Question and Answer Period
Mr. Tarui:	I should now like to introduce a time of questions and answers with Dr. Stallman.
Participant A:	I am afraid my question is not directly linked to the issue of free software, but I should like to ask Dr. Stallman's opinion on the matter of networks, and national governments' observation of networks. Dr. Stallman asked a question of Professor Sakamura a little while ago, and in a similar way, I was very opposed to national government observation of my personal activities through being connected to a national government network. We hear, however, that terrorists can put many people in danger through creating limitless networks through Internet connections, and this makes me think that steps need to be taken to stop this sort of activity. I realize this problem carries a lot of dilemmas with it, but I should be interested to hear Dr. Stallman's opinions.
STALLMAN:	Well, one very simple way to protect the appliances in your house from being attacked by terrorists through the network is not to connect the appliances in your house to any network. I agree, in fact, this is a paradox that has been known for 2000 years or so: Who will watch the watchman? We have to depend on governments to protect us from outside enemies and from terrorists of various sorts or other criminals.
	At the same time, the government is a terrible danger to our freedom as we are now seeing in the United States and in other countries. The United States has passed laws that have done a terrible blow to our freedom. In the United States you are not even allowed to travel on a train or a bus between states without showing your papers. "Let me see your papers, please" is a term that we associate with dictatorships, police states, not with the United States, right? But in the United States that's what I hear. I can fly from Germany to Italy and I don't have to show my passport or any identification, actually, but to get on a train from Boston to New York I'm being forced to show identification, and I resent that. Every time I see it I feel the presence of an occupying army, the U.S. Army, in the United States. But the fact that this occupying army is the U.S. Army doesn't make it any better.
	So, what do we have to do? Some people have pointed out that the best way to make sure information about us is not misused is to make sure it is not collected, in the first place. That's the reliable safe way, and that's why I have some concerns about pervasive computing because pervasive computing means the information gets collected and it's much harder than to make sure it will not be misused, and that's why I do not carry a cellular phone with me. I don't own one. I borrowed one here but, in general, I don't have one and it's specifically because they can be tracked, and they are going to be tracked. The United States has plans to measure by triangulation the precise location of every cellular phone, and I expect that that will go in a permanent database that the police can check years later. You may not remember where you were on the evening of the 23rd but the police can just look. I don't like that. I don't want the police to have that much power.
Mr. Tarui:	Let us hear from the next questioner.
Participant B:	I should like to ask for a little more detail in regard to patents, software patents. Mr. Stallman talked about patents in his speech. I always consider that patents are not that good a thing in themselves, when I read much of what is written in regard to them. I realize that they are necessary to make sure, for example, that I myself as a small company don't lose out to those with lots and lots of capital. Personally I think that patents are now awarded for just anything, and I think that this way of operating is not terribly helpful. I should be interested to know what Dr. Stallman thinks.
STALLMAN:	That's not true, I'm sorry to say. It is a widespread myth that patents help@to "protect" a small company from large competitors. They don't do that and the reason is that the large companies have many patents themselves and their normal approach is to make everyone cross-license with them. Now, we have a naive idea of the patent system which is that if you invent something then it will be this new device and there will be one patent on this new device and you will have that patent. That may have been true for mechanical devices in the 1800s. It may even be true, to some extent, in some fields today, but it's far, far from the truth in software. The truth in software is that if you write a program you have to use many different ideas, and each one of them might be patented by somebody else, so if you, the small company, develop a software package you were the one who is likely to get sued for patent infringement by various others. And suppose there are a few new ideas in this package which you thought of, well, maybe you could get a patent on those new ideas, and suppose, say, IBM or Toshiba wants to compete with you, they have a lot of patents. They probably have patents on some of the other ideas used in your program, so if you try to use a patent to stop them they just say: Look at all these patents we have that your product is infringing, so please cross-license with us. So you do, and once you cross-license they are now allowed to compete with you. That's why the patent system doesn't really help small companies, not in the domain of software. A typical package is often so big.
	In a different field of engineering where a product is a simpler thing and where any previously known patented ideas may be there used in components that you just buy, in that case maybe it does work that way. Patents affect different fields differently and it's absolutely vital to judge the fields, each based on its own needs. To start with the arbitrary assumption that all fields should be treated alike is not rational.
	The best way to understand how patents affect software is with an analogy. The analogy is symphonic music because a symphony is also very long, has a lot of things, a lot of stuff in it, a lot of details that you have to write and work hard to make them all work together in a good way. But in the various parts of the symphony you are going to use various musical ideas.
	So imagine if in the 1700s the various governments of Europe had decided they would promote the progress of symphonic music by allowing musical idea patents so any musical idea that you could state in words could be patented by somebody.
	Now imagine that it's 1800s and you were Beethoven and you want to write a symphony and not get sued. That writing a symphony that you don't get sued for is going to be harder than writing a symphony that sounds good. Because there would be thousands of patents you have to thread your way around to avoid getting sued, and if Beethoven had complained about this, the defenders of the patent system would have said: Ah, that's just because you are not creative. Why don't you think of some of your own ideas.
	Well, as it happens, Beethoven had a lot of new ideas compared with most composers, but those new ideas had to be used with the old ideas because the old ideas made up the language of music. If he had not used them he would have made something that nobody would make any sense of, nobody would listen to. If you could even do it, then nobody is so brilliant he can reinvent all of computer science from zero and produce something totally different, and if you did it and you developed software, nobody would use the software. They would say: This is too strange. I don't see any of the familiar features that I'm used to. I don't want to learn to use this. The idea of encouraging everybody to reinvent all of computer science is silly. It's not what we want.
Mr. Tarui:	Thank you very much, Dr. Stallman.

Remarks

Forum