Free Software and Beyond: Human Rights in the Use of Software and Other Published Works

Free Software and Beyond: Human Rights in the Use of Software and Other Published Works  (2007)  by Richard Stallman

Gothenburg, Sweden, 2007-05-16

The Speech

[Richard Stallman:]

Hello, finally.

This is an unusual talk in which I'm going to start by explaining the ideas of free software, and then go on to talk about how they extend to some things other than software.

People have been taught mostly when judging what software to use, to judge it superficially in terms of practical convenience only, how powerful is it, how efficient is it, how convenient is it, how reliable is it, what is its cost, pure matters of practical usefulness. Well, of course they count for something, but if that's all you think about you're overlooking the most important thing, which is: how does it treat your freedom and your community, the ethical aspects of using a particular program. People are not taught to even to pay attention to them, or even that they exist. And that's what the Free Software Movement is all about.

'Free software' means the software that respects the user's freedom, respect your freedom. Software that's not free is proprietary software, that is, software that tramples the user's freedom. Proprietary software keeps the users divided and helpless. Divided because everyone is forbidden to share with anybody else, and helpless because the users don't have the source code, so they can't change it, they don't control what it does. They can't even verify independently what it really does.

But free software respects the user's freedom. What does that mean? We have to be more specific. It's easy to say "I'm in favor of freedom", but until we identify which freedoms we defend, we're not saying something coherent. Free software means the user has four essential freedoms:

• Freedom 0 is the freedom to run the program as you wish, for whatever purpose.
• Freedom 1 is the freedom to study the source code of the program, and change it so the program does what you wish.
• Freedom 2 is the freedom to help your neighbor, the freedom to distribute exact copies of the program when you wish, up to and including republishing it.
• And freedom 3 is the freedom to contribute to your community. That's the freedom to distribute copies your modified versions when you wish, up to and including publishing them.

If you have all four of these essentials freedoms, then the program is free software, meaning that the social system of the distribution and use of a program is an ethical system. But, if one of these freedoms is substantially missing, then the program is proprietary software, non-free software, user-subjugating software. So the social system of its distribution and use is unethical, and that means you certainly shouldn't use it, and it shouldn't be developed and distributed, nor that it should be distributed in a way that tramples people's freedom. The same code could be released as a free program or as a non-free program, because this isn't a technical question. The difference here is the difference between two ways of treating the users of the program, it's not about what the program does. It's an orthogonal independent question, a social question rather than a technical question, and therefore a very important question.

But why are these four freedoms essential, why define free software this way? Freedom number 2, the freedom to help your neighbor, the freedom to distribute exact copies to others when you wish, is essential on basic moral ground, so you can be a good member of your community, living an upright life. If you use a program which doesn't give you this freedom number 2, you are in danger of falling into a moral dilemma at any moment, whenever your friend said "That's a nice program, could I have a copy?". At that moment, you will be faced with a choice between two evils. One evil is to give your friend a copy, and violate the license of the program, the other evil is to deny your friend a copy, and comply with the license of the program. Once you're in the dilemma, you ought to choose the lesser evil, which is to give your friend a copy and violate the license of the program. Now why is this the lesser evil? The reason is if you can't help doing some kind of wrong to somebody, it'd better to pick the wrong that falls on someone who deserves it. Someone who has done wrong. Now we can assume that your friend is a good member of a community, of your community, a good friend and normally deserves your co-operation. This is not the only possibility, there're also people who are bad members of the community and are not nice to you, but that's an easy case, if one of those people ask for your help you could simply say "Why should I help you?", or at least you can. But that leaves the other case which is also possible that it's a good friend, and a helpful person, someone that you ought to help. By contrast the developer of the proprietary program has deliberately attacked the social solidarity of your community. So if you can't help doing wrong to one or the other, do it to him, the developer. However to be the lesser evil doesn't make it good. It's never a good thing to make an agreement and break it. Now, there're some agreements that are inherently evil, like this one, and keeping them is worse than breaking them, but still it's not good to make the agreement and break it. And if you do give your friend a copy, what would we have? We will have an unauthorized copy of a proprietary program. And that's a pretty bad thing, it's almost as bad as an authorized copy. So, if you've fully understood this issue, what you should really do is make sure you're never in this dilemma. There are two ways too do that, one is: don't have any friends. That is the method suggested by the proprietary software developer. And the other method is: reject proprietary software. If you don't have a program which denies you freedom number 2, then you don't have to worry about what you would say if someone asks for a copy of it. That's my solution. If someone offers me a program under the condition that I promise not to share it with you, then no matter how attractive that program might be, I won't take it, I say my conscience does not allow me to agree to that condition, that would be a betrayal of the rest of my community. It's important to note something special about this particular freedom. This is the freedom for you to act ethically. If you agree to give up this freedom you have done wrong. Of course the developer is the kingpin of this scheme of wrong, but everybody who participates, joins in doing wrong to everybody else. So, although primarily you're the victims of the developer, you're also partly a perpetrator.

So this is the reason why freedom number 2 is essential, but what about the others? Freedom 0 is essential so that you can have control of your own computing. It may be surprising but there're proprietary programs that restrict even the use of authorized copies, for instance they restrict who can use them, or on which computer or how much or for what purpose, and this is obviously not having control of your own computing, that is to say not having control of your own life. So obviously you must have freedom 0. But that's not enough, because that's the freedom to do or not do what the developer already implemented the code to do, so the developer still has power over you, it's just that instead of exercising it through a license, he exercises it by making the program refuse to do certain things. So, in order to really control your own computing, you need freedom number 1, the freedom to study the source code and change it so the program does what you want. With this freedom, you decide, not him. If you use a program without freedom 1, you can't even tell what it is really doing. Many non-free program has malicious features designed to spy on the user, restrict the user, even attack the user. Spy-ware is quite common, one proprietary program that spies on the user that you might have heard of is called Microsoft Windows. When the user of Windows, and I wouldn't say 'you' I'm sure because you wouldn't use a program like that, when the user of Windows uses the menu feature to search her own files for a word, Windows sends a message saying what word was searched for. That's one spy feature. But there's another. When Windows XP asks for an upgrade, it sends a list of all the programs installed in the machine, another spy feature. Now, Microsoft did not announce these spy features, people had to figure them out, and it wasn't easy. For instance the list of installed programs is sent encrypted, it was not easy to figure out what information was in that message. So it's possible that there're other spy features that haven't been figured out yet. We don't know how much spying Windows is actually doing. But spying on the users is not limited to Windows. In fact, Windows Media Player also spies on the users, it reports whatever the user looks at. Total surveillance. But please don't think that only Microsoft is so evil that it would do such a thing, because Realplayer also spies on the user, and I think Realplayer did it first, after all, Microsoft usually imitates rather than inventing them. The TiVo also spies on the user. That's an interesting case, because the TiVo uses a lot of free software, it has a small GNU+Linux operating system, and some people in the free software community applauded when the TiVo came out. But, the TiVo also uses some non-free software, and it spies on the user. Now, this shows that it's not enough to set the goal of "using free software" because that means using some free software some of the times, and that's not enough. If you want to keep your freedom the goal has to be "use only free software all the time". If you use a non-free program you don't have control. But malicious features goes beyond mere spying, there's also the functionality of refusing to function. When the program says "I don't wanna show you this file", even though it's on your computer, "I don't wanna let you copy part of this file", "I don't wanna print this file for you, 'cause I don't like you". I'm not talking about errors here, these're intentionally implemented functionality to restrict the user. They are known as "Digital Restrictions Management", or DRM, the intentional feature of refusing to function for you. Microsoft Windows is perhaps the World's champion in restricting the user. In fact, the new version of Windows called Windows Vista is primarily designed to advance the restriction of the user. That's why we have a site called 'BadVista.org', which is a campaign for people, even those who are continuing to use Windows to at least not move to this nastier version of Windows. But Digital Restriction Management is not limited to Microsoft. Apple does a lot of it. iTunes is designed to restrict the user, and in our protest campaign against DRM which you can find in 'DefectiveByDesign.org', we have targeted Microsoft, we've targeted Apple, we've targeted others, and we will continue to do so. Google does DRM. When I asked them "Please make the Google Earth client free software" right now, that service cannot be used with free software. They said that the reason it was non-free was for the sake of restricting the user. Google Earth is DRM.

But malicious features go beyond that. There're also malicious features to attack the user, back-doors. One proprietary program you may have heard of that has a backdoor is called Microsoft Windows. When Windows XP asks for an upgrade, Microsoft more or less knows the user's identity, which means Microsoft can deliver to him an upgrade designed specifically for him, which means Microsoft can take total control of his computer and do whatever it wants. Now, we can deduce the existence of this backdoor from known fact, but it may not be the only one. A few years ago, in India I was told that they had arrested some people working on development of Windows, and accused them of working for Al-Qaeda as well, installing another backdoor that Microsoft wasn't supposed to know about. That attempt apparently failed. Was there another that succeeded, we have no way of knowing. But we do know that Microsoft was caught in 1999 having installed a backdoor for another even more violent terrorist organization, the United States government, the same government that kidnapped somebody from Sweden and sent him to be tortured. And we know that Windows Vista's designed to extend Microsoft's capability to attack the user. For instance, it has a feature that allows Microsoft to send out a broadcast that will tell every computer to refuse to work with a particular model of I/O device. Any model of I/O device, they can send a command that tell everybody's computer to stop working with it, and for those people whose computers don't work any more, tough on them. That's Microsoft's attitude.

