progress. I think that it will be easier to make a deal about patents in GPL3 than to make a deal about DRM in GPL3.
But both deals are hard at the present time, and there’s a lot of yelling and arm waving going on on both subjects. So that’s, I think, how I would run it all out.
[Unknown:]
Thank you.
[Eben Moglen:]
OK.
[Unknown:]
OK, my question is, you’ve been talking about the content industry, like the movie industry and the recording industry, more or less as a monolithic entity.
And I’m wondering, how monolithic is movie recording industry and so on, is there any hope of any wedges appearing there?
[Eben Moglen:]
Well, four companies control 95% of the world’s popular music, and eight studios between them control about 90% of the English language movie business in studio manufactured movies, and about 40% of the manufacture of so‐called “independent” movies in English.
So, part of the problem is, it’s a highly concentrated industry. Another part of the problem is that it’s an industry of inconsiderable size and no economic importance whatever.
Music and movies together, produce in direct sales in the US economy about 42 billion dollars. If you let them puff up to maximum size, by counting every possible dollar as revenue and not noticing that they don’t pay royalties on any of the rest of it, they’re about 80 billion. For comparison purposes, IT in consumer electronics in the US is about 825 billion, OK?
The very fact that they’re so economically unimportant, in a way increases the threat that they pose. Because all they make is image. But they make a lot of image. Right?
And they get a lot of people sitting there slack‐jawed in the movie theater while some scene painter says that BitTorrent is going to put scene painters out of business, and starve set workers to death, right?
So, in a way, it’s precisely the disconnect between the reality and the movie version which is their strength.
Yeah, independent cultural production has real meaning now, right? There are recording companies splitting the net roots income with musicians, and there are movie making activities going on that are profoundly anti‐propertarian in nature and Indymedia is happening and Google Video and Flickr are changing how we think about the distinction between the producer and consumer in cultural property.
But it’s still a very small playpen with some very big people in it, in their own judgements, and they still play as though they had the whole fate of world civilization in their hands, and by and large, because there are a few of them and they hog the microphone they’re very widely believed.
[Unknown:]
Thank you. I wanted to ask a brief, or maybe not so brief, technical question about patent strategy. In particular, I’ve seen, in the past few years, some prominent patents, people have gone to the patent office and asked for a reexam, of the patents, I think the FAT patent and a couple of other ones. And I’ve spoken to patent attorneys who’ve told me that that’s a very risky strategy, because in the reexam process the patent holder gets the last word in, you don’t get to challenge, they get to come in with, and respond to your arguments, and you can’t respond to them, you can’t cross‐examine, and at the end it’s, you know, quite possible you’ll end up strengthening the patent rather than hurting it.
[Eben Moglen:]
Well, that’s a correct analysis so far as it goes.
That’s to say, reexamining carelessly, or taking your shots poorly, can result in making a patent that looked weak before look stronger afterwards.
And it is true, that the ex parte system of reexamination, the “outside the lines” system of reexamination means that the patent holder has very substantial procedural advantages. The trade is that you have a very low cost of entry to the system.
So, if you compare that to an inter partes, adversary procedure over the validity of the patent, say, in the European Patent Office, you come to the conclusion that, yeah, you take additional risk of the party sweet‐talking the examiner, but on the other hand it costs you maybe 1% or 2% as much as mounting a full‐scale partisan attack on the patent against full resistance.
But the other thing I have to tell you is that patent attorneys also have a tendency to over‐estimate the risks of reexamination, because they prosecute patents. Because every time you reexamine a patent you’re pissing on a patent lawyer.
And that’s not a risk that we care a whole lot about.
PUBPAT’s basic philosophy was this: “We are going to shoot at the most important patents in the world, about which we are certain to be 100% right.”
So the first one we did was a basic biotechnology patent which had been triplicatively patented by a university owner, trying to extend indefinitely its power over a very basic biotech invention. The university at stake expected to make 120 million dollars over the life of the patent, and we blew it up, the patent owner was Columbia University, and we did that to prove that we would not take any softer a view when my own employer was at stake than we would in any other case.
The second one we did was the Microsoft FAT filesystem patent.
And the third one that we did was Lipitor.
At each of those we had complete and total victory at the first level in the reexam. We were right.
Lipitor is worth ten billion, with a “b”, dollars a year to Pfizer.
The third patent on Lipitor, the one that said “Hey, you remember that chemical that was already invented twice for lowering cholesterol? We crystallized it, could we have another 20 years please?” That patent.
The one that says “Same chemical, same use, but new, crystalline form!”, right?
That patent was worth seven additional years of ten billion dollars a year to Pfizer. That’s seventy billion, with a “b”, dollars.
Now you can be sure, that after we for five thousand dollars have blown up that patent, that Pfizer is going to spend a certain amount of money talking to the PTO. Right? I suspect, in the end, they might even succeed in convincing the PTO.
Microsoft did. We blew up one of six FAT filesystem patents absolutely justifiably, Microsoft went into whispering conference with the examiner, on appeal, and the examiner has now reinstituted the patent altogether, on the basis of two arguments, both of which had been previously rejected by the patent office in writing.
So, in due course, we’ll reexamine again. And we’ll say, “No.”
“You can’t put back the patent on the basis of an argument already rejected on the file wrapper. Who are you kidding?”
And we’ll win again. And they’ll spend more money, and they’ll get it back again, maybe, or maybe not. But, let me point out two things.
One. You may not collect royalties for infringement of a patent under reexamination. And you may not have an injunction to prohibit distribution of infringing goods of a patent under reexamination.
That’s to say, each time we file one of those requests, we make two years of time. And each time they spend ten million dollars putting back what we did for five thousand, we achieve a little bit of tactical advantage.
Not to mention, occasionally pointing out to people that you can do seventy billion dollars’ worth of health care reform for five thousand dollars, if you’re not the United States congress, right?
Now, all right, you may be right, we may have strengthened the FAT filesystem patent a little bit; I don’t think so. But even if we did, if was worth doing what we did in order to take those chances. But I agree, it’s a calculated enterprise.
We don’t think we’d had been better off if camera manufacturers and CompactFlash card manufacturers had been paying royalties to Microsoft these past three years, and if they were gearing up to say the Linux kernel could not read or write MS‐DOS disks, because of the patent, without a license.
So, OK, “Take your lawyer’s advice, don’t try this at home.” I guess would be my answer.
Thanks.
