This is the raw text of an interview
with Eben Moglen, Legal Counsel, Free Software Foundation, 3 March 2004 done by Kathy Bowrey Senior Lecturer in Law at UNSW in Sydney.
The interview was done by kathy in the process of writing her new book Law & Internet Cultures, (Cambridge University Press UK)
Concering the reproduction/IP (yes rms I use and will continue to use that term 
) issues and this interview: Kathy
has given me permission to reproduce the interview. Eben Moglen
suggested it be done with a Creative Commons licence requiring
attribution. I don't use CC licences and have avoided them to date. But
you can see where the various participants stand in regard to the work.
Thus my position on IP matters is that you may link to this interview and credit Kathy Bowrey, Eben Moglen and myself, but if you wish to reproduce it in any other form or cite it please contact me and discuss it.
The
text may contain some errors which I am in the process of proofing. But
I have gone ahead and put it here now as I have referred to it in my Time Machines and the Constitution of the Globe paper, because of recent interest in the question of the enforceability of the GPL in jurisdictions
other than the US, and I wish to use the material as part of some
thinking I am doing about the GPL and the State of Exception (ala
Agamben). Interesting in regard to all of this is the opinion of Professor Hoeren a Visiting Fellow at Oxford regarding the 2004 German preliminary decision on the GPL.
THE INTERVIEW
KB: The
two themes I am playing around with is trying to work out if the
commons that open source creates is a technical commons, but when it
comes to law, whether it is still a form of private ordering - in the
sense that it basically relies on contract, private negotiation. If
it doesn’t go to court there is no accountability or scrutiny about
that side of things. I’m not necessarily critical of that. I’m
not sure what I make of that.
The other thing I‘m not sure about is when technical people talk about open source licensing they think about law in the same way that lawyers do. When you talk to engineers about licenses they seem to have a very very shallow understanding of legal concepts altogether and legal process as well. I find that very interesting, drawing on the work done in relational contracts theory that talks about the way in which business people use law and how it has very little to do with the formal legal understanding.
They are the two themes.
In terms of over seeing the FSF licenses, how do you deal with licenses offshore?
EM: Well, it’s a very important point and let’s start recognising that this is the primary reason that I don’t use contractual principles at all. The goal of the GPL and secondarily the L-GPL is to use pure copyright, in a minimum tool-set guaranteed to exist by the Berne Convention. It is true that there are inevitable adjective ancillaries to any copyright system containing the full set of Berne principles, including local understandings about license interpretation, as well as substantive areas of law ancillary to the license and I will come back to them in a moment. But the goal of the internationalisation of the license is to do something that no other business on earth is stupid enough to do. I am trying to do business everywhere on earth with a single copyright license. The goal of that license is to use only copyright principles and only that that Berne must provide, where it is providing anything at all. So, my assumption is that any Berne member system contains the principle that as to software, in order to copy, make derivative works and redistribute, modify or unmodify software you must have a license of some kind.
I’m assuming in other words that it is always infringing conduct everywhere to make permanent tangible copies, modify and create derivative works, and to distribute. And I always will have distribution before enforceability.
So all that I do is bring an infringement action. It is the defendant’s responsibility to prove license and the only credible license for the defendant to plead is my license, because code is not otherwise available except under that license.
So I don’t recognise an offshore/onshore.
But the problem is that under Australian law it is not clear who would actually have standing to sue, or how much you would have standing to sue.
Why would the copyright holder not have standing to resist infringement?
Because the way in which I understand it, and as other lawyers here describe it, they way in which you’re being a copyright owner is probably that you’re simply one acting on behalf of a number of copyright owners to the extent that everyone has contributed code because it comes under compilation copyright in Australia.
No. Not for my code. I have an assignment from every single person. And a disclaimer of work for hire. And the country of first publication is the United States. Under Berne the copyright status of that work was determined by the state of first publication. It is a treaty violation for the Australian courts to assert that I am not the copyright holder under Berne.
So as long as you locate code, you can claim ownership to it and avoid that question?
It’s not even a question of location. You have again pointed out that the way free software is assembled, put together in the first place, is even more important than the license terms of its distribution. I would not want to have the legal arrangements restricted to the licensing arrangements for the purpose of legal analysis. For the purpose of legal analysis you have to understand how the work was assembled. The FSF principles of assembly are as important as the other.
What your Australian lawyer colleagues have said is correct with respect to works that are compilation works. And with respect to that the Linux kernel is the best possible example. And it would be right to speculate that enforcements on infringement on the Linux kernel might be a problem in Australia. It might be a problem anywhere in the world, including the United States. But when Linus began putting the kernel together he was in Helsinki in 1991, and he was a 19 year old college student, and he had an even more intense disdain for lawyers than he now has. And I’m not going to quoted anywhere on earth about what I think about that.