So, I've mentioned various non-free programs that have malicious features. We can divide non-free programs into two classes: those in which we know of malicious features, and those in which we do not know of any. Now they may have some that we haven't found out, or they may have none. But we can never be sure that a program without freedom number 1 has no malicious features, 'cause there's no way to check. We either know it's malicious or we don't know. So you can never trust a program without freedom number 1. All those programs demand blind faith from the users, and none can ever justify any trust. But what about the ones that don't have malicious features, I'm sure there are some, we can't be sure which ones they are but I'm sure there are. What about them, what can we say about them? Their developers may not do this particular kind of wrong, but they're humans, so they make mistakes. Their code has errors, and the user of a program without freedom number 1 is just as helpless facing an unintentional error as facing a deliberate malicious feature. If you use a program without freedom number 1 you're a prisoner of your software. We the developers of free software are humans too, we also make mistakes, and our code also has errors. The difference is, when you come across an error in our code or any part of its design that you don't like, you're not a prisoner, you're free to change it, because of freedom number 1. We can't make ourselves perfect, but we can respect your freedom, and that makes all the difference.

But freedom number 1 is not enough. Freedom number 1 is the freedom to personally study and change the source code of the program, that's not enough because there're millions of computer users that don't know how to program, and that are not able to personally exercise this freedom. But even for programmers like me, freedom 1 is not enough because there's just too much software, in fact there's just too much free software, for any one person to study it all and master all the source code and personally make all the changes that he might wish. So the only way we can fully take control of our software is to do it working together, co-operating, and for this we need freedom 3, the freedom to contribute to your community, the freedom to distribute, including publication copies of your modified versions, when you wish. This enables one person to make an improvement, and then publish it so someone else can start from there and make another improvement, and publish it so someone else can start from there and make another improvement so that all the users get the benefit of a community of people working together. If there're a million users that want a certain change, or even just would like a certain change, although they may not be aware of this, but suppose they would like this change if it was made, in a free program. We can assume that there are a few thousands in that million that know how to program and sooner or later a few of them will make the change and publish their modified version and all those millions people when they see it and they want it, they can install that version, including those that don't know how to program, because they don't have to write this code, they just have to install a new version. And thus you can see that all the users get the benefits of the four freedoms. Every user can exercise freedom 0 and freedom 2, the freedom to run the program as you wish and the freedom to distribute exact copies. You don't need to know how to program to do these things, every user can. Freedom 1 and 3, they entail programming, the freedom to study and change the source code and then publish the modified version, so any given person can exercise these to the point that he knows how to program. Now that's not an 'all or nothing' choice, just as many people learn to do some kinds of maintenance on their cars, even though they don't run to be professional mechanics, many people can learn a certain amount of programming so that they can make easy changes, even though they don't learn enough to be professional programmers. It's quite useful to be able to make easy changes. But it's true that not everybody can exercise these freedoms making hard changes, but when programmers do so, all the users can then install the new version or not. So all the users get the benefits of the fact that all users have freedom, and the over all results of these freedoms is democracy. A free program develops democratically under the control of its users. Now I'm not talking about the usual kind of electoral democracy, we don't ask all the users to vote and make a decision that's binding on everybody. Rather, every user gets to choose what to do, every user can make a change or not, every user can install this changed version or not, and some total of all their decisions, of all these individual decisions, is society's decision. So if we have a free program, a lot of people want progress in this direction, lots of people will make changes going in this direction, producing a lot of progress. If a few people want progress in this direction, probably they'll make some changes in this direction from time to time, but probably not so rapidly, because it depends how enthusiastic and determined they are. And if nobody wants progress in this direction, then nobody can force it on you, because with free software nobody has power over anybody else, nobody is in a position to impose features on you that you don't want, but with proprietary software yes they do. They can and they do because they have power, power that nobody should have. A proprietary program develops under the power of its developer. The developer has total power, and the program imposes that power on all the users. It's a kind of dictatorship, or perhaps more like a feudal domainwhere the developer is the lord, and all the users are the serfs, and unless they escape, they have to obey.

But what if there're just a thousand users of a free program that want a certain change and suppose none of them knows how to do this sort of work. They can still take advantage of the four freedoms to get the change they want, here's how. They can get in touch with each other and make an organization and the idea's that when they join they pay dues, so the organization collect this money and then hire programmers to make the change. So the organization will have to talk with various programmers to figure out who to hire, so it could ask a group of programmers, "What would you charge?", "When could you do it?", "Please, show us your portfolio so we can judge if you could do this job well", and then it could ask the same questions of a different group, and another group, and then decide who to hire. And this shows us something very interesting. Free software brings with it a free market for all kinds of support and services. This is quite ironic considering that the developers of proprietary software like to call it 'communist'. But in fact they are the ones who have a sort of (??) command economy system. Support for a proprietary program is a monopoly, because only the developer has the source code, so only the developer can make any change. The users want any sort of change has to beg and plead or try to use influence and connections. "Oh please, oh mighty developer please make this change for me". Sometimes the developer says, "Pay us and we will listen to your problem". If the user does that, the developer says, "Thank you, in 6 months there will be an upgrade, buy the upgrade and you'll see if we've fixed your problem, and you'll see what new problems we have installed for you". But with free software, anybody that has a copy can study the source code, master it and begin offering support. Thus, all users that think that good support is very important, should insist on using free software so that they can get their support in a competitive free market. And by the way, another interesting thing is that free software respects private properties, whereas most proprietary software developers do not. Most proprietary software developers say that your copy is not your private property, you're not allowed to own a copy. They claim to own everything. Their idea of private property is: it all belongs to them. By contrast with free software copies really is yours, it's your property and you're free to do with it all the ethical things that you could do. Anything you could do respecting other people's freedom, you're free to do.

Now this fact about a free market brings us to a paradox, because usually when there are a choice of products to do a job, we say there's no monopoly. But when it's a choice between proprietary software products, yes there're monopoly, because the user who chooses this product, then falls into this monopoly for support, but if the user chooses this product, he falls into this monopoly for support. So really it's a choice between monopolies, and the only way to escape from monopolies is to escape from proprietary software, to escape to the free world.

And that's the goal of the Free Software Movement, for everyone to escape to the free world. The free world is the new continent that is built in cyberspace. It's impossible to live in freedom in the old world of cyberspace, where every program has its lord. The only way that we can be free is to escape from there, to migrate to the new world. Because the new world is a virtual continent, it has room for everybody, and we hope everybody will come. And because there never were any indigenous people in cyberspace, we didn't get this continent by displacing any. You're all welcome, and when you all come to the free world, we will have achieved our goal, the liberation of cyberspace.

