RMS Speech at MIT, Chrompet

a guest Nov 7th, 2011 93 Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
  1.                         Transcripts from the Speech of               13th, March, 2002
  2.                                 Richard M. Stallman
  3.                                                 at the
  4.                         Madras Institute of Technology,
  5.                                 Chromepet, Chennai
  8. After the Welcome Address (which was unfortunately not recorded on the
  9. tape) the Chief Guest, Mr.Richard M. Stallman took his hands off his
  10. laptop and came forward to deliver his speech. Here is the transcript:
  12. So I guess I have to stand exactly here inorder for it to work right.
  13. please raise your hands if you cannot hear me. oh oh, nobody is laughing
  14. I guess that means the sound system is not working.  [Laughter]  now
  15. what is *this* microphone for? oic. so I can use *this* instead. there's
  16. only one. this is not *stereo*. well, ok. we'll see what happens.
  18. If i hunt around enough I should be able to play music with this.
  19. [Laughter] well, The subject of this talk is the free software movement.
  20. but really this subject is an ethical, political question. the question
  21. is "what rules should society have for using software?".
  23. But now well it sounds like its still working. alright. I'm hoping to
  24. find a place where we won't have the feedback going on. I'm going to get
  25. feedback from you at the end of the speech but I don't want feedback
  26. from the speakers. I don't think their opinion is very thoughtful and I
  27. don't think they have much useful to add to what I'm saying. [Laughter]
  29. So the question is "what rules should the society have for using
  30. software?". Now most of the time when people consider this question they
  31. work for software companies and they address it from a self serving
  32. point of view. they ask "what rules can we impose on every one else to
  33. make them pay or sell their money?". Now I'm sure you are familiar with
  34. the answers that they come up with.
  36. Now, I had the good fortune in the 1970s to be part of a community of
  37. programmers who shared software. and because of that I was led to
  38. address that question from a different direction to ask "What rules make
  39. for a good society for the people who use software?" and so I reached
  40. completely different answers.  let me tell you a little bit about what
  41. life in that community was like.  The community included programmers at
  42. some of the best universities. even programmers of computer companies
  43. sometimes participated. and in this community if you wrote a program,
  44. you shared it. that was our way of life. nobody forced this. nobody
  45. demanded this. but it was our way of life and so everybody did it.
  47. The lab where I worked, the AI lab at the other MIT, was perhaps the...
  48. in a way the deepest part of this community because there, all the
  49. software we used was the community software. it was all free software.
  50. we had an entire Operating System. the Incompatible Time Sharing system
  51. or ITS for short was developed by the community and mostly by us and
  52. therefore all the software that we used, we could and would share with
  53. anybody we wanted. so if you walked past another hackers console and you
  54. saw something interesting you'd say "hey what is that?" and he'd say "oh
  55. this is the new foobar program that we just got from stanford and its in
  56. the foobar directory." so you'd look in that directory and you'd find
  57. the executable you could run with also the source code. which you could
  58. study to learn how they solved those problems. and if you ran the
  59. program you might...[interrupts]
  61. To the MIC-TESTER: can you do anything about this feedback. there's so
  62. much. there's got to be a way to solve this problem. maybe if you turn
  63. of those microphones?. he should be able to turn them off... ok we'll
  64. see if this helps.
  66. in running the program you might encounter bugs or you might have
  67. ideas for new features. so you could go to the source code and fix the
  68. bugs and add more features. you could even cut out a piece of that
  69. program and put it into someother program that you were writing.  we
  70. used to call this "cannibalising" the old program which was a joke
  71. because it doesn't destroy the old program when you do this. so you
  72. could use the program not just by running it but in all the various ways
  73. it could be useful. The software we developed was available to everyone.
  74. It was part of human knowledge and because of that I could feel that I
  75. was on humanity's team. I was not working against other people, trying
  76. to beat them or stop them. I was working for the good of everyone. and
  77. that enabled me to feel good about my work.
  79. and so.. over the years the system grew the way a city grows. you know,
  80. you see somelines of code and say "oh by their style I can tell these
  81. were written in the 1960s". well, in other areas you see whole
  82. neighbourhoods that've been built recently and a program would be passed
  83. from one person to another to another they would keep on improving it
  84. over the years.
  86. but then we got a taste of  what life was like for most computer users.
  87. the people who did not belong to a community like ours. that happens
  88. when XEROX gave MIT a laser printer. now this was a very handsome gift.
