Possible solutions to copyright problems

Now, how long should copyright last? Well, nowadays the tendency in publishing is for books to go out of copyright faster and faster. Today in the US most books that are published are out of print within three years. They've been remaindered and they're gone. So it's clear that there's not real need for copyright to last for say 95 years: it's ridiculous. In fact, it's clear that ten year copyright would be sufficient to keep the activity of publishing going. But it should be ten years from date of publication, but it would make sense to allow an additional period before publication which could even be longer than ten years which as you see, as long as the book has not been published the copyright on it is not restricting the public. It's basically just giving the author to have it published eventually but I think that once the book is published copyright should run for some ten years or so, then that's it.

Now, I once proposed this in a panel where the other people were all writers. And one of them said: "Ten year copyright? Why that's ridiculous! Anything more than five years is intolerable." He was an awardwinning science fiction writer who was complaining about the difficulty of retrouving, of pulling back, this is funny, French words are leaking into my English, of, of regaining the rights from the publisher who'd let his books go out of print for practical purposes but was dragging his heels about obeying the contract, which says that when the book is out of print the rights revert to the author.

The publishers treat authors terribly you have to realise. They're always demanding more power in the name of the authors and they will bring along a few very famous very successful writers who have so much clout that they can get contracts that treat them very well to testify saying that the power is really for their sake. Meanwhile most writers who are not famous and are not rich and have no particular clout are being treated horribly by the publishing industry, and it's even worse in music. I recommend all of you to read Courtney Love's article: it's in Salon magazine right?

AM2 (Audience member 2) Yes

RMS: She started out by calling the record companies quote pirates unquotes because of the way they treat the musicians. In any case we can shorten copyright more or less. We could try various lengths, we could see, we could find out empirically what length of copyright is needed to keep publication vigourous. I would say that since almost books are out of print by ten years, clearly ten years should be long enough. But it doesn't have to be the same for every kind of work. For example, maybe some aspects of copyright for movies should last for longer, like the rights to sell all the paraphernalia with the pictures and characters on them. You know, that's so crassly commercial it hardly matters if that is limited to one company in most cases. Maybe the copyright on the movies themselves, maybe that's legitimate for that to last twenty years. Meanwhile for software, I suspect that a three year copyright would be enough. you see if each version of the programme remains copyrighted for three years after its release well, unless the company is in real bad trouble they should have a new version before those three years are up and there will be a lot of people who will want to use the newer version, so if older versions are all becoming free software automatically, the company would still have a business with the newer version. Now this is a compromise as I see it, because it is a system in which not all software is free, but it might be an acceptable compromise, after all, if we had to wait three years in some cases for programs to become free... well, that's no disaster. To be using three years old software is not a disaster.

AM3: Don't you think this is a system that would favour feature creep?

RMS: [airily] Ah that's OK. That's a minor side issue, compared with these issues of freedom encouraging, every system encourages some artificial distortions in what people, and our present system certainly encourages various kinds of artificial distortions in activity that is covered by copyright so if a changed system also encourages a few of these secondary distortions it's not a big deal I would say.

AM4: The problem with this change in the copyright laws for three would be that you wouldn't get the sources.

RMS: Right. There would have also to be a condition, a law that to sell copies of the software to the public the source code must be deposited somewhere so that three years later it can be released. So it could be deposited say, with the library of congress in the US, and I think other countries have similar institutions where copies of published books get placed, and they could also received the source code and after three years, publish it. And of course, if the source code didn't correspond to the executable that would be fraud, and in fact if it really corresponds then they ought to be able to check that very easily when the work is published initially so you're publishing the source code and somebody there says alright "dot slash configure dot slash make" and sees if produces the same executables and uh.

So you're right, just eliminating copyright would not make software free.

AM5: Um libre

RMS: Right. That's the only sense I use the term. It wouldn't do that because the source code might not be available or they might try to use contracts to restrict the users instead. So making software free is not as simple as ending copyright on software: it's amore complex situation than that. In fact, if copyright were simply abolished from software then we would no longer be able to use copyleft to protect the free status of a program but meanwhile the software privateers could use other methods--contracts or withhlding the source to make software proprietary. So what would mean is, if we release a free program some greedy bastard could make a modified version and publish just the binaries and make people sign non-disclosure agreements for them. We would no longer have a way to stop them. So if we wanted to change the law that all software that was published had to be free we would have to do it in some more complex way, not just by turning copyright for software.