This, the paradox I told you about, is an instance of an important principle. Freedom is something much bigger than having a choice between a few fixed options. Freedom means having control of your own life. To have the choice between proprietary software packages is being able to choose your master. Freedom means not having a master. And in the area of computing, freedom means not using proprietary software. It was in 1983 that I arrived at these basic ideas and I decided that I wanted to make it possible to use computers in freedom. For my sake and for other people's sake, I wanted to make this possible. But how? At the time it was impossible because the computer is useless without an operating system, and all the operating systems of that day were proprietary. It was therefore the second step after buying the computer was to sign an unethical contract to get the executables of a non-free operating system. What can I do to change this? I was just one man, not particularly famous, with no political experience, and no reason to think I'll be any good at it. But there was one thing that I was very good at, and that was developing software, especially operating systems. So I realized there was a way I could solve this social problem through technical work. All I had to do was develop another operating system, and then I as the author could make it free, and then everyone could have freedom by using this system. Thus I was aware of an important and growing social problem most people didn't recognized as a problem. I had the skill necessary to try to correct it, and probably no one's going to do it if not me. Therefore I had been elected by circumstances to solve this problem, it was my duty. It is as if you see someone drowning, and you know how to swim, and there's no one else around, and it's not Bush, then you have a moral duty to save that person. Well actually that statement is too strong, there're a bunch of people that work for Bush in various countries that you wouldn't have a duty to save. Anyway, in this case the job that needed doing was not swimming it was developing software, a lot of software. So, I decided I would develop a free software operating system or die trying, of old age we can presume, because back then the Free Software Movement that I was just starting had no active enemy. Of course most people didn't agree but they just laughed and walked away, so the obstacle was not opposition, it was a large pile of programs we would have to write in order to have a free operating system. I said 'we' because the next decision I made after I decided to start the project was to ask other people to help. But I didn't know when I started this if we would ever have a complete free operating system, nobody knew. Lots of people said, "Well it's a great idea but it's so big a job you'll never do it". I couldn't say I knew they were wrong, what I said was "I know we have to try". Because when it's a matter of fighting for freedom, you can't afford to wait until victory is within reach before you start a struggle. If you wait that long you will miss most of the opportunities when you could have won. You have to start at a point when you don't know what's going to happen. So I decided to develop this operating system, I decided to make it a UNIX-compatible operating system for technical reasons, so it would be portable, and so that lots of UNIX users would find it easy to switch. And then I just needed a name, and I wanted a humorous name because then it's more funny. And it was a tradition in the free software community that I belonged to in the 1970's that when you're developing something that's compatible with, or inspired by some existing program, you can give it a name which's an recursive acronym saying that your program is not the other one. So for instance in 1975, I developed the first Emacs text editor. It was a programmable extensible text editor, and afterward there was about 30 imitations. Some of them were called this or that Emacs, you know, the most obvious boring kind of names. But one was called 'fine' for "FINE Is Not Emacs", and there was also 'sine' for "SINE Is Not Emacs", and 'eine' for "EINE Is Not Emacs", and 'mince' for "MINCE Is Not Complete Emacs". And then version 2 of eine was called 'zwei' for "ZWEI Was EINE Initially". So of course I looked for a recursive acronym "something is not UNIX", but I couldn't find a word 'blank-i-n-u', and if it doesn't have another meaning, it's not funny. So what could I do? I thought I could make a contraction, get rid of the 'i', and have "something's not UNIX", 'blank-n-u'. So I tried every initial, anu, bnu, cnu, dnu, enu, fnu, GNU! Well, gnu is the funniest word in the English language, why is it used for so many jokes? The reason is according to the dictionary, the g is silent and it's pronounced 'new', so anytime you're going to write 'new' you write it 'g-n-u' and you got a joke. Maybe not a very good joke, but there're lots of them, and sometimes it does make a good joke, for instance when I was a child there was a funny song based on the word gnu. So given a specific meaningful reason to use the name GNU for a particular project, I couldn't resist. But as the name of our operating system, please do not follow the dictionary. If you talk about the 'new' operating system, you'll get people very confused, given we have been working on this system for over 23 years, and people have been using it for about 15 years, so it's not new anymore. But it still is and always will be GNU, no matter how many people mistakenly call it 'Linux'.

How did that confusion get started anyway? During the 80's, the GNU project developed piece after piece of this operating system, and occasionally we found a program that was free and we could plug it in and it would do a needed job. And in 1990, we had almost all the system but one essential major component was still missing, the kernel. We started developing a kernel in 1990. I chose a design that I thought would enable us to finish the job quickly, and the Free Software Foundation hired programmers for working on it. I'm not completely sure why but this kernel still doesn't really work well enough to recommend it. There seems to be some fundamental problem with the design in fact. It's too bad, but fortunately we didn't have to wait, because in 1991, Linus Torvalds then a college student wrote another kernel using the traditional monolithic design, and he got it to barely work in less than a year. This kernel was called Linux, and initially it was not free software, it license was too restrictive. But in 1992, he changed the license, he adopted the GNU General Public License, that being one of the free software licenses, Linux then became free software. Now at this point you might be wondering, what is a free software license? Well, today's copyright law considers every program a literary work, and by default copyright law says people are not allowed to copy or change or distribute a program. So how can we make a program free? The only way to do it is to put on a statement from the copyright holders authorizing the four freedoms, and that statement is call the free software license. So free software licenses are based on copyright law but instead of using it for the usual purpose to restrict people, we take note of copyright law so as to make it clear to assure people that we respect their freedom. There are various different free software licenses, because any such statements that grands the four freedoms qualifies as a free software license. Please don't write another however, we've got too many and it'd be better if we didn't have quite so many. So please use an existing one rather than writing your own. In any case there're two main categories of free software licenses. All free software licenses have to permit freedoms 2 and 3, which is redistribution without changes or with changes. But there're more than one way they can be redistributed. Some free software licenses permit people to redistribute the program putting on additional restrictions, even making it non-free. So there're free programs which have non-free versions, and the users using those versions don't have freedoms, because a middle-man took it away. And I'd already seen this in 1983, so when I launched the development of GNU, I decided I was gonna make sure this didn't happen, I feel it was futile to develop a free operating system if a middle-man could strip of the freedom, and most of its users wouldn't get the freedom. That would fail to achieve the goal of liberating cyberspace. So I wrote a free software license, the GNU General Public License or GNU GPL, which says, you're free to distribute exact copies, you're free to distribute modified versions, but when you do, you have to keep this license unchanged and make the source code available to the user. In other words the user must get from you the same freedom that you got from us. The way we make sure that every user get the freedom is we tell the middle-man, that is anybody who might redistribute it, "you are not allowed to strip off the freedom, you must respect the freedom of others, you must pass on to them the same freedom you got from us". This technique is called copyleft. So a copyleft license grants the four freedoms but says when you redistribute you must always respect the freedom of others. There're also free software licenses that don't have this copyleft. But in fact the GNU GPL was the most popular of all, used for about 70% of all free software packages. It's not just used for packages developed and released by the GNU project, anybody can use it, and in 1992 Linus Torvalds decided to apply it to his kernel, Linux. So once Linux was available as free software, the combination of the GNU system and Linux made a complete free operating system. Linux filled the last gap in GNU, and as a result, the goal, the initial goal of the GNU project that we had set out for almost a decade before had been reached, for the first time there was a free operating system for modern computers, for the first time it was possible to use the PC without giving up your freedom. But at the same time the people who started using this combination got confused. They thought that the whole thing was Linux, they didn't realize that Linux was just one important component, they thought that the whole things has been developed by Linus Torvalds and it existed because primarily of his vision of the World. And that's why this mistake is so harmful. You see, if you call the system 'Linux' you're not just denying the principle developers the credit. Of course that's bad, but it's not the worst thing in the World, it's not the most important ethical issue that we face. There's something much more important at stake here, and that is your freedom. You see, when people think the system is Linux, and they think it exists because of Mr. Torvalds' vision of the World, they tend to follow his, and what is his vision? He doesn't agree with these ideas about freedom and social solidarity, the freedom to co-operate which I've been telling you about today, he never supported them. He has a completely different view in which the important values are just the practical values, powerful, reliable, convenient, software that's cheap. The same values that most people are taught to apply to their choice of software. So when people think that he developed the system they tend not to pay attention to the Free Software Movement which really did develop the system. Of course he's got the right to his view, he's got the right to stay his view and promote his view, but not with our work as the basis. The people who use the GNU+Linux should be aware that the system they're using was developed for the sake of their freedom, for the sake of your freedom, you have to know this, because if you don't defend your freedom, you're likely to loose it. Freedom is frequently threatened, we frequently have to defend it, and when people don't defend it, well you can see the result just by looking at the United States, where basic human rights have been taken away by our own leaders in the name of protecting us from another secondary enemy, and of course their reach is worldwide, they have taken away the human rights here too.

But in order for people to defend their freedom, they have to value it, and in order to value their freedom they have to know what it means. In other areas of life, people have been discussing the ideas of human rights for centuries. There has been long time of debate for what human rights should be, and a long time to spread these ideas around the World. But that does not mean that we always succeed in defending them. But at least we have a base with which to try. But computing is a new area of life, only about, it's only about 15 years, even in the riches countries, that most people participate in computing. So that's not much time for most people to debate about what human rights should be for the users of software, and to the extend that there has been any discussion of that question, it has mostly been under the domination of the large companies that develop proprietary software. People started using computers with proprietary software, they took for granted that that's how it is, everyone around them was saying this is how it's has to be, and only the Free Software Movement was saying anything else. Only we were saying not just that we'll win if you share software, but you should be free to share software. Only we were saying that you deserve to have the control over the software you are using. So the only place where anyone can get an idea that there's another side to that debate is from the Free Software Movement. They won't get it from Linus Torvalds. And nowadays there's a term for the point of view that he and some, and many others agree with, they don't say 'free software', they use the term 'open source'. Many corporations have associated the term 'open source' with a political position which doesn't criticize proprietary software on ethical ground, but only suggests that if you let the people to share and change the software, it's likely to produce practically superior software, and to the extend that they are right I think that's a bonus and I do appreciate software that are powerful, and reliable and convenient. But that's a secondary matter, and we shouldn't get distracted by that from things that really matter like freedom and social solidarity. So you can help us spread the word about this, in just one second, all you have to you is saying "GNU slash" before you say "Linux", or type "GNU/" before you type "Linux" when you're talking about the entire combination. Of course if you're talking about the kernel that Mr. Torvalds wrote, you should call it Linux, he's the developer of that, we should use the name he gave it. But he's not the main developer of the whole system, we're, the name we gave it is GNU, and when you use that right name, you also give us a chance to spread the word about freedom. Of course you might want to do more than just mention our name, you might want to explain these ideas to other people. But that takes times, I'm sure it'd at least take you 10 minutes to explain the ideas of the Free Software Movement. And maybe you won't have time to do that everyday. But if you say "GNU slash", that only takes a second, you can afford to do that. Now, you can't explain the philosophy in one second. But what it will do is give greater effect to our explanation of the philosophy, because when people become aware that this system which is treated by Microsoft as its most dangerous competition, was developed by the Free Software Movement for the sake of everyone's freedom, when they realize it is the GNU system, and then they see articles about GNU talking about freedom, then they'll pay more attention to what we say, because they'll realize that they are the users of our work. If they think they are Linux users, then when they see articles from GNU talking about freedom, they're likely to say, "Well, why should I pay any attention to those fanatics. They're just on the fringe of our community. I'm a Linux user". How ironic it is. They think we're few fanatics on the fringe because they don't know we developed their system. But if they hear you calling it GNU/Linux, then they'll know, and then we'll have a chance to try to convince them, a chance that they pay attention to our arguments, and that they may come to appreciate freedom, and then they'll fight for freedom, and freedom may win.