  89. it was the first time anybody outside XEROX had a laser printer. it was
  90. actually a high speed copier that had been turned into a laser printer
  91. by adding a laser attachment. now this was a very fast printer and it
  92. printed a page a second and it had high resolution and straight lines
  93. came out nice and straight. but it had a flaw. it frequently got a paper
  94. jam. now then if it was a copier maybe that was okay cause there would
  95. have been somebody to fix it when it jammed. so it wouldn't have been
  96. jammed for long. but as a printer it was off by itself and often no body
  97. would pass by and fix it for a long time. so it would stay jammed for
  98. maybe an hour. it was a real problem. now when we discovered, when we
  99. recognised that this problem existed we knew a solution. because our
  100. previous printer which was slow and low resolution and tended to make
  101. straight lines come out crooked also got paper jams. and since we
  102. couldn't improve the printer itself, we being programmers not printer
  103. enggrs, we added features to the software to compensate for the
  104. problems. for instance there was a feature that everytime a file
  105. finished printing the system would display a message on that user's
  106. screen saying "your file foo has been printed". so you had to wait
  107. because the printer was slow but you didn't have to wait extra just
  108. because you didn't know that your job was finished. and there was the
  109. second feature that I recall adding which was: anytime the printer got
  110. in trouble the system would search the print queue and make a list of
  111. people waiting for printing and it would display a message to each one
  112. of them saying "The printer is in trouble. go fix it." now if you got
  113. that message you were not going to ignore it because you would know that
  114. only a few people are going to get that message and you didn't to want
  115. to take the risk that the printer would stay jammed. so you would go
  116. straight to the printer. the printer is still jammed. but a minute later
  117. 2 or 3 people would arrive. one of them atleast would know how to fix
  118. the problem and would teach the others. so essentially the system became
  119. self-correcting. we treated the user as a part of the system. and we
  120. added end-to-end feedback and we obtained reliable operation for the
  121. entire system even though the printer components were still unreliable.
  122. after all thats what feedback is for.  
  125. so we solved the problem and when we saw that the new printer had a
  126. similar problem we thought of using a similar solution. but there we ran
  127. into a stone wall. because we were able to add these features to the old
  128. printer because the old printer was controlled by a free program. we had
  129. the source code. we could make any changes at all limited by our skill
  130. as programmers. but the new printer was controlled by a proprietary
  131. xerox program. we couldn't add any features. we were stuck completely.
  132. we were prisoners of our software. so we just had to suffer with it. so
  133. you'd type the command to print a file and you'd go back to work cause
  134. you know its going to take a long time. a while later you'd notice the
  135. time oh its been half-an-hour... well, I don't desperately need it yet
  136. and its probably not printed yet so I'd go back to work. a while later
  137. you'd notice the time. oh its been a whole hour. Maybe its printed now.
  138. so you walk upstairs. you go to the printer and see its been jammed the
  139. whole time.  so at that point you fix the jam and you go back to work.
  140. and a while later you'd notice the time oh its been half-an-hour and now
  141. I really need the print-out. I'd better go and see. so you go upstairs
  142. to the printer and see it printed 200 pages of other people's stuff
  143. which was about 3 minutes of printing for this fast printer and then it
  144. jammed again. and at that point you'd say "I'm going to stand here and
  145. fix it everytime it jams and so I'd get my output". constant
  146. frustration.
  149. but what made it even more boring was to realise that we could have
  150. fixed the problem except that XEROX was not letting us fix the problem
  151. because they wouldn't let us have the source code. then I heard that
  152. somebody at Carnegie Mellon had a copy of that source code.  eventually
  153. I was visiting Carnegie Mellon for someother reason. so I went to his
  154. office and said "Hi, I'm from MIT. could I have a copy of the printer's
  155. source code". and he said "No. I promised not to give you a copy".
  156. [Laughter]. I was so stunned as well as angry that I couldn't think of a
  157. way to express it and do justice to my anger. all I could think of was
  158. to walk out of his office without another word.
  161. but I thought of that afterwards. you see... his refusal to help us.
  162. essentially his denial of co-operation with his colleagues was very bad
  163. for us at the AI lab at the other MIT. because we never got that source
  164. code, we were never able to solve this problem and the printer just was
  165. frustrating to use for several more years until we replaced it.
