On Monday, May 31, 2004 at 20:15 +0100, Malcolm Tyrrell wrote:

Anyone else up for a bit of transcription? Minutes 75 and upwards are still avaliable.

I have done up to the end of Patricia McKenna's presentation (up to Ciaran saying "any questions"). That's 80 minutes. I'll post once I've done a basic proofing. There are a few indistinct bits here and there but it's mostly intact.

If someone could do the questions that would be just fine! There's 20 minutes of questions, even doing 5 minute chunks would be a great help.

[glenn, I've cc'd you since the fsfe-ie archive isn't updating - hopefully there's been no communication problem...]

Glenn Strong Glenn.Strong@cs.tcd.ie writes:

...

If someone could do the questions that would be just fine!

ok, I've done from 78:00 to the end:

Is there a canonical document to proof read?

There are some "???" left in my transcript, but my ears are now repeating the same non-comprehensions. I'll proof someone elses. Special care should be given to the words of the MEP candidate guests before we tag the document has final.

==================8<=======================

Patricia McKenna: Best of luck, thank you. [applause]

ciaran: Will you take questions?

RMS: What did you say?

ciaran: Will you take questions?

RMS: sure.

ciaran: I think we only have one microphone.

RMS: The best way to do this is you stand in one place with the microphone and anyone that wants to ask questions form a line. And that way we always know who's next, it's not a matter of arbitrarily picking a hand. It's much better that way, and it's less hard than getting the microphone to people.

And please speak loud and clear, I won't understand you if you don't.

AM1: I'll try, but I'm Italian though so the language is going to be an issue but... I give it a try. I have a couple of questions, ?? topic ?? a couple of months ago and ?? better answers from you, well, and ?? could be patented, ??

RMS: I'm having trouble hearing you. Please try to pronounce every consonant clearly so I can recognise your words. I'm hard of hearing, make an effort please.

AM1: So, how does it work the patents in different countries?

RMS: Each country has it's own patent office, it's own patents, with it's own criteria. And these patents then restrict everything done in that country.

AM1: I have an idea about that, if I write software and I put it on the Internet, so my software isn't patentable in Europe, but then it's covered by some US patents, should I be worried

RMS: Probably not, because, they might try to sue you in the US, but you can probably just thumb your nose at them. That is, even if they win in the US, they can't enforce it on you to collect any money if you don't have any assets there - but if this happens you really should ask a lawyer what to do.

AM1: Second question is, is it a problem [??...attachments...??]

RMS: No. No. I think the problem is patenting software ideas, because, to write a program you have to combine lots of ideas, and you can't do that if the ideas are not available to be used and combined.

[1:22:33] AM1: Is the problem not patenting in general, not in software

RMS: I don't agree. Every patent covers an idea. Every patent says: "here's a certain idea which you're not allowed to use". Whatever it says in the patent, that's an idea. An idea is any product of thought, anything you can think - now, there are people who use the word quote-idea-quote in a strange way, trying to make a distinction. They like to claim that it's impossible to patent an idea, only something else, but they're using the word idea in a peculiar way. Whatever is patented it is an idea. And they're also wrong. They're trying to claim that only a specific implementation of something can be patented, but that's not true in any field. But the thing is, if patents were really very specific they would be irrelevant. The reason they matter is precisely because someone else might get a patent that covers a program that you're going to write. That's why he wants that. If it didn't cover your program, he wouldn't gain anything by having that patent.

AM1: That's it?

RMS: If he only wanted to cover the code he wrote, he doesn't need a patent, he has copyright covering the code he wrote.

AM2: Thank you for taking my question, it is: There is an organisation in the United States called the <a href="http://www.pubpat.org">Public Patent Foundation</a>. Can you offer an opinion on that organisation and it's work?

RMS: Well that organisation is attempting to cause re-examination of certain patents. Now if this gets rid of - if this succeeds in getting rid of some harmful patents, that would be good, but I don't think they can get rid of all of them that way. So it's like, y'know, somebody going around with a mine detector removing occasional mines. It's a step in the right direction but it's not going to make it safe to walk around.