So that's why when reporters ask me for an interview, I say, "I'll do it if you promise to call the system GNU/Linux and never just Linux, and if you promise to describe what we're doing as free software, and not connect us with the term 'open source'". It's said that all publicity is good publicity, but in fact if they put in somebody else's name, it's not good publicity for you. If they made these errors that would be great publicity for Linux and open source but I've never worked on Linux and I don't support open source, I don't agree with that philosophy, so I'm not interested in doing an interview to give them publicity. I'm not going to try to tell reporters what they should say, I don't make conditions for what they should say about GNU, or what they should say about the ideas of free software, but I do insist that they call the two things by their right name if they want my participation. Otherwise they'll have to write their article without me. And I hope you too will help.

Now, one of the important battle grounds for free software nowadays is the schools. There're proprietary software companies, including Microsoft, that try to convert schools to instruments for directing societies into permanent dependence on non-free software, and they do this by donating gratis copies of their non-free software to schools, or almost donating. And why do they do this? Is it because they are idealistic and they want to promote education? I don't think so. The reason is, they want the schools to teach the students to be users of their software, essentially they want to addict the schools to addict their students to the software, so that after they graduate, it will be too hard for them not to pay. The schools create the dependence, and then the students have this dependence for their whole life. That's the plan, the plan of these proprietary software developers. And it was actually Bill Gates who first announced a plan like this, he said people using unpaid copies would become, would get sort of addicted. And some day Microsoft would make them pay. Now schools would not agree to accept donations of addictive drugs to inject into their students, and they shouldn't accept donations of software that creates the dependencies either. Schools have a social mission, and that mission is to keep society strong, free and independent. And in the field of computing that means to teach people to use free software, not teach them to use proprietary software, every school should teach free software exclusively. There's another deeper reason for this, and that is for the sake of programming education, and that is because there are some people who are natural born programmers. At the age of 13 or so, we want to learn everything about what the computer is doing. So if we use a program we want to learn how it does that. But when the student asks the teacher "How does this program work?", if it's proprietary he can only say "I'm sorry, it's a secret". And thus, there's no opportunity for education. But if it's free software the teacher can say, "Well, this is much as what I know, and if you want to learn everything, here is the source code, read it and you'll understand everything". And that student will read it because he's eager to understand everything, he's fascinated. And the teacher can say, "If you come across any point that you can't figure out, show it to me and I'll figure it out and explain it to you". And this way that student learns something very important, how not to write code. Every time he reads something that's hard for him to understand, it must be very unclear, so he'd better not write things that way. And this is crucial for learning to program well. There's no need to teach these natural born programmers to program, for them it's obvious, they read a manual on programming and they know how to program. But to learn to write good code, you have to read lots of code and write lots of code. So first you read a big program, and then you write to change it, adding some new features or fixing a bug, and you do it again and again and again a hundred times, and then you know how to write good software, at least how to write good clear code, that other people can maintain, and you learn how to make improvements in other people's code, which in a bigger part of the job than writing programs from scratch. So the schools must use free software, so that these people have the opportunity to perfect their natural abilities. But there's a deeper reason which applies to everybody and starting at the first level of schools, and that is for moral education, teaching people to be good citizen, teaching the spirit of good will, teaching the habit of helping your neighbor. Every class has to have a rule: "Students, if you bring software to class, you may not keep it for yourself, you must share it with the rest of the class". This rule will teach people to be people who share. But the school has to follow its own rules in order to set a good example, the school must bring only free software to class. Those of you who are students should organize in your schools, look for allies among the students and the faculties and the staff and press the school to start migrating to free software, and keep on migrating until it gets all the way there. There's no excuse for a school to teach somebody to be dependent on proprietary software.

Now, people, at the end of my speeches, would often ask me, many years ago, whether these ideas apply to anything other than software. They would say, "What about computers? What about cars, or chairs and tables? Shouldn't they be free too?". It's a silly question, because if you really want to ask the same questions, you've got to use the same definition in terms of the four freedoms, and what do we see? Well, the freedom 0, the freedom to use the physical objects, generally you do have that freedom if it's yours and there's freedom 1, the freedom to study and change the source code, well there's no source code, uh, we've already run into a difficulty in applying the question. But you could study and modify the physical objects, and in almost all cases you're free to do so to the extend that's possible, and that could be more or less. Modifying a car, well there're a lot of modifications that can be made to cars. Modifying a computer, well it's harder, it's impossible to modify a chip, you'd just destroy it in the process. And what about the freedom to put it into your copier and make copies and distribute them? There's no copier for physical objects, so it's a meaningless question. And likewise to copy your modified versions, as long as there're no copier that's a meaningless question. So the question that they were asking was just silly. But it was just silly about physical objects or other things that we can copy and modify, and there're other kinds of works that we might have copies of in our computers. So it makes sense to ask the question about these other kinds of works, should they be free, should they have the four freedoms? That doesn't mean the answer has to be the same. Since this has been a long talk already, I would avoid, I would skip the part of this where I condemn the World's governments for having done exactly the wrong things, for showing that they're on the side of the entertainment companies, rather than on the side of their citizens. They have been increasing constantly the power that the entertainment companies have over the general public, constantly reducing the part of the four freedoms that we still have. But what should they do? What would a democratic government do? Well, copyright law is meant to be an artificial incentive for writing works that then benefit the public. In order to get this benefit, we've traded away freedom, which back in the age of the printing press we couldn't exercise anyway, so it was no loss, and therefore in the age of the printing press, this trade, which was the copyright system, was, its function essentially was an industrial regulation. Ordinary people didn't have printing presses, we couldn't copy anyway, except with a pen and ink, and copyright law was never applied to individuals copying with pen and ink, it was only applied to printing and publication. So, the result was copyright law was pretty much uncontroversial, easily enforced, and arguably beneficial. It was uncontroversial because it didn't restrict the ordinary reader in any significant way, so they had no reason to complain. It was easy to enforce because it only had to be applied to publishers, and it's easy to find who's publishing a book, you find somebody selling copies and you say, "Where did they come from?". You don't have to invade everybody's home or everybody's computer to do that. And it was arguably beneficial because in this trade, the user gave up something they couldn't use, so if they got something of some value in return they would gain.

But in the age of the computer network, this is not true anymore. Even if the words of the copyright law were unchanged, it would apply differently in the world of the computer networks. What used to be a restriction on the publishers exercised by authors with a public benefit as the intended result now becomes a restriction on the public exercised by the publishers in the name of the authors. And thus, copyright no longer operates as an industrial regulation, and therefore it is no longer uncontroversial, no longer easily enforced, and no longer beneficial, not as it is. Today enforcing copyright law which has become a (?) system requires draconian punishments. What democratic governments ought to do is get rid of this as it is. But this does not mean I advocate the abolition of copyright. What I advocate is a reduction in the extend of this trade. Governments have traded away on our behalves freedoms which we now want to exercise, so we've got to keep them, but we could still have a reduced copyright system, because there're still some part of the freedom that we could afford to trade away. But there are other freedoms that we must keep, so that we can exercise them. So let me be specific about my recommendation.