  167. but it was very good for me in a paradoxical way because it taught me an
  168. important lesson. A lesson which is important because most programmers
  169. fail to learn it. you see.. he had promised to refuse to co-operate with
  170. us, his colleagues at MIT but he didn't just refuse to co-operate with
  171. us.  chances are he refused he refused to co-operate with you too. and
  172. chances are he did the same thing to you as well. and I'd expect he also
  173. refused to cooperate with you.  infact he prolly refused to co-operate
  174. with most of you here today.  the exceptions being some of you who
  175. weren't born yet. because that was in 1980 or so. because he had
  176. promised to refuse to co-operate with just about everybody alive on
  177. earth at that time, he had signed a non-disclosure agreement. now this
  178. was my first direct encounter with a non-disclosure agreement. I was the
  179. victim. I and my Whole lab were the victims. and the lesson I learnt was
  180. that non-disclosure agreements have victims. they are not innocent. they
  181. are not harmless. they are hurting somebody.
  183. Now I was lucky to learn this lesson. most programmers first encounter a
  184. non-disclosure agreement when they are invited to sign one.  and there's
  185. always some kind of goodies, some temptation ... something they are
  186. going to get when they sign. so they make up excuses to ignore the
  187. ethical issue of what they are doing. they say "He'll never get a copy
  188. anyway so who shouldn't I join a conspiracy to deprive him". they say
  189. "this is the way its always done. who am I to question it?" they say "If
  190. I don't sign this. somebody else will". various excuses to gag their
  191. consciences. but when somebody invited me to sign a non-disclosure
  192. agreement my conscience was already sensitised. because it couldn't
  193. forget how angry I was when somebody had refused to share with _me_ the
  194. source code the source code that my lab needed. and I couldn't turn
  195. around and do the same thing to somebody else who didn't deserve it
  196. anymore than we did. so I said. thank you very much for offerring me
  197. this nice piece of software. But I cannot accept it in good conscience
  198. on the conditions that you have set. so I'm going to do without it. no
  199. thank you. and so I have never knowingly signed a non-disclosure
  200. agreement for generally useful technical information such as software.
  202. Now there are other kinds of information which raise different ethical
  203. issues. for instance there is personal information. a totally different
  204. subject. you know if you wanted to talk with me about what was
  205. happenning between you and your girlfriend and you ask me would I please
  206. not tell it to anybody That I could agree to. because that is not
  207. generally useful technical information. atleast it probably isn't. now I
  208. could imagine that you might reveal to me some wonderful new sex
  209. technique  [Laughter]  and then I might feel a moral duty to pass it on
  210. to the rest of humanity so that somebody could make use of it. but if
  211. you just wanted to talk with me about the usual soap opera stuff.. you
  212. know... who hurt, who's feeling how and how the the other one responded
  213. and who's angry and things like that those details that your life are
  214. those are not something that other people need to know to inorder to see
  215. how to live their lives better. so its okay for me to keep those secret
  216. for you.
  218. but when it comes to generally useful technical information. the stuff
  219. of science and engineering. the mission of these fields is to develop
  220. that information for humanity. if we conceal it we are betraying that
  221. information of our field. this after a few years of thought I came to
  222. the conclusion that this was wrong and that I decided that I would not
  223. do it. but during the same period of time a serious of calamities fell
  224. on my community and ultimately wiped it out. my community was destroyed.
  225. perhaps the final blow was when digital discontinued the PDP-10
  226. computer. because the entire time sharing system was written in
  227. assembler language for the PDP-10. so when the PDP-10 was discontinued
  228. our 15 years of work turned into dust and blew away. now thats a pretty
  229. bad blow in itself but the consequences were even worse. because the
  230. only operating systems of any kind from modern computers were
  231. proprietary. to get a copy you had to sign a non-disclosure agreement.
  232. so the only way you could get a modern computer and use it  was to
  233. betray every one else in the field. to do exactly what I had concluded
  234. people shouldn't do. so that put me in a moral dilema. I could not go on
  235. working on my field the way I had been doing it before because that path
  236. had been blocked off. it was no longer being available. it depended on
  237. being a part of a community that had its own body of software that it
  238. could share.