So, overall I would recommend we look at the various kinds of works and the various different kinds of uses and then look for a new place to draw the line: one that gives the public the most important freedoms for making use of each new kind of work while when possible retaining some kind of fairly painless kind of copyright for general public that is still of benefit to authors. In this way we can adapt the copyright system to the circumstances where we find it we find ourselves and have a system that doesn't require putting people in prison for years because they shared with their friends, but still does in various ways encourage people to write more. We can also I believe look for other ways of encouraging writing other ways of facilitating authors making money. For example, suppose that verbatim redistribution of a work is permitted and suppose that the work comes with something, so that when you are playing with it or reading it, there is a box on the side that says "click here to send one dollar to the authors or the musicians or whatever" I think that in the wealthier parts of the world a lot of people will send it because people often really love the authors and musicians that made the things that they like to read and listen to. And the interesting thing is that the royalty that they get now is such a small fraction that if you pay twenty dollars for something they're probably not getting more than one anyway.

So this will be a far more efficient system. And the interesting thing will be that when people redistribute these copies they will be helping the author. Essentially advertising them, spreading around these reasons to send them a dollar. Now right now the biggest reason why more people don't just send some money to the authors is that it's a pain in the neck to do it. What are you going to do? Write a cheque? Then who are you going to mail the cheque to? You'd have to dig up their address, which might not be easy. But with a convenient internet payment system which makes it efficient to pay someone one dollar, then we could put this into all the copies, and then I think you'd find the mechanism starting to work well. It may take five of ten years for the ideas to spread around, because it's a cultural thing, you know, at first people might find it a little surprising but once it gets normal people would become accustomed to sending the money, and it wouldn't be a lot of money compared to what it costs to buy books today.


So I think that in this way, for the works of expression, and maybe aesthetic works, maybe this could a successful method. But it won't work for the functional works, and the reason for that is that as one person after another makes a modified version and publishes it, who should the boxes point to, and how much money should they send, and you know, it's easy to do this when the work was published just once, by a certain author, or certain group of authors, and they can just agree together what they're going to do and click on the box, if no-one is publishing modified versions then every copy will contain the same box with the same URL directing money to the same people but when you have different version which have been worked on by different people there's no simple automatic way of working out who ought to get what fraction of what users donate for this version or that version. It's philosopically hard to decide how important each contribution is, and all the obvious ways of trying to measure it are [emphasis] obviously [/emphasis] wrong in some cases, they're obviously closing their eyes to some important part of the facts so I think that this kind of solution is probably not feasible when everybody is free to publish modified versions. But for thise kinds of works where it is not crucial to have the freedom to publish modified versions then this solution can be applied very simply once we have the convenient internet payment system to base it on.

With regard to the aesthetic works. If there is a system where those who commercially redistribute or maybe even those who are publishing a modified version might have to negotiate the sharing of the payments with the orginal developers and then this kind of scheme could be extended to those works too even if modified versions are permitted there could be some standard formula which could be in some cases renegotiated, so I think in some cases probably possible even with a system of permitting in some way publishing modified versions of the aesthetic works it may be possible still to have this kind of voluntary payment system.

Now I believe there a people who are trying to set up such voluntary payment systems. I heard of something called the street performer's protocol. I don't know the details of it. And I believe there is something called GreenWitch.com [transcriber's note: URL uncertain] I believe the people there are trying to set up something more or less like this. I think that what they are hoping to do is collect a bunch of payments that you make to various different people, and eventually charge your credit card once it gets to be big enough so that it's efficient. Whether those kind of systems work smoothly enough in practice that they'll get going is not clear, and whether they will become adopted widely enough for them to become a normal cultural practice is not clear. It may be that in order for these voluntary payments to truly catch on we need to have some kind of... you need to see the idea everywhere in order to... "Yeah, I outta pay" once in a while. We'll see.

There is evidence ideas like this are not unreasonable. If you look at for example public radio in the US, which is mostly supported by donations from listeners, you have I believe, millions of people donating, I'm not sure how many exactly but there are many public radio stations which are supported by their listeners and they seem to be finding it easier to get donations as time goes on. Ten years ago they would have maybe six weeks of the year when they were spending most of their time asking people "Please send some money, don't you think we're important enough" and so on 24 hours a day, and now a lot of them have found that they can raise the contributions by sending people mail who sent them donations in the past, and they don't have to spend their airtime drumming up the donations.

Fundamentally, the stated purpose of copyright: to encourage righting is a worthwhile purpose, but we have to look at ways of ways to achieve it that are not so harsh, and not so constricting of the use of the works whose developments we have encouraged and I believe that digital technology is providing us with solutions to the problem as well as creating a context where we need to solve the problem. So that's the end of this talk, and are there questions?

Index | Introduction | Problems | Kinds | Solutions | Discussion

Douglas Carnall's home page

Richard Stallman's home page

Email comments to douglas@carnall.demon.co.uk

Transcription by Douglas Carnall 10 July 2000.

Copyright 2000 Richard Stallman

Verbatim copying and distribution of this entire speech transcript are permitted in any medium provided the copyright notice and this notice are preserved.