AM2: Sorry, just following on to that, I do believe that it will - possibly their work will heighten the awareness of the problems with patents, and

RMS: It may. On the other hand there's a danger that people will think, "Oh, well they'll get rid of some of the invalid patents, and then the system will work". Well, I think we should get rid of all the bad software patents, which means all the software idea patents, because they're all bad because they all restrict how you can use your computer.

Some of them may be legally invalid if the law is carefully applied, but other may be legally valid if the law is carefully applied, but they all are harmful so I call them all bad. Unfortunately, only the ones which would be invalid can be eliminated perhaps by this method. Also, there are [??audio-skip??] reexamination, it leads to a procedure that goes on between the patent office and the patent holder, and the patent holder may get the chance to rewrite the patent - so that it avoids the prior art that has been shown but it still covers what a lot of people want to do.

The way it works is: you come up with prior art, which is essentially proof that people used an idea that falls was within the scope of the patent before the application was made. And then question is: how far away does something have to be from this point that you can show was published before, in order for there still to be a patent on it. And so that becomes a question of: what is the threshold that the patent system uses? And in fact their threshold is very low, the result is that points of prior art that you can show are sort of like, American soldiers in Vietnam or Iraq, they control the ground they stand on, and that's all. So, you may be able to prove that this particular idea was known fifteen years ago, but they can still draw the patent to go around it, and cover the things that people want to do today.

AM3: Hi, I wanted to here - programming language, it is a language - so theoretically you can say that when we program, we express ourselves. By software patents, I would say actually, we limit free speech. Would you agree on that?

RMS: I would agree but I don't think that legal argument would convince courts by itself to rule against software patents. Not once you have EU directives overriding constitutions, and if there's an EU directive that authorises software patents, I'm afraid that this kind of argument about human rights won't get you anywhere.

AM3: And what actually, ...in the US also.

RMS: I really don't think so. Maybe someone someday will get a chance to try it. I am not a lawyer. The questions of what a court might or might not think, I'm not really that interested in, I'm the wrong person to ask.

I'm trying to talk about what's wise, sound policy, and what's foolish policy. As a programmer for more than 30 years, I think I can say something about that. But you should ask a lawyer about what courts are likely to believe or not believe.

AM4: Quick question, I hope it is anyway, I think one of the most famous patents is on the progress bar.

[1:28:52] RMS: Well, that's one example of a European Patent Office software idea patent.

AM4: I'm just curious, is there any way of circumventing that, for example...

RMS: I haven't seen the text of that patent, so I have no idea. The only way you can tell how many other different things are covered by that same patent is to talk with a lawyer, looking at the text of the patent.

AM4: So if instead of having a progress bar, you had a little picture drawing itself...

RMS: I have no way of knowing this. We can't tell from what we know. If you get that patent, and look at it with a patent lawyer, maybe you could figure it out.

AM4: And there in lies the problem

RMS: Well, that's a part of the problem. But, y'know, I wouldn't say that's the whole problem. Not at all. Just the fact that you're not allowed to use a progress bar, of a standard everyday sort is a problem.

AM5: Can I, an algorithm published be patented, if published in a journal, or does it have to be in the patent process beforehand?

RMS: It depends on the country, but in general in Europe, once the idea has been published, then any subsequent patent application is considered invalid.

That doesn't mean that the patent office knows that you've published it - so they might accept the patents application and give you the patent. At that, given proof that you published it, that is, ten years later there's got to be a copy of the publication - that you could use to prove it really has been published, that would be evidence, you could take it to court to prove the patent is invalid. If you've got the money to go to court.

ciaran: is there one question from the other side?

RMS: people can - it works so much better when people make a line.

AM6: The question is, how does the two hundred, three hundred people in this room lobby, the current and MEP candidates to try to get the second reading, the way we want it?