  240. so what was I going to do? the most obvious option was to accept that
  241. the world had changed. To adapt myself to it. To start signing
  242. non-disclosure agreement and start using these non-free operating
  243. systems. and I'm sure MIT would have had me developing non-free
  244. softwares as well. I thought about that and I realised in that way I
  245. could have fun programming and I could make money but at the end and I
  246. had to look back at my career and say "I've spent my life building life
  247. to divide people". and I would have been ashamed of everything I had
  248. achieved. So I looked for another option. and it was easy to find one. I
  249. could leave the software field and do something else. to many
  250. programmers that seems to be unthinkable. they say "THe people who hire
  251. programmers demand this, this and this. If I don't do it I'll starve."
  252. thats literally the word they use. "starve". Well I had no other special
  253. skills. but I'm sure I could have become a waiter. not at a fancy
  254. restaurant but I could have been a waiter somewhere. There are two
  255. things to know about being a waiter. one is as a waiter you are not
  256. doing anything wrong. there's nothing evil about being a waiter. well at
  257. most restaurants.  [Laughter]  and the second thing to know is as a
  258. waiter you are not going to starve. but I realised that for me being a
  259. waiter would be no fun. and it would be wasting my skills as an
  260. operating system developer. It avoid misusing the skills. Developing a
  261. non-free software would be misusing my skills. its better to waste them
  262. than to misuse them. but still its not the best thing. so I decided to
  263. look around for someother option. what could an Operating System
  264. developer do that would be ethical?  that would make the world a better
  265. place? And I realised that an Operating System developer was what we
  266. exactly needed.
  269. The moral dilema existed for me and any other computer user because all
  270. the other existing operating systems were proprietary.  if an Operating
  271. System developer were to write another Operating System, this is your
  272. free to share. this would give everybody a way out of the moral dilema.
  273. I concluded that I had been elected by circumstances to do this job. the
  274. job had to  be done. I knew that. nobody else was paying attention.
  275. nobody was going to do the job if I did not. and I had the skills
  276. necessary to do the job. so I realised I had to do it. I decided I would
  277. develop a free Operating System or die trying. presumably of old age. so
  278. this led to a bunch of technical design decisions. what kind of system
  279. should it be? well I had seen one entire Operating System turn into dust
  280. and blow away because it was written for a particular kind of computer
  281. that got discontinued. I didn't know what kind of computers would be
  282. popular in 5 years or 10 years. I knew it would take years to get this
  283. job done. and I didn't want to take the risk that the same thing would
  284. happen again.  clearly the system had to be portable. well, I knew of
  285. portable Operating System that was a success and that was UNIX. so I
  286. decided to follow the design of UNIX. that way there would be a good
  287. chance. I could write a system that could work and be portable.
  290. and further more I decided to make the system upward compatible with
  291. unix. why? because users don't like incompatible changes. I knew that if
  292. I took all the best ideas seen in various system and added my own
  293. favourite ideas I could have developed my dream operating system. but
  294. that would have been incompatible with other systems. I knew what users
  295. would say. They would have said "well this is very nice but it would be
  296. too much work to switch over and we can't afford. so we are not going to
  297. use your system." now at that point I could have made an excuse and I
  298. could have said "Well I offerred them freedom and they didn't take it so
  299. its their fault". That would have been sufficient as an excuse. But I
  300. wanted more than an excuse. I wanted to start a community where people
  301. would actually come and enjoy the benefits of living in liberty and
  302. having a community. so to do that I had to make a system that people
  303. would decide to use. Compatibility with a popular system was a good way
  304. to make it easy for people to switch to this system once it was done.
  307. Now UNIX consists of many components which communicate through
  308. documented interfaces or more or less document. so to be compatible with
  309. UNIX you had to replace each component one by one.  which means that the
  310. initial design decisions were all made. except for one: What range of
  311. target machines would we aim for?. Now UNIX was designed to run on 16
  312. bit machines. but that the small address space of those machines made it
  313. extra hardwork to get all the programs to run in that small address
  314. space. well I realised this was going to be a big job and we have to
  315. make it try to make it easier. One way to make it easier was not to
  316. support 16 bit machines. I figured that by the time this is done 32 bit
  317. machines would be the norm and so it would be okay if we didn't support
  318. 16 bit machines. and indeed thats what ultimately happened. by the time
  319. we had a GNU system that could run every body was getting 32 bit
  320. machines.
  322. so the design decisions were made so all that we needed was a name.