First of all, I recommend shortening the period of copyright down to 10 years. I recommend that as a first step, after wards it could be adjusted up or down once we see the results. Now I once proposed this in a panel discussion with some authors at a science-fiction convention, and I expected flames, and I got some. The award-winning fantasy writer sitting next to me said, "10 years! That's outrageous, anything more than 5 years is unacceptable". I was surprised too. I naïvely imagined that the entertainment companies were telling the truth when they said that they demanded ever more power for the sake of the authors. The fact is they don't care about the authors, and they are (?) the authors all the time. This award-winning author had a contract which said with a standard course, that if the books when out of print, the right reverts to him. Well, for practical purposes his books were out of print, but the publisher wouldn't admit it, so he was in a legal dispute with his publisher, trying to reclaim the right to distribute his own books, so that people can read them. And this, of course, is the most basic motivation of authors, they want people to read their work. And copyright was stopping that from happening. He knew that more than 5 years of copyright was no use to him, because his book would be out of print by then anyway. That's the normal case at least in the US, the typical book is (?) within 2 years and out of print within 3. Why should there be up to 150 years of copyright in some cases. And, of course, if medicine advances, then copyright ending 50 years after the death of the author could go on forever. So I propose 10 years as a first step, and then we might decide we need a little more or we might decide we could do with less. But the other question is, what activities should copyright restrict? For the answer of this, I distinguish 3 broad categories of works, not according to the medium, but rather according to their social usefulness, how do we use them, in what way do they contribute to society. They all contribute but in different ways. The first category is the works of knowledge, the works of practical use, works that is used to do a job in your life. These include programs, recipes, educational works, reference works, text fonts, and probably some other things as well. And what they have in common is, you don't look at them to admire them, you look at them so that you can do something, say, get you computer to run, cook a dish, learn a subject, find what a certain word means, print a document. So, these works must all be free, because the same reasons that I explained before apply to all of them. If you use a work to do a job in your life, and you don't have control over what's in that work, then you don't control your life. So you must be free to adapt that work to your needs today. But you're probably not alone, in wanting to do the work in certain way, there're probably other people who have needs like yours, so you must be free to publish your modified version, so that they can get a copy of your version and use it too if they like it. So all four of these freedoms should be available to us for all of these works of practical use. These four human rights are the human rights that the users of software must have, but also these other works. And people might say, "But if we do this, and if there're no restriction on redistribution, will these work be produced?". Well, 20 years ago that might have seem like a real question, now we can have the answer, the answer is, yes. We have the example of the free software community, we have the example of all the recipes people circulate, we have the example of Wikipedia and other free reference works, and now we have free text book projects as well that're starting to catch on. We can see these produced works of practical use, the answer is yes. We shouldn't believe that we must deny ourselves essential freedoms because that's the only way we have any work at all. Of course, if society doesn't produce enough free works, governments can always fund the development of more. This has been a function of governments for at least a century if not two, to fund the development of useful knowledge, and make it available to the public. But let's turn to the next category, works that state the views of certain parties. Now here my answer is different, I don't think modified versions of these works contribute to society, all they do is misrepresent the authors. So I propose a compromised copyright system which says that everybody is free non-commercially to redistribute exact copies. But modifications require permission and commercial use require permission. So this compromised copyright system would provide revenues more or less as the current system does. Then there's a third category, works of art and entertainment. And for this category it took me a long time to reach a satisfactory conclusion about the question of modification, and the reason is modifying a work of art, it isn't misrepresenting anybody, because it's not the mission of this work to represent what somebody thinks, the values art or entertainment contribute to the society is simply in its impact. So it's not out of the question to let people modify it, but there're arguments on both sides, there's an argument against allowing modification, which is that a work of art can have an artistic integrity, and modifying it could destroy that. I think that this is sometimes true, but it's not as true as authors would have it, and to see this you only need to look at how willing they are to let Hollywood butcher their work in exchange for enough money. Meanwhile there're good arguments in favor of modifications as a way to contribute to art. Consider for instance the folk process, which produces works of great richness that were process of repeated modification and adaptation by many people. But if we want to consider only named, known author, consider Shakespeare. Many of Shakespeare's plays were based on the stories of other plays that were published not that long before. If today copyright law had been in effect, those plays would have been illegal. They would never have been produced or published, they would not have been allowed. And the copyright holders would have said, "Well, Shakespeare just wants to do a rip-off of my play, and that would be junk, there's no reason why that should be allowed". And the public would have had no reason to, no basis to be sure that this was not true, not having seen these works ever. But since we've seen them and we can say they're among the master works of human literature. So, modifying other works of art can be contribution to art, it can even sometimes, although of course great art are rare, but it can be great art. So we should let people modify other art. But, it's not urgent, it's OK if people have to wait 10 years before they publish their modified versions of previous works of art. If you use the work to do work in your life, to do a practical job, you must be free do change it today and you must be free to publish the changed version today, but if it's just a matter of making something that is interesting and beautiful, we can afford to wait, as long as we don't have to wait a tremendously long time. So I propose for the works of art and entertainment the same compromised copyright system that I proposed for the works that state people's opinion or experiences or point of view, namely everyone is permitted to non-commercially redistribute exact copies, but for modification or for commercial use, they have to get permission. Now I think commercial use should be defined in a rather broad fashion, not just sales of copies, but commercially displaying the copies, playing them in your airplane for instance. That's commercial use, if the airplane is run by a company, and they should have to pay, why not? I'm not worried about the airline companies' freedom to show movies, but your freedom to make a copy of the movie for a friend, that's important. Your freedom to share with others across the Net, that's important. Now this compromised copyright system will provide money, but will it be enough? In the case of music we have to recognize that nearly all the musical groups that have record contracts actually get nothing when you buy their record. The contracts of these record companies are so exploitative so nasty that in theory a certain faction of what you pay goes to the musician, but in practice they never get it. And this is even if they don't cheat the musicians, this is if they follow the contract, the musicians never get it. So almost all musicians won't lose anything anyway. Now you might ask, "Does this mean record contracts are useless for musicians?". Sometimes but not always, because the record company gives them publicity and that means they have more concerts and more people come and that's how they make money and they can sell things at their concert and they make money that way. The publicity really is useful, but there's a better way, a healthier way to give musicians publicity, and that is, by sharing their music. It's only the long established super stars that actually get money if you buy their records. You see they have finished their first record contract, and at the end, since we assume they are famous, they have clout and they can negotiate another contract where they actually get paid. It's quite rare but it happens, it's a small fraction that is, of course there're probably hundred of them but that's out of the many many thousands who don't, who haven't got there. So yes a super star will make less money, but that's not a disaster. They will live. And you might wonder, "What about movies, movies cost so much to make, right? Don't they need to be able to take away our freedom and squeeze every possible dime out of us?". No. In fact more that half of these stated costs are for publicity, not actually making the movie. And these costs are also inflated as a way of keeping most of the money within the studio instead of returning it to investors. So, and then we realize that a lot of uses of movies is commercial use, and would be covered by these compromised copyright system, and they would have to pay. So, the movies would still bring in a lot of money, maybe half, OK then they can do less publicity. And then they could still make the movies. Or they might make movies that focus on stories of people doing things instead of special effects. They might even make better movies that way. Do keep in mind that there's systematic reason why Hollywood makes crap, it's not that those people don't know their own business, of course they do know their own business, they know the rules of how things have to be done, and those rules systematically produce crap. Now that doesn't mean they should be forbidden to do it, I'm not in favor of censorship, but the question here is whether the continued production of social crap is so socially important that it justifies taking away our freedom. And I think the answer is "No".

Meanwhile some of you might be looking for a way to support the art better than we do now. If we just allow sharing I don't think we'll lose anything important. But we might want to do more, we might want to have a system that really supports the arts well, which we don't have now. So I have two proposals to make, one proposal is a tax on various things connected with using the Internet or computers, and this tax could, the tax money could be distributed directly to those who make the works, distributed based on popularity, but not in linear proportion to popularity. So for music I propose the amount that the musician gets should be based on that musician's popularity but not linearly so, and the reason is a super star can be a thousand times as popular than another fairly successful musician. But it's not efficient to give the super star a thousand times bigger share of the tax money. So I propose using a function that (?? ??), so that if the ordinary successful musician is here, and the super star is way down there, the super star might get 10 times as much, but not a thousand times as much, if we avoid in this way giving most of the money to super stars, we can support a much larger number of interesting musicians adequately with the same total tax money. So this system would work much more efficiently than the present one, in two ways, first of all because, of the money that goes to musicians, it wouldn't be giving most of it to super stars, it would be using it efficiently and second all the money would go to musicians, say the money for music would all go to musicians, and composers, it wouldn't mostly go to businesses. The major record companies from what I've read, pay 4% of their income to musicians. And we can expect another similar amount to composers, well that's 10%, that's very inefficient, this system would give all the money to the musicians and composers. So if it collected one-fifth as much money from us, it could do twice as good a job. And that's not to mention the effect of shifting it from super stars to everybody else, and that would do a far better job of supporting music itself. And we would get rid of the hype factor, which poisons music. Right now money, how these companies will make more money, control music in a way that's not good for music, not good for anybody, except those who get rich from it. Let's put an end to it. However I have another alternative way to support the arts better than we do now, and that is with voluntary payments. Suppose every player has a button which you can push if you like and it would send 1 Dollar or however, I don't know how much a kronor is worth but the point is an amount of money like that sent to the people who made the work, and you can do this when you want, or not, when you don't want. But lots of people really do appreciate good works of art, and are happy to send a little bit of money. The main reason we don't do it now because it's so hard. Now I would be happy to send a dollar to the musician who made music I like, but I don't know how to do it, I have to find where to send it and I have to have a credit card, it wouldn't be anonymous so I don't do it, out of the question. But if I could just push a button and it would anonymously send a dollar from me to those musicians I would be happy to do it. And I think this way we could, because people feel good doing it because it's not a lot of money after all, I think a lot of people would do it. And if not enough people did it, instead of a nasty vicious propaganda campaign trying to tell people that sharing is bad, we can have a friendly, warm PR campaign "Have you send a dollar to you favorite musical group this month? Why not? It's only a dollar you'll never miss it, and don't you own their music?" People would response to this, they would realize that, yes they're asking me to do something that is good to for me to do and I'll never miss it, why not? I think we get a lot of people who might even send a dollar every week. What the current system provides, well I read in the US the average person spends 20 dollars a year on music, and if it's going to the major records companies, about 1 dollar goes to musicians, that's 1 dollar per person per year, any other system that wants to do a better job just need to transmit more than 1 dollar per person per year. So I was really happy, since I was campaigning for this for many years, I was really happy to see the creation of the Piratpartiet, to see these issues actually become public issues for election, made me very happy. I was not happy with their name, and the reason is I think we have to reject this propaganda word that attempts to say that sharing is the equivalent of attacking ships, so I urge people please don't call unauthorized copies "pirate". I can see their point, you know, if they want to take this propaganda word and throw it back in the enemy's teeth, well that can sometimes be good, if it works, I won't complain. I'm not entirely happy with their program though. Now obviously it differs in some ways from what I've said, but that doesn't mean it's bad, but there're 2 specific points where I have concerns. Their program would backfire in the case of software in a really bad way, because their program says that everything would go into the public domain after 5 years, so if Microsoft distributes the binary of a program, that binary would go into the public domain after 5 years, and if we in the Free Software Movement, we release a program as source code, the source code would go into public domain after 5 years, which means that after 5 years, Microsoft could take our source code and put it into a proprietary program, copyleft would only work for 5 years. Well, I would say that's OK if it works both ways, if we could also take the code of a 5 years old Microsoft software and put it in our free software. But that wouldn't work because Microsoft doesn't ever release the source code, Microsoft only releases the executable. So after 5 years the executable would go into the public domain, but we would never be able to use that code in writing free software. So this program as currently designed puts free software at disadvantage compared with proprietary software. Proprietary software would never go into the public domain in a useful way. This doesn't mean we have to be total against their program, the problem can be fixed. And the way to fix it is to require that when they release a program, any program, anyone releases a program or at least when anyone commercially releases a program meant for use in connection with business, we have to brought about the interpretation of commercial use, they must deposit a copy of the source code with the State, and 5 years later it will release that source code in the public domain, and if it doesn't work to build the binary from that source code, that's fraud. So this way, both free software and proprietary software would really go into the public domain after 5 years. I'd say yes to that deal, but they've get to make that deal not be biased against free software. This problem is not hard to fix but they've got to fix it.