So I agree with you about that issue. But I think that properly assembled code does not raise that issue in any particular place, provided that the country of its first publication which conveys the copyright status under Berne is a country in which the assignment that was used is fully recognised.
You might want to play that out in your mind a little bit with respect to works containing significant German contributions, not first published in Germany. Where a contributor is making a contribution in a copyright system in which only certain rights of authorship can be alienated, such that for example in a system that divides moral rights and exploitation rights, only the exploitation rights have been alienated, if those have then been incorporated in a work, and your nationality under Berne was set somewhere else, it could be a problem. And of course therefore as you can imagine actually to think these questions through involves some fairly tricky places. Code contributed by people in the UK, Germany, France, in Australia and South Africa and in the United States, first published in the UK, infringed in Iceland- what’s the outcome?
And I won’t for an instant claim to have gone through the entire matrix of possibilities, because they are infinitely large. But I will claim that with a little careful transactional planning questions can be rendered unimportant.
In terms of enforcing your code overseas, how do you actually manage that? Is it the case that people have a watching brief, or that you wait for complaints to emerge?
A very serious problem and it’s one of the reasons why I find that securing co-operation is easier than bringing law suits. There are some countries in which it would be, as a matter of adjectival practice, hard. I cannot for example have pro-bono lawyering in Germany where the fee schedule is mandatory. There are countries in which infringement could happen in which I have no lawyers. This is one of the reasons that we have resisted the translation of the GPL into other languages as an official matter. Because my position has always been that once an official translation was published, that in effect created a closed legal universe in which the official translation was now a document with a nationality that would have to be practiced by lawyers in that nationality’s legal system. And so I have tried to insist with people that before an official translation is published an entire legal infrastructure for compliance and enforcement needs to be in place. That is regarded in certain nationalist entities, I won’t mention the French particularly, to be a significant legal difficulty. They think it to be American and Anglophile and all that crap. And what I say is, look, if you publish an official translation of that license, your going to be on your own with respect to enforcement of it in your courts, cause your going to domesticate it. That can cause problems in relation to adjectival law in particular settings, for example where a French statute says that public acquisitions occur on French language documents. Or with respect to consumer protection laws in Germany where for consumer protection to attach the consumer has to buy on a German document. In general I try and finesse those questions on official translations. But the game at some point might not be worth the candle anymore. Then the question is can we advise, can we oversee, can we assist, but we can’t actually practice in those courts and we have to do our best to help people who appreciate the situation.
How formal and how informal are your contacts in other jurisdictions?
They vary. They vary. We have close relationships with lawyers in some jurisdictions which amount to formal relations, and in others we have no better than the fact that we are seen as people who might be helpful because we know something. I have never practically assumed that I could be in control of litigation anywhere in the world than in my own place.
If it goes to litigation in other jurisdictions, is one of the concerns that the national legal system will in fact start to, even in the English speaking world, changing the contract through the process of legal interpretation and putting local flavours onto to it and twisting it in ways that perhaps you were not intending?
This is the very reason why I have resisted contractualisation completely because contract law is totally non uniform around the world. All it takes is one jurisdiction, which I refer to in my speeches as Unfreedonia, to create the difficulty. By sticking to what I take to be a minimum Berne toolset I hope that I am pursuing closer to harmonised globalised copyright law system. But of course that is inevitably wrong because within the Berne toolset there are many possible differences. The UK copyright laws’ view that the transient copies made by a computer during the execution of software are copies, potentially infringing copies, causes me enormous degree of unhappiness.
Our legal system does that as well, but it has a savings clause if you like, that is still a bit uncertain in terms of exactly in terms of operation what it means.
Yes exactly. From our point of view here at the FSF that implies the possibility that people need licenses to execute computer programs. And that is ideologically unacceptable to us. I don’t want to have to operate to contain infringements, in a culture in which in order to do so I have to concede the licenseability of the execution of code. I do what I can therefore, when I have infringement situations which seem to suggest venue in the United Kingdom to find an alternative. That of course may be no more than a venue question, as infringement is via internet after all.
In your view how common are (a) complaints and (b) the initial stages toward litigation?
Well in my view the answer is that with respect to the portfolio maintained for the FSF complaints are on the scale of some 150 a year, of which investigation creates a reasonable likelihood of an infringing activity of one sort or another in half, and of the vast bulk many dozens a year are immediately adjustable on first contact with the infringer. There are probably somewhere in the neighbourhood of a dozen cases a year that represent something that evolve toward an adversary situation. Of those so far at least 50% are adjustable, given goodwill on both sides or given the leverage which we have as a copyright holder in a system with strong mandatory relief rules, preliminary injunction style. Moreover it is the rare defendant, as I have said in public on many occasions, who actually wants to plead my license and argue that he understands it better than we do about whether he is violating his terms.