  323. well, we hackers, generally look for funny or mischievous names because
  324. think of people being amused by the name is half the fun of writing the
  325. program. so we also had a hacker tradition when you are writing a
  326. program thats similar to some existing program which is something we
  327. often had to do back before the days of portable programming. you know
  328. there was an existing program for some other computer system and you
  329. wanted something like it you had to write another one. so when you were
  330. doing that, writing a program similar to an existing one, you could give
  331. it a name that was a "recursive acronym". which said "This program is
  332. not that one". so there were many TECO text editors. and they were
  333. generally called something or the other TECO. but one hacker called his
  334. program TINT for TINT Is Not TECO. The first recursive acronym. well, we
  335. thought that was so much fun, we started making more. In 1975 I
  336. developed the first EMACS text editor.  The programmable, extensible
  337. display editor. and there were many imitations of EMACS and some were
  338. called something or the other EMACS.  but one was called FINE. for FINE
  339. Is Not EMACS. and there was SINE for SINE Is Not EMACS. and EINE for
  340. EINE Is Not EMACS. and MINCE for Mince Is Not Complete Emacs. and then
  341. EINE was almost completely rewritten by not quite and the new version
  342. was called SWEI for Swie Was EINE Initially.  [Laughter].
  344. So I decided to look for a recursive for 'Something Is Not Unix'.because
  345. I didn't have any cleverer idea. so I tried the obvious four letter
  346. approach and I discovered that none of them was a word. They didn't seem
  347. funny. so I tried a contraction so I could make a three letter acronym.
  348. I started substituting letters. ANU, BNU, CNU, DNU, ENU, FNU, GNU!. well
  349. GNU is the funniest word in the english language... so that had to be
  350. it. Now why is the word GNU used for so many jokes? The Reason is
  351. according to the dictionary the 'G' is silent.  so its sounds like
  352. 'New'.  so infact when people were asking the question "whats GNU?" long
  353. before there was a GNU system. But now it has a new answer when someone
  354. asks you "Whats GNU?" you can answer 'GNUs Not Unix'. And look at this
  355. you see it sounds like you are being obnoxious telling the person what
  356. it is not instead of what it is. But infact you are giving the one and
  357. only correct answer  [Laughter] .  Anyway when its the name of our
  358. Operating System please pronounce a hard 'G' pronounce it 'gah-nu'. If
  359. you talk operating the 'new' operating system you would get people very
  360. confused.  You see we've working on it for 18 years now so its not new
  361. anymore!  [Laughter] but its still is and always will be 'Gah-NU' no
  362. matter how many call it Linux by mistake.
  365. So we had a name. we could start work. In january 1984 I quit my job at
  366. the other MIT. and started working on pieces of GNU. Now I had to quit
  367. my job because if I'd kept work at MIT, the MIT administration could
  368. have said they owned everything I wrote and I would have had to beg and
  369. plead with them about  precisely how to release the software. I wouldn't
  370. want that to happen. I didn't want to take any risk that my software
  371. would not be free. so I took them out of the equation by quitting my
  372. job. and I've never had a job since then. but the head of the AI lab was
  373. nice enough to let me keep using the facilities.
  375. so I began using the one and only Unix machine at the AI lab to start
  376. developing GNU. now at the time I thought we would develop all of these
  377. pieces and only then would people start to use it. Thats not how it
  378. happenned. In september 1984, I started working on GNU Emacs. which was
  379. my second implementation of the programmable text editor. by early 85 it
  380. working well enough that I could use it for all my editing and that was
  381. very convenient you see I had no intentions of learning to use 'vi'. so
  382. until that point I did my editing on other computers and transferred the
  383. files through the network to the UNIX machines to test them. once GNU
  384. Emacs was running I could actually do my editing on the Unix machine.
  385. and so infact had been other people. other people who had been emacs
  386. users wanted to have an emacs to run on their unix machines started
  387. asking me for copies. so I had to work out the full details of how to do
  388. distribution.
  390. well, Of Course I put a copy in the FTP server directory and that way
  391. people on the net could get copies but in 1985 most programmers were not
  392. on the Internet. so they were asking me how could they get copies. I
  393. could have said "I want to spend my time writing more pieces of GNU. Not
  394. writing Mag Tapes. so please find a friend who is on the Internet who
  395. will download it and put it on tape for you." And they would have found
  396. somebody sooner or later. Every programmer knows every other
  397. programmers. but I had no job. and I was looking for someway I could
  398. make money through my work on Free softwares. So I announced "Send me a
  399. 150 Dollars. and I'll mail you a copy of GNU Emacs". And the orders
  400. began dribbling in. by the middle of the year they were trickling in. I
  401. was beginning to get 8 to 10 orders a month. which if necessary I could
  402. have lived on. So I had a free software business that was successful
  403. enough for me.