Patricia McKenna: First of all, we don't know what the new Parliament will be, and how people will vote with the new countries coming in and what the position will be, -- we had an interesting vote just before we broke up because the ten accession countries had MEPs were in the Parliament, and had the right to vote - and that was the third time we were asked to vote on the PNR, that was the issue of the data being handed over the the U.S. without coming over proper EU data protection laws.

We were very worried that the Parliament was going to this time vote by majority to accept what the commission was doing, but interestingly enough it didn't.

So, I'm just wondering next time, it depends on, I suppose, the make up of the Parliament, first of all, particularly in relation to the new countries, and how they're going to be influenced by the lobbying from the big multinational interests.

In relation to the Irish MEPs themselves, I suppose it's important to try to get some sort of commitment from all the people that are running in Ireland, I'm not sure how many candidates are running in total but I think it's about twelve in Dublin and thirteen in the East. But it would be really important to try and get commitment from them, but at the second reading it's going to be even more difficult to get a majority that's required to ensure that the existing position of the Parliament is upheld.

But also it's not just the Parliament itself, it's very important that that's your priority to get the Parliament to adopt a similar position to last time. Even though we didn't think it was great, I have to say, it was better than what was being proposed, but the second thing is to get the government, and that includes local government, [????] there's been very little public debate on this, at a national level. And how it's going to effect small and medium-sized enterprise, in relation to this.

I think there's two challenges, [????] and also to our government.

AM7: I just wanted to, get your opinion on, you seem to have focused on software patents, as if ?? you couldn't draw a line ?? many patents were drawn against physical systems, but they could be applied to a software version.

RMS: Well they shouldn't be.

AM7: Well, that's ??

RMS: I can't hear you, what are you saying? It's really unfair for you to say things I can't hear, and then I'm supposed to respond to them.

AM7: A classical example would be the concept of semaphores and signals. ?? railways.

RMS: What? What? what?

AM7: The system of semaphores and signals. They were part of a system for railways.

RMS: That has nothing to do with anything going on in a computer. Patenting a mechanical system for moving up a bar to tell a train what to do, that patent wouldn't cover anything going on in a program. That's a patent on a mechanical device.

AM7: But if ?? concept ??

RMS: You're saying this, I don't if it's true what that patent covered, I haven't seen it, have you actually read that patent?

AM7: It's expired now, but it would have covered semaphores and signals in Unix.

RMS: Then, I guess we have to say that those patents shouldn't be allowed either. Allowing someone to patent building a signaling machine to display lights to tell a train what to do, okay, I'm not an expert on how to run railways, but such a general patent that would cover what goes on in software would have to be rejected, or else we have to perhaps say we would have to say that it doesn't apply to the software.

AM7: ?? business methods ??

RMS: well, that's exactly part of the problem, allowing software idea patents allows patenting of business methods and that's a stupid thing in itself.

[1:35:55] AM7: ?? ??

RMS: I don't know what you mean, business methods shouldn't be patentable in any domain.

AM7: But ??, I guess there are fuzzy areas.

RMS: There are fuzzy areas everywhere, they are not an important problem. Courts are constantly dealing with fuzzy areas. That's what courts are for. And so basically, when people want a policy to stretch all the way across, they make the idea that to have to draw any line in between would be impossible, but that's what courts are for, to interpret the lines that are drawn between many cases.

Consider the fuzzy line between drunk and sober. There's no one natural right place to draw a line between drunk and sober. And I can say this is - you can see this is absolutely true. It's not just that we can't see the place to draw the line, we know there's none. Does that mean that laws can't distinguish between drunk and sober? obviously not. It just means that there's going to have to be a line drawn somewhere, and it isn't going to be at one place which was the only right possible place. That's alright, we need a line between there somewhere. We don't want to treat drunk drivers and sober drivers alike. So we'll have a line somewhere, and the courts will figure out exactly some place for it to go, and it won't be the only possible place, y'know if they'd put the line a little bit this way or a little bit that way it would be okay too.

AM8: Just one question, I was talking to people who are involved in drawing up some of these documents. And I put the case of, "software patents are bad", all this, but they say things like "we don't have a choice, the WIPO demands that we implement software patents".