So, to end I'd like to mention a few websites that you should look at. The website of the GNU system and the GNU project and the Free Software Movement is called gnu.org. The Free Software Foundation also has a website, it's fsf.org. The Free Software Foundation of Europe has a different website, fsfeurope.org. And you can join these organizations which means paying dues, and supporting these organizations both with your name and your money. You can join the Free Software Foundation through fsf.org, and you can join the Free Software Foundation of Europe through fsfeurope.org. Please join, we need your support. There're also the two specific campaign sites, DefectiveByDesign.org, we hold protests in the street which get press attention as well as letter-writing campaigns. We have turned the term DRM into one that the DRM conspirators are afraid of and uncomfortable with using. They're starting to propose to change the name, because we have made that name stink the way it deserves to stink. But we haven't won yet. Someone proposed they would call it DCE, which maybe stands for "Digital Chain for Entertainment". Go to DefectiveByDesign and sign up and help make our protests bigger. We also have the site for campaign against using Windows Vista. Now of course people who want freedom shouldn't use Windows at all, you've got to defenestrate your computer, which mean either you throw Windows out of the computer, or you throw the computer out the window. But for those who are not yet convinced to escape to the free world, we've got to try as a first step convince them not to let Microsoft put on the new tighter chains with the new improved locks that are harder for them to get out of. And that's why we have the site BadVista.org. So to end, oh I should also mention that we have some stickers to give away on the table outside, and the Free Software Foundation is also selling various things, we're selling lapel pins that looks very nice for, I think it was a hundred kronor and we're selling keyrings for, how much? 15 Euros is what we usually (?). OK, and there're also copies of my book "Free Software, Free Society" that's for? Buy these thing and you will support us, because the money goes to the FSF or to FSF Europe, and I'll also sign books if you buy them.

People have occasionally accused me of having a 'holier than thou' attitude. I think they're mistaken. When I see someone who doesn't strive to promote free software, my goal is not to show I'm better than him, but to convince him to become better like me. But I do have a holy attitude, because I'm a saint, it's my job to be holy. I'm saint IGNUcius, of the Church of Emacs, I bless your computer, my child. Emacs started out as a text editor which became a way of life for many users because they could do all their work without ever exiting from Emacs. Emacs has been extended over the years to do almost everything, and ultimately Emacs became a church, with the creation of the newsgroup alt.religion.emacs. Today in the Church of Emacs we have a great (?) between two rival versions of Emacs, and we also have saints, but fortunately no Gods, instead of Gods we worship an editor. To be a member of the Church of Emacs you must recite the confession of faith. You must say: "There is no system but GNU, and Linux is one of its kernels". The Church of Emacs has certain advances compared with some other churches I won't name. For instance, to be a saint in the Church of Emacs does not require celibacy. Thus, if you've been searching for a church in which to be holy, you might consider ours. However, it does require living a life of moral purity. You must exorcise whatever proprietary evil operating system have possessed any computer under your control, and then install a holy free operating system, and then only install free software on top of that. If you make this vow and you live by it, then you too would be a saint, and you too may eventually have a halo if you can find one because they don't make them any more. People sometimes ask me whether it's a sin in the Church of Emacs to use the other text editor, vi. It's true that "vi vi vi" is the Editor of the Beast. But, using a free version of vi is not a sin, but a penance. And sometimes people ask whether my halo is an old computer disk. It's no computer disk, it's my halo, but it was a computer disk in a previous existence. So, thank you and now I will answer questions.

[Applause]

Q&A

[Coordinartor:]

OK, thank you all very much. [Unintelligible]

[Richard Stallman:]

Oh it's simple, we should have a microsoft, a microphone, sorry. You should have a microphone in one place, and people come up and make a line, and then it's always obvious whose share is next.

[Coordinator:]

Yes, this is what I am suggesting. I would also like to ask the people who have brought chairs in here, because the fire inspection unit is not very happy about that. And while people who come here to ask questions, those who have brought chairs, would you please take them outside, I haven't seen anyone on this side, but I think there are some on that side. But as Richard suggested, people who have questions to ask, please come here, and I'll give you the microphone.

[Richard Stallman:]

I'll do the book signing outside after the questions are done, and someone needs to keep track of the money for FSF Europe and for the FSF.

[Coordinator:]

We'll take care of that.

[Richard Stallman:]

OK, thanks.

[Coordinator:]

First question.

[Audience member 1:]

Hi, I have a question, how about we all sing the Free Software song together?

[Richard Stallman:]

Well, who else here knows how to sing it? (I do.) OK, we'll have to, I think we'd better work on doing it together. We can do that together but I think not now, it will take time for us to learn to sing it together well and so on. We can do that after.

[Audience member 1:]

OK.

[Audience member 2:]

Hello Stallman. I have a question about the GPL license. When you distribute some source code with a program under the GPL, it must always be GPL, right? (Yes.) [Unintelligible] for distribution. What if I have a company where I keep one program within the company and don't distribute it...

[Richard Stallman:]

Then in some sense the question of what its license would be if you distribute it is moot. You don't have to distribute your copy, your version...

[Audience member 2:]

But what if I make it, and I use it in my company, and they don't have the source code written for themselves.

[Richard Stallman:]

You mean, you make it, and then you distribute to the company. Then you have to give them source code, the GPL requires it.

[Audience member 2:]

But for example I use the OpenSSL which would be incompatible with the GPL as license.

[Richard Stallman:]

Well, you couldn't merge those two programs and distribute the result, because you'd have to violate one or the other, they're incompatible licenses. However if you would change it privately then you don't have to worry about that.

[Audience member 2:]

But it's on a server.

[Richard Stallman:]

It doesn't matter, if they're not combined in one program, it doesn't matter if they have different licenses.

[Audience member 2:]

But [Unintelligible] on a server, because it has [Unintelligible].

[Richard Stallman:]

On a server, no, no it's not. We think of using a program on a public server as public use of the program, of course there're private servers too, and using a program on a private server is private use. But there's a different license called the Affero GPL which says that if you use the program on a server, you have to make the source code available to the people who access that server. And if that's the public it means you have to make the source code available to the public. But the ordinary GNU GPL doesn't require that. Next question.

[Audience member 2:]

Thank you very much.

[Audience member 3:]

My question is a little bit shorter. In free software, if you choose the right not to exercise your freedom, but it doesn't give you the right to give them up, doesn't that mean open source is more free...

[Richard Stallman:]

No, this is a total confusion, because the distinction you're trying to make between free software and open source is just a mistake. All the free software licenses are considered open source license. I think this looks like the common confusion when people think that free software is synonymous with being under the GPL. It's not true. Most of the open source licenses are free software licenses, and as far as we know all the free software licenses are open source licenses. So, the most permissive license of all, the X11 license is a free software license, it's also an open source license. The different between free software and open source is the different between criteria for license. We insist on a greater level of freedom than open source does. But they do accept our licenses, so the GNU GPL, they would call it an open source license. I don't because I don't support open source, the big different is in the philosophy. I think the only way to make your question make any sense is if you want to compare two licenses, not compare two philosophies.