  405. Now part of the reason I could have lived on that is that I have always
  406. made a practice of living cheaply. Most americans if they start making
  407. 'this' much money they immediately look for how they can spend 'this'
  408. much money. [Laughter]  so they start buying houses and cars and boats
  409. and airplanes and rare stamps and art work and adventure travel and...
  410. children...  [Laughter]  all sorts of expensive luxury and of course
  411. once they get them they dont think they can live without them anymore.
  412. so the result is they become puppets of money. whoever has the money
  413. they have to obey.  They've lost the freedom in their lives. but if you
  414. resist getting accustommed to these expensive habbits then you can
  415. decide what you want to do with your life. you can do what you think is
  416. important. you can make a contribution to the world. instead of just
  417. struggling all the time for money.
  419. But people sometimes used to ask me before I started forestalling them:
  420. "What do you mean its free software if it costs 150 dollars".  Well, the
  421. english word free has multiple meanings. One meaning refers to price and
  422. another meaning refers to "freedom". When I speak of Free Software I'm
  423. referring to freedom not price. so think of free speech, not free beer.
  424. [Laughter] . when you start talking about free software to people it
  425. doesn't hurt the first time to explain that I mean 'free as in freedom'.
  426. to help prevent any confusion. so some people got their copies of GNU
  427. emacs and they didn't pay me because there was nothing on the FTP server
  428. to collect any money from anyone anybody could download it. some got
  429. their copies on a tape from me and they paid me. and some got their
  430. copies indirectly from somebody else who had a copy and maybe they paid
  431. that somebody else I dont know but they didn't pay me. whether they paid
  432. the somebody else that was between them and it was not in my business.
  433. so GNU Emacs was gratis for some users and paid for for some users.  but
  434. for all of the users it was free as in freedom.
  437. because all of them had certain crucial freedoms which makes the
  438. definition of free software. so let me now infact get to the hard issue
  439. and give you the detailed definition of Free software. because its after
  440. all its easy to say I believe in freedom but you are not addressing the
  441. hard issue that way. the hard issue is which are the freedoms that are
  442. important - the freedoms that we should safeguard and which are the
  443. secondary freedoms which have to give way when they conflict with the
  444. primary ones. because different ideas of freedom _can_ conflict. you
  445. know your freedom to swing your fist ends ends where my nose begins.
  446. because thats the matter of which freedoms are primary and which are
  447. secondary. so the definition of free software represents a conclusion
  448. about which freedoms are primary. let me give it to you now. a program
  449. is free software for you a particular user if you have all of the
  450. following freedoms.
  452.   Freedom#0 is the freedom to run the program for any purpose in any
  453.   way.
  455.   Freedom#1 is the freedom to help yourself by studying the code to see
  456.   what it does and then changing to suit your needs as you wish.
  458.   Freedom#2 is the freedom to help your neighbour by distributing copies
  459.   to others.
  461.   Freedom#3 is the freedom to help build your community by publishing
  462.   improved versions so others can get the benefits of your improvements.
  464. If you have all of these freedoms the program  is free software for you.
  465. Now freedom#0 doesn't require much comment. Its pretty clear that if you
  466. are not even allowed to run the program anyway you like thats a rather
  467. restriced program. even most most software will let you run it anyway
  468. you like, even though its restriced in other ways. and also the way the
  469. law is setup, if you have freedoms 1, 2 and 3 then freedom 0 follows as
  470. a consequence. so the freedoms that really distinguish Free Softwares
  471. from typical softwares are Freedoms 1, 2 and 3. so I'll go into more
  472. depth explaining why those freedoms are more important and what they
  473. need:
  475. Freedom#1 is the freedom to help yourself by studying the code to see
  476. what the program really does and then changing it if you like to suit
  477. your needs. to make this freedom feasible you have to be able to get the
  478. source code. yes its true its possible to study the binary by
  479. disassembling but thats terribly hard and people only do it as a last
  480. resort of desperation. so For this freedom to be really be meaningful
  481. you must have access to the source code. so access to the source code is
  482. a free condition of FreeSoftware. Now who can make use of this freedom?