Of the people you would be complaining about, what is the range or scale of the operations? Is it right a across the board or is it …?
Yes. Some of those are small private organisations and some of them are large organizations. Adversarility is unlikely in large organizations at the moment because almost all the major players building information technology systems understand the benefits from free software and they face enormous difficulties with vendors, suppliers, workers for being the people trying to kick over the apple cart. There are SCOs in the world of course.
In light of that then, how often do you think those kind of cultural controls are used as opposed to the more formal legal processes?
I think they are enormously important and they are of growing importance. I think that that represents not so much private ordering as opposed to legal principle however, as the equally famous distinction between rules and norms.
In terms of when it actually looks like it is going to litigation what are the practical obstacles in terms of going ahead with that kind of stuff?
Litigation?
Yes
Well it has costs. Right. Those costs are material primarily.
I guess what I am coming around to is the question of, off the record, your attitude to courts and the legal system in terms of what level of trust you have in them, in …?
Oh, I don’t think it is a probably actually. My attitude about technology litigation is that the side that explains most clearly wins.
Ok.
And I believe that we have both clarity and a comparatively simple story. In general what we want is the courts to treat computer programs like they would any other copyrighted works. We have a comparatively small number of situations in which in order to prevail we have to explain clearly to courts why should do something different with software than if it were a book. But in general my effort has been explaining that. To be sure, the other side always wishes to argue when it contemplates becoming the other side that we are doing some really new and different thing. But it is always possible for me to stand there and say, Judge, we are achieving a different social purpose in something very familiar legal technology. There are not entitled to differentiate us according to our purpose. We are copyright holders and we seek only what copyright law always and everywhere guarantees us we may have.
So how do you see that in terms of the speech school of copyright which has been very unsuccessful when its gone to litigation?
Big pause.
Well, look, I mean, I think the primary problem that we’re asking about here is how to coordinate the relationship between a theoretical proposition and what actually happens on the ground. And I don’t feel terribly worried that the courts are going to reach a new theoretical basis for copyright on a time scale which somehow makes it impossible for me to do the work of saying, “he took my client’s computer program judge and you should give it back’.
I
think we are actually operating here and now on a time scale where
copyright law does exactly what we want in the way that we want it
and by the time we have got to a very different story we will have
embedded our principles because we will have embedded our software.
Let me put it in the shortest possible way. Five years from now
there’s going to be not a government on earth that isn’t using
our stuff. There won’t be a court system on earth that won’t be
using our stuff. Every judge will be aware of the fact that if the
system breaks, his computer breaks. All I have to do is stand off
until then.
So it is kind of like an infiltration argument really?
In a sense, although I’m not willing to be quoted on that particular phrase anymore than I am willing to be quoted that it is a viral thing.
Interruption
So how important do you think legal theory or jurisprudence is?
Not particularly.
But isn’t is sort of important to feed into public understandings of what is going on? I mean if you are going to rely upon norms and normativity it needs some kind of loose…
And yet as you have pointed out the primary community to which this is addressed is not concerned so much with the jurisprudence as with the code like characteristics of legal principles. The hackers tend to seethe system in terms of self-executing principles. The businesses that are in the area tend to see the matter in terms of profit and loss.
Sure if the guy on the other side says I ‘m a cancer on the face of American life in the end there has to be an answer to that. But the answer to that isn’t “I’ve got a brand spangled fancy new copyright theory. He’s wrong. And he’s trying to end human life. I’m just the same as the music business.” He’s wrong.
No. I appreciate that. What do you think happened with the WIPO discussions on open source?
What is happening is essentially parties have learned, including Jamie Love from the Center for Democracy and Technology, to move the developing economies toward a better understanding of their international political role, That free software is a human capital development program is now well understood outside of the US. Organizations that are based on one country one vote, and which are usually controlled by patronage, payments to small poor country representatives, are now becoming less controllable. Because there are a large number of countries whose delegations are interested in discussing what owners in the North do not want discussed. That gradually edges the US government toward taking increasingly dictatorial positions about what is and what is not discussable, thus feeding into an increasing understanding that free software is a way of resisting American cultural imperialism. I didn’t invent that dynamic and I have no control over it’s operation and I don’t sign up for any normative approval of it to notice that it is pretty good for my interests.