[Audience member 3:]

My point is precisely that open source supported by Linus Torvarlds does give you all the rights...

[Richard Stallman:]

You're mistaken, you're just confused, what you've just said already nonsense. It's not wrong it's not right it's nonsense, to say that open source gives you certain things that's nonsense. Open source is a criteria for various licenses, free software is a criteria for various licenses, and mostly they're the same licenses. But so when you say free software does or doesn't give you certain rights to do certain things, that depends which license we're talking about.

[Audience member 3:]

Doesn't open source supports all GPL licenses, plus commercial.

[Richard Stallman:]

You're confused, contrasting GPL licenses with commercial ones that's nonsense. Companies release software under the GNU GPL. Sun releases software under the GNU GPL. The distinction of commercial versus non-commercial is a different question and has nothing to do with the choice of licenses. I think you're trying to talk about the difference between copyleft licenses and non-copyleft licenses. Is that what you're really talking about?

[Audience member 3:]

Not really.

[Richard Stallman:]

Then I'm thinking you don't know what you're talking about. What you're saying doesn't make any sense at all. However, there're people who prefer non-copyleft licenses because, they say, the non-copyleft licenses allow middle-men to subjugate users, and they say that this gives those middle-men more freedom, and I disagree, what it gives them is more power. We have to distinguish freedom and power in order to think clearly about freedom. Control of your own life is freedom. Control of other people's lives is power. All free software licenses respect the users' freedom. The permissive licenses that basically say you could do absolutely any thing to this software, they also allow the middle-man to take power over the end-users. The copyleft licenses don't allow that because we do more than just respect the freedom of all the users, we actively defend everybody's freedom against any possible middle-man that might want to take it away. So the very permissive licenses permit situations that, permit conducts that end up taking lots of people's freedom away. Consider for instance, Apple. Apple decided to develop a new version of Mac OS that was based on a free operating system, and they started with the BSD system, not the GNU+Linux system. And the reason is that most of BSD uses a very permissive license, namely either the original BSD license or the revised BSD license. And both of those are very permissive licenses, and they permit proprietary versions. And Apple initially released the lower half of the system, the part that's not graphical and thus not really exciting as free software, and they call it Darwin, and so they asked the community to help debug it for them, and then they put it into Mac OS X as proprietary software, and they let Darwin die. Is it more freedom or is it less freedom? If you're taking a rather Riddle's mind's definition, if you want to say that Apple's ability to do this constitutes to freedom, then it's more freedom. But when you look at the users it's less freedom, because the people using Mac OS don't have freedom. So, I'm not interested in using an idiot (?) mathematical definition of freedom. I define freedom as having control of your own life, and what Apple did is exercise power and take away other people's freedom. The GNU GPL is designed to stop that, it's designed to defend everybody's freedom, to control her own life, and it's not designed to give people power to control anybody else's life. Next question?

[Audience member 4:]

Hi, my name is David, I'm a web developer, I think there could be a few web developers in here. I'm interested in hearing about the Free Software Foundation's stand on web software and web communities, because one thing I see happening is that rather then downloading and installing the programs on our own computers which is where we have the common ground we can discuss licenses for copying modifying etcetera. We run programs on web servers.

[Richard Stallman:]

When people talk about web applications, they're confusing two different kinds of thing. Some of them is a program you get from a website and your browser runs it. Those'd better be free software of course, and you shouldn't always have to get it from the website, you should be able to make your own version, change it and run it when you talk to that website, and we should set up our browser to make that easy to do. But there're also the web applications that run on a server. And if you're using somebody else's copy of a program to do your computing then you don't control your computing. So this is a different problem with the same bare consequence, that you lose control of your computing. So you shouldn't use a web server to do your computing, ever, because if you do, you've let that web server have power over you. There're other things that web servers do, for instance they give you information, there's no problem there, or they may let you search their data, that's OK too, or they may be doing a computation that you participate in but isn't yours, like say Wikipedia. I sometimes edit web pages on Wikipedia. I don't control the computation that's done there, but there's no reason why I should, that's not my computation, it's a joint collective computation that I chose to participate in. It's Wikipedia's computation and Wikipedia does have control of it, 'cause it's free software running on Wikipedia's servers. So that's OK. But to use something like, say Gmail or Hotmail where it's your computation on your data and someone else controls it, that means you've given up control and that's dangerous. And I've read recently people think it's used to do terrible surveillance to keep, you know, I read that, I'm not sure whether that's certainty or speculation but that if an user of Gmail gets an email with a URL in it, and they click on that URL, it actually goes through Google's site, and then they know who it is, who's looking at that website. This just shows how loosing control of your own computing can be dangerous. But although this problem leads to the same bad consequence, structurally it's different and the solution is different. Free software doesn't solve this problem. If you're using somebody's server to read your email, and you don't control it, you have a problem even if he's running entirely free software, even if you can download the exact source of all software he runs from his server, the fact is you still don't control his copy. Getting the source code would enable you to run your own copy but if you fail to run your own copy it doesn't help you.

[Audience member 4:]

Can we expect to see the Free Software Movement to address this problem [Unintelligible] try to build up [Unintelligible] structure as Google has been doing?

[Richard Stallman:]

I think the FSF can't do it, but the point is having another large organization in the same position doesn't solve the problem, because the problem is not that Google is inherently evil, the problem is you're using a copy which you don't control.

[Audience member 4:]

So basically, web applications are inherently evil?

[Richard Stallman:]

What?

[Audience member 4:]

Web applications are inherently evil then?

[Richard Stallman:]

Yes, for doing your computation, you shouldn't use a web application. You should use software that you can install on your computer, you should do your computing on your computer.

[Audience member 5:]

Hello, my name is (?) and I'm working as manager for small (?) and IT department for a company in Gothenburg. I have a question but also a request. I would like to hear you're view on standardization [Unintelligible] I would like the Free Software Movement to make their own standardization organasition. Because I've been actively for 10 years now trying to implement more and more free software where I work, but the thing is that if you have one commercial software, proprietary software in one end, it tends to need another in the next end, and so on and so on and so on, and there's no gap in between where you can put in better and free software. So what we need is more standardization I think.

[Richard Stallman:]

Actually I disagree. I understand your goal, I agree with you about the problem, but I don't see a solution that way, and the reason is you can lead Microsoft to a standard but you can't make it not stink. The reason that...

[Audience member 5:]

No, but if there's a standard you can compete...

[Richard Stallman:]

Well, but Microsoft won't use the standard.

[Audience member 5:]

No, I don't even think about them actually.

[Richard Stallman:]

The thing is there are standards, and in fact many proprietary programs do follow standards, just as most free programs follow standards that are applicable. Not, of course we don't follow them slavishly. When POSIX, I was in the POSIX standards committee, and I actually coined the name POSIX, but when they insisted on saying the standard for listing a directory was to give the size of a file in unit of 512 bytes. I said, "GNU is not going to follow this standard by default, users want to see sizes in 1K unit, and that's what we're gonna do".

[Audience member 5:]

That option POSIX_ME_HARDER, the best option ever [Unintelligible]

[Richard Stallman:]

Right. I decided we would make an environment variable called POSIX_ME_HARDER and if you define that environment variable, then it will use 512 byte size unit. And somebody else convinced me that people might take offense if we call it POSIX_ME_HARDER, and so I should call it POSIXLY_CORRECT instead. And I went along with him and now I regret it. But the thing is, so we don't follow standard slavishly, but we follow them when we think that the users like it. And some proprietary software developers also do that, but there're some that think it's more in their interest to refuse to follow the standard. That's why Microsoft turds use an incompatible secret file format. They published a standard, and then once they saw that they were dominant they thought, "Well, we can really kill off all competitions if we switch to a non-standard secret format".

[Audience member 5:]

But you've just said it, that's why it's important not to leave (?) the decision to them, that's why...

[Richard Stallman:]

No, no, you got it all wrong. They didn't. Standardizing doesn't help when a company figures that they don't have to follow the standard. The problem is not that Rich Text Format came from Microsoft, I believe that's true, right? The problem is once that they were dominant, they decided that they were better off using a non-standard secret file format. And no matter how good a standard is, and no matter who develop it, that's not going to help you much if a company thinks it's in a position to thumb its nose completely at the standard. And how do they get into that position? Well, from being proprietary software. You see, free software almost always follows relevant standards, and the reason is the users are in control and that's what they usually want. So, if we were to depart from a standard, some users would make the program follow the standard and people would switch to that version, and I can predict this. I don't have to wait for it to happen, I just know that the users want me follow standard so I do. That's what happen when the users are in control. But standards just don't do any good when companies figure that it's, developers of proprietary software think that it's in their interest to ignore the standards. So, in order to make standardization really happen thoroughly, we've got to reject proprietary software. Meanwhile I should point out that another obstacle to good standardization is software patents. Companies that develop proprietary software love to propose standards and get them adopted and then come out to say, "Look we have a patent on it all along, so you're not allowed to implement this standard". Of course wise governments don't allow software patents and they avoid this problem.