  484. Well first of all, what changes do I need? well you could fix bugs, you
  485. could add new features. you could translate all the error messages and
  486. output into tamil. you could port it to a different computer system.
  487. anychange you want to make, you should be free to make. who can take
  488. advantage of this freedom. clearly any skilled programmer can make use
  489. of this freedom. But not only they. Any business that uses software can
  490. directly take advantage of this freedom. Now maybe there are no
  491. programmers in the company because what they do is make clothing. That
  492. doesn't matter if they want the program changed they can go to a
  493. programming company and say "How much will you charge for these changes
  494. and when can I have it done?" and if they don't like the answer they get
  495. over there they can go ask another company and say "When can you have it
  496. done?". because one of the consequences of FreeSoftware is that there is
  497. a free market for all kinds of supports and services. and the result is
  498. you can expect better support and service for free software. for a
  499. proprietary program, support is a monopoly. because only the company
  500. that owns the program in general can give you any support. except for
  501. the most superficial kinds. so the result they don't have to care and
  502. they know it. they'll tell you "Pay us and we'll let you report a bug".
  503. and if you do that they'll tell you "in six months there will be an
  504. upgrade. Buy the upgrade and you'll see if we've fixed this bug and
  505. you'll see what new bugs we gave you"  [Laughter] .
  507. so the support for proprietary software is typically lousy and its
  508. interesting to note that even if there's a choice of several different
  509. proprietary programs to do the job, once you've chosen a program the
  510. support for that program is always a monopoly. so you are choosing
  511. between several monopolies. well, with Free Softwares you'll get a free
  512. market for support. Of Course, in general you have to pay for it.  Free
  513. software doesn't mean zero price.  thats not the issue at all.  were not
  514. to eliminate paying for things and we think its fine when programmers
  515. get paid to provide support for programs. infact thats the kind of free
  516. software business that I did for the second half of 80s. But the
  517. important thing is that every body has got the freedom the use the
  518. program, to get support from wherever they like, to offer support when
  519. they wish. people can also benefit from this freedom if they value
  520. security and privacy on their computer systems.  because when you have
  521. the freedom to check what the program does you can see if it has a
  522. Trojan Horse. you can see if it has a surveilance feature, now if you
  523. don't have the time to check every program you use, but there's a
  524. community of users and people are checking various parts at various
  525. times. the result is if there's a malicious feature, it might get
  526. caught. and if there are accidental bugs because most programmers wont
  527. put in malicious features but we all make mistakes.  so bugs are always
  528. to be expected. If there's a bug, people can catch that too and fix
  529. the result is that you can trust the software better because it is
  530. not blind trust.
  532. with a proprietary program all you can do is put blind faith in the
  533. developer and often they don't deserve it. Microsoft put a surveilance
  534. in some version of Windows.  It would report what was on your harddisk.
  535. and I think people got very angry and they took it out. and I heard
  536. there are other proprietary programs that are popular which has
  537. surveilance features in them too.  There's also suspicion that there's a
  538. back door in windows because there are symbols called NSAKEY1 and
  539. NSAKEY2. People suspect that, maybe, these have to do with a backdoor
  540. that was provided for the NSA. No one knows and there's no way to find
  541. out either. and finally any intelligent person can take advantage of
  542. this freedom.
  544. now most people are not going to learn to be skilled programmers but
  545. anybody can learn a little programming which is enough to make simple
  546. changes.  and thats useful by itself. and if you are the kind of person
  547. whose strength is getting along with people you are not a technical
  548. person well then you probably have a lot of friends.  and some of those
  549. friends are programmers so when you want a change made in your program
  550. you can convince one of your friends to do it. so everybody can take
  551. advantage of the freedom to change programs. Now if you don't have this
  552. freedom that causes practical material harm to the society. because
  553. people are stuck using software that doesn't do what they want and maybe
  554. even snoops on them. and they cant fix it. They are prisoners of their
  555. software. but it also causes Psycho-Social harm. that affects people's
  556. morale, their enthusiasm for their work. you see, if you have to use a
  557. program thats painful to use its not good.  and you are not allowed to
  558. improve it, its going to be frustrating.  its going to be frustrating
  559. over and over. well people who've experience this repeated frustration
  560. they tend to learn to stop caring thats the way you can protect yourself
  561. from feeling frustrated. If you don't care whether you don't get any
  562. work done or not then you are not going to get frustrated when you can't
  563. get any work done. but when ...