Yes but at the same time it shifts political focus to bi-lateral agreements and things like that which…
Oh but you must take for granted that IBM is a big, rich company that bribes as many Congressmen as Microsoft. And one of the things that therefore happens is that when Microsoft wishes to see the anti-free software position put on the bi-lateral schedule at the US Trade Representative, and IBM’s guys come around and tell the USTR you should drop that off the agenda, we don’t want it there, the USTR says well we’ve got a very full schedule to discuss bi-laterally with the Chinese, and this is going to make many as many problems for us as it is benefits, we don’t need to put it on the schedule. It is easier for Lois Boland to say nonsense about why WIPO shouldn’t discuss free software, than for the USTR to turn itself into an emissary for Microsoft to the Chinese Government.
I guess so.
I understand your point that the rhetoric of bi-lateralism will then develop as an alternative, but there are limitations on how far it can develop in the real world of the political economy.
Why that is a problem for us in Australia is that, in the recent free trade agreement, of which the text is not out yet, it is supposed to be, that basically our government just completely shafted all the IP agreement for some kind of promise about other sectors of the economy. So for example, we go the 70 year copyright term even though the government’s own competition review of it talking about the implications and pointed out that it is kind of disastrous in terms of balance of payments, there are very few benefits, etc. etc., and there is not going to be any debate about it. We are just getting that. And so, when it moves to the bi-lateral agreements from an Australian perspective what happens is that the IP policy is not decided on the grounds of IP policy at all but it is just a bargaining chip in relation to other aspects of the economy, which is very dangerous for IP in Australia.
I don’t disagree with you but the problem is that you’ve gone from celebrating the Queen’s birthday to celebrating George Washington’s birthday. If you want to not be part of the US Empire you have to get a new government that’s all. That’s a local political issue that isn’t about a question of free software anymore than its about the question of nuclear weapons.
Well its the question about the ability to determine the agenda, and it is not necessarily clear that a change of government would actually change that aspect. In fact it probably wouldn’t.
Well, that may be correct in the sense that the foreign trade implications are such that every government has to take the same position vis-a-vis the agricultural exports or something like that, but it does not follow that a new world information order is beyond the possibility of a left wing government in Australia, if you can conceive of such a thing actually to have.
True, but it is just that that idea is depressingly inconceivable at the moment.
Oh no but that after all is not less true in Cameroon. Whether we are going to have a new world information order in the 21st century is a question to be determined by the macro politics of all the fate of ownership ideas in the 21st century. Its true that if the AIDS numbers in Australia began to approach Botswana levels that would have an effect on the willingness to trade off intellectual property principles against agricultural exports, but heaven forbid that that would be the root to learn.
Ok. Going back more generally by way of finishing off, how would you then describe the legal strategy of the FSF?
The legal strategy of the FSF is to continue the pursuit of the turning of copyright into legal structure to encourage the diffusion of knowledge and the useful arts. We are actually attempting to fulfil the 18th century conception of copyright, by using existing doctrines to facilitate the exchange of knowledge and access to the opportunities for human improvement represented by the free circulation of knowledge. We are paradoxically or ironically using the doctrines of copyright to achieve its historic purpose in a world where copyright had ceased to achieve its historic purpose owing to technological change, and was becoming primarily a reservoir for monopoly rent extraction. In doing that what we do is defend and improve the license, by having it meet the threats to freedom at or over the horizon as early as we can, and to encourage the making and distribution of better software at lower prices. It is a very straight forward capitalist proposition and it is driven to success, not primarily by our cleverness or ingenuity, but by capitalism’s need referred to in the original Communist manifesto to reinvent the mechanisms of its production all the time.
Yey but you’re also starting to get criticism along those lines aren’t you in terms of debates within the movement itself. It’s not necessarily something the rest of the world wants.
Sure. That’s good. This conversation is important.
Does it bother you that it could be swept up into an Anti-Americanism?
My own concern is not with the vindication of American goodness or the capitalist root to the freedom of free software. What Lula da Silva (Brazillan President) thinks about the value of free software is as important as what I think. And it is as important as Sam Palmisano (IBM's chief operating officer) thinks. And it is as important as what Thabo Mbeki thinks. I believe that if we secure the free circulation of knowledge in the 21st century humanity will be benefited. Those people who think that the unfree circulation of knowledge in the 21st century benefits humanity are entitled to their opinion. We are actually running a civil rights movement.
So can you also just sum up for me why it isn’t a form of private ordering?
Well if by private ordering we generally mean contract principles I would say that it’s legal basis is firmly based in the public law side of copyright. It is firmly based in the availability of the courts to prevent public harm arising from infringement of copyright, not based on that contractual consent based activity which we typically mean when we talk about private ordering.
But of course, since I’m a legal realist I have a difficulty with the public/private distinction to start with.
Yes I know, I appreciate that.