RMS: There is not WIPO treaty about this. WIPO are thinking about a substantive patent law treaty - they're making it all up, they do this constantly. Some of them may claim that TRIPS, or TRIPES as I prefer to call it, it's the "Trade Restricting Impediments to Production, Education, and Science". Some people will claim that TRIPES requires software patents but it doesn't. They make all sorts of bizarre all sorts of claims hoping that they can escape from the conversation without looking like they're mistaken on the spot. They say, they say -. I gave a speech on this topic last Friday in London, and messages were read out from someone from the UK Labour party and the Conservative party, and they were trying to give the impression that the directive as the Commission wrote it does not authorise software idea patents. Well, it's not true but they hope they could get away with the pretense. If this directive goes through, there'll be no hope of ever revising it because to revise it we need to get all three branches of the EU to work together. To block it we only need one maybe but to revise it we need all three. So they figure if they can get it across there will be nothing we can do and it won't matter if they're seen to have been saying utter nonsense.

ciaran: Ok, we have to close the hall, do you have any closing comments?

RMS: Ok, well: No I don't.

[applause]

ciaran: I just want to make a closing comment myself. The website for IFSO is ifso.ie, it should contain information for how to work on software patents, or on the patents directive. And also other information on the legislative process and how to lobby the European Parliament and the European Council. So if you want to help, take a look at the website.

[1:40:57] RMS: I have two websites I want to recommend also, about this issue: www.ffii.org - that's the site with the most information about the situation in European regarding software idea patents. And then there's another site which is very good for introducing people to the issue, and that is softwarepatents.co.uk

ciaran: Ok, sorry, one second, we have one last MEP candidate who has just: Ivana Bacik

Ivana Bacik: Sorry, and sorry to be taking your time and I'm very sorry to be arriving late, I was out canvasing. My name is Ivana Bacik, I'm running for the Labour party for the European Parliament elections this year. As many of you probably know, I teach law in Trinity, in the law school in Trinity, but I'm just here because this is an issue I feel strongly about. I would like to lend my support, and as much support as I can from the party as well, to the campaign for free software. Thank you, and sorry to take up your time, thanks.

On Tue, Jun 01, 2004 at 07:44:37AM +0100 or so it is rumoured hereabouts, Conor Daly thought:

I'll attempt to proof minutes 0-30 this morning...

Minutes 0-15 proofed, upload later.

Conor

On Tue, Jun 01, 2004 at 10:01:43AM +0000 or thereabouts, Conor Daly wrote:

Here's minutes 0-15 proofed. I've attached both a proof copy so you can see my edits and the finished copy.

And minutes 15-30 proofed. Same manner but I've marked them up with some HTML tags.

Conor

On Tuesday, June 1, 2004 at 15:41 +0100, Brian Brazil wrote:

On Tue, Jun 01, 2004 at 06:47:50AM +0100, Ciaran O'Riordan wrote:

...

[1:40:57]

<Small snip> > softwarepatents.co.uk

Site doesn't exist. There is a softwarepatents.com though.

Hmm, you're right. - google finds (lots of) links to it, but none seem to be later than 2003. Perhaps it has expired and not been replaced. It's clearly what he said in the talk. I'll make that link in the transcript go to a footnote offering some alternatives (suggest away - softwarepatents.com and swpat.ffii.org are two I'll use).

Also found softwarepatents.org.uk but some has messed up the permissions by the look of it.

On Tuesday, June 1, 2004 at 10:32 +0000, Conor Daly wrote:

On Tue, Jun 01, 2004 at 10:01:43AM +0000 or thereabouts, Conor Daly wrote:

...

Here's minutes 0-15 proofed. I've attached both a proof copy so you can see my edits and the finished copy.

And minutes 15-30 proofed. Same manner but I've marked them up with some HTML tags.

Excellent stuff, Conor, thanks. I've updated the working draft with all these changes.

http://www.cs.tcd.ie/Glenn.Strong/Documents/ifso/rms-2004-05-24-transcript.h...