  565. [audio tape flipped over.. a few minutes of speech is lost]
  568. for beings that can think and learn sharing useful knowledge is the
  569. fundamental act of friendship, when these beings use computers these
  570. acts take the form of sharing software. if you don't have this freedom,
  571. if a program has a owner and this owner and this owner by whatever
  572. method has setup a situation that every user has to pay to use the
  573. program
  575. [someone leaves]
  577. RMS: leaving so soon? I hope it wasn't something I said
  579. Audience: [Laughter]
  581. well, if the program has an owner who has established a situation where
  582. every user must pay to use the program then this creates a financial
  583. disincentive, discouraging the user of the program. because some users
  584. will say "Alright, I'll pay" and they will use the program and the
  585. others will say, "Its too much I'll never mind. I'll do without it". and
  586. everytime somebody says "Never Mind, I'll do without it", the program is
  587. going partly to waste. but the work it takes to write the program to any
  588. given level of power and quality is the same regardless of the no. of
  589. users. Infact it might be even harder if you have fewer users helping
  590. you by reporting bugs. so, the same work is done. but only a part of the
  591. potential benefit is achieved. The rest is deliberately inflicted waste
  592. which is practical material harm to the society. but because its
  593. inflicted by forbidding people to help each other it causes a psycho
  594. social harm which affects the spirit of co-operation, the spirit of
  595. good-will, benevolence, the willingness to help other people, just
  596. because you see that they could use your help. This spirit of good will
  597. is society's most important resource. we depend on this so that we can
  598. have a liveable society instead of a doggy-dog jungle. and because this
  599. resource is so important the world's major religions all talk about the
  600. importance of this spirit of good will of helping other people. for
  601. thousands of years moral leaders have encouraged the spirit of
  602. benevolence. so what does it mean when we see major social institutions
  603. telling you that you are not supposed to help your neighbour. they are
  604. polluting the society's most important resource which is something that
  605. society cannot afford. what does it mean when they say that if you share
  606. knowledge with your neighbour you are a "pirate". They are saying that
  607. helping your neighbour is the moral equivalent of attacking a ship. if
  608. you don't believe that, reject that word. its a propaganda word. and
  609. what does it mean when they start making harsh punishments for anyone
  610. who shares with his neighbour how much fear is it going to take to get
  611. people to stop helping their neighbours. do you want your country to be
  612. pervaded by that level of fear? I certainly don't. I hope you dont
  613. either. this i think is the most important reason why software should be
  614. free. because we must encourage people to help their neighbours, not
  615. discourage them. When I was a child going to school, the teachers were
  616. trying to teach us to share. they said, If you bring candy to school you
  617. can't keep it all for yourself, you have to share with the other kids.
  618. they were trying to teach us to share with other people. now the US
  619. today is the world leader in trying to stop people from sharing useful
  620. information with their neighbours. [Applause]. but the US is not the
  621. first country to make an effort to stop sharing. The soviet union did
  622. that too.  Trying to crush the forbidden copying and sharing which was
  623. known as "Samisda"[?] You know dissidence did the same. you get a copy
  624. you put six carbons in your typewriter and you a copy and then you hand
  625. out those six copies to other people and they would type up more copies.
  626. The soviet union despite all its vicious repression was never able to
  627. completely stop the undergroud sharing that their people did. But they
  628. tried hard and they used several different methods. First guards
  629. checking all copying equipment, thats why people had to it with carbon
  630. papers with typewriters because for every piece of copying equipment
  631. there was a guard check what you copied. Second : harsh punishments.
  632. Those who were caught doing forbidden copying could be sent to siberia
  633. for years, were put in prison. Third: To help catch people they ask for
  634. informers, They ask everyone to wrap on their co-worker's and their
  635. neighbours to the information police. Fourth to help catch people
  636. collective responsibility. "You. You are going to watch  that group. If
  637. I catch any of them doing forbidden copying, you are going to prison. so
  638. watch them carefully." and fifth, Propaganda starting in Childhood: To
  639. teach everyone that only a vicious enemy of the people would do this
  640. forbidden copying. The US today is using all five of these methods to
  641. crush forbidden copying and sharing. First: Guards checking copying
  642. equipment. Well in copy store they have guards checking what you copy,
  643. to make sure you don't do forbidden copying.  But to have human labour
  644. checking what you