[Audience member 5:]

Well, thank you. We disagree but anyway...

[Richard Stallman:]

We disagree about method. I'm all in favor of standardization efforts, that's why I participated in some.

[Audience member 5:]

And I'm all in favour of free software and Free Software Foundation.

[Richard Stallman:]

So our only disagreement is, I'm saying if we want to make sure software follow the standard, we need to insist that the users be in control of the software, and the users will make sure it follow the standard.

[Audience member 5:]

But I as a buyer has at least a slight bit of power because I can at least request the standard, I can't request [Unintelligible]

[Richard Stallman:]

Well you follow...

[Audience member 5:]

I follow the Free Software Foundation's standard on this point. Then I would buy the program

[Richard Stallman:]

You could do both. You could do one or the other. They both help, but of course if you wanna live in freedom you should get rid of the proprietary software in your system. But they both help, and, but the other point is it isn't crucial who develop it, as long as it's a real standard and not patented, free software can implement it and will. And the reason that we can't interoperate with turd formats that way is that Microsoft doesn't want to have interoperability, and if you say to Microsoft, "Please allow interoperability", they're likely to say, "No, we'd rather not". Tough on you. You know, you can fight with them and sometimes you might win. Anyway I think we should go to the next question.

[Audience member 5:]

Thank you.

[Audience member 6:]

Hello, I would (?) about patents. We heard that Microsoft claims that the free software, open source software, Linux...

[Richard Stallman:]

And GNU, don't forget GNU.

[Audience member 6:]

Yes, the question is about patents. We heard that Microsoft claims that lots of software, free software, Linux software, open source software and (?) use patents that belong to Microsoft.

[Richard Stallman:]

That's (?), you know, you don't use a patent, you just implement some algorithm or technique or feature that's patented. The thing is, in the US, every large program probably infringes hundreds of patents. That's to show how stupid the software patent system is, how dangerous. Every large programs is going to infringe hundreds of patents. Now, a couple of years ago, a lawyer who supports our community did a study, and he reported he had found 283 different software patents of the United States that are infringed by something in the code of Linux. That's not the whole GNU/Linux system, that's Linux the kernel. He didn't say which patents they were, because if he had done that, he would cause legal difficulties to the developers of Linux, he didn't say. But Linux is just an example of a large program, it would be the same for every large program, there's no way to develop powerful software except by infringing patents, and the conclusion is we have to abolish those patents. Or at least protect software from claim of infringement. Next question.

[Audience member 7:]

Is it true that Eric Raymond is the antichrist of the Church of Emacs?

[Richard Stallman:]

Um, no. Next question. We wouldn't have an antichrist because this is not a Christian church. We would have an antiemacs.

[Audience member 8:]

Yes, if I would write a program, of course I'd like to have users but if I write a program for Windows, for example, I would get more users. Is there a way to bypass this...

[Richard Stallman:]

Work to make GNU plus Linux more popular than those is one way. You can also make your program portable and have it able run on various systems and then you get even more users. But I do want to point out that, although I share your wish, you know, I want my software to be popular too, that we shouldn't let this desire for personal success overcome our support for freedom.

[Audience member 8:]

OK, and one more question, I have before, a couple of years ago, I tried to use the GNU/Linux system, but I find it rather slow, more slow than Windows, how come...

[Richard Stallman:]

Well that surprises me because generally I've heard the opposite, but since I don't use Windows, I really don't know. Lots of people seem to find it adequate, so I guess that's good enough.

[Audience member 9:]

Hello, I'm Alexander, I'm a student at Chalmers. I was wondering what is new in GPL3?

[Richard Stallman:]

I'm afraid that that would take along time to answer, and the reason is there is no particular one thing. The basic idea of GPL version 3 is the same as GPL version 2, establish and defend freedom for all users. So the changes are all in details for many different reasons, so there's no simple answer.

[Audience member 9:]

I was thinking more about the hardware part, something about the restriction in hardware or (?)

[Richard Stallman:]

OK, so you're talking about one particular thing. There's a practice that we call 'tivolization', where they embed software in their hardware products, they design the hardware so that if the user modifies the software, the hardware shuts down. It's named after the TiVo because TiVo does this. The TiVo contains GPL-covered free software, some GNU programs and Linux, and they provide you with the source code as the GPL version 2 requires. But if you modify the source code and compile it and install it in the TiVo, the TiVo doesn't run at all, it shuts down. This means that they are respecting in a literal sense freedom number 1, but in a real practical sense they do not. And that is why we decided in GPL version 3 to prohibit that. So if they want to design the machine such that some sort of checksum or key is needed to run a given version of a program, they have to give you the means to make the checksum from your version so that or the key to sign your version, so that your version can run in your machine.

[Coordinator:]

OK, so let's have the last and final question.

[Audience member 10:]

I'm one of the authors of web community for free art, that is free as in freedom. Today we require the author of the art to license the work under Attribution-ShareAlike, which was sort of like GPL, and my question here is, do you think—and the reason that I chose this license is I assumed that it was easier to understand for an artist because it doesn't speak about 'software', it speaks about 'work'--so my question to you is, do you think that it's a benefit for us to use the GPL instead.

[Richard Stallman:]

I'm not sure I'd have recommended the GPL for visual art, and I recommend the GNU Free Documentation License which isn't only for documentation. But that license as such is a free copyleft license, there's nothing wrong with that. The reason I don't support Creative Commons thesedays is that Creative Commons has other licenses which are so restrictive that they should never be used for anything. As I've said, I think the minimum freedom for art is the freedom to non-commercially redistribute exact copies, and most Creative Commons licenses give you that much freedom, so this is not free art, it's share-able art, and I think that ethically that's sufficient. But there're some Creative Commons licenses that don't even permit the non-commercial sharing of exact copies, for instance there're the developing countries licenses, and there're some of the Sampling licenses that do not permit sharing. And when I discovered them, I had to say I can't support Creative Commons anymore. And the reason is, people tend to imagine that Creative Commons actually describes a policy. Creative Commons has no license policy, it just makes licenses all sort of different ways, these licenses don't have any particular principle underneath them. And Creative Commons is perfectly honest about this, they don't say that they defend any given freedom, they say that their goal is to aid copyright holders to using their power more flexibly, they don't say that there's any freedom that the public should have, and so what we find is that there're a lot of people who imagine that if they say, "We're using a Creative Commons license", they have said something substantive about their policy. They don't realize that that's not true. And so, while there're some Creative Commons licenses that are good for bigger or smaller ranges of uses, I don't think we should support Creative Commons. The problem is that Creative Commons focuses people's attention on their brandname, and away from the specifics of the license used by any work. You see lots of works that say, "Licensed under a Creative Commons license", they don't say what that actually permits, instead they just have a link, so if you follow the link you can find out what's permitted but they don't directly tell you. So, instead of telling you what their license policy is even in the most general terms, they just tell you that they're using a certain brand of license. I think that's a mistake, I think if they want to use that license they should say, "Licensed to permit sharing", or "Licensed for non-commercial copying" or some other brief description of what their actual policy is, and of course you can follow the link to find all the details. Instead of focusing on a brandname they should focus on what they've actually decided. By the way, I didn't mentioned it before but there's a term you should never use. And that is the term, quote, intellectual property, unquote. That propaganda term carries bias and confusion. The bias is obvious, if that term is applied to, say, copyrights, it encourages you to imagine this is some short of property that authors simply have to have no matter how much it hurts society. The right way to think about copyright is to think, what rules do we want to have in society for using published works, and which rules will lead to good results for a society. But the entertainment companies don't want you to think about the question that way, they want you to suppose that copyright is some sort of natural right that we simply have to suffer with no matter how much harm it causes, because they're entitled. So to reject that bias we should reject the term. But the term is even worse than that it's not just bias, it's also confusion. Because this term is defined to include many different laws that are not similar. For instance, patent law is also called intellectual property. But patent law is nothing like copyright law, they're different in almost every point. They apply to different things, they have different results. So the result is anybody who start to think about, quote, the issue of intellectual property, unquote, is already so confused that he hasn't got a chance of thinking clearly about copyright law or about patent law or about trademark law or about any of the other laws. So the way to encourage clear thinking is to call each law always by its name and never mix them up, always talk about one of these laws then we can think about that law based on what it actually says, based on the consequences that it actually has, instead of mixing it up with other laws that're totally different. So anytime someone talk about, quote, intellectual property, unquote, he's either confused or trying to confuse others. If you want to see the propaganda use of this term, look for it in Microsoft's statements, you'll get a very good clear picture of how they're trying to lead people's attention away from the specifics of any given law, and choose some vague confused idea that it's a natural right we just, a vague natural right that we simply have to suffer with.

[Coordinator:]

Thank you Richard.

[Richard Stallman:]

So, I guess we're done now, finally. So happy hacking, I'll go out now to sign books, and please join FSF and/or FSF Europe. We need you to actually actively support us, it's not enough just to be privately in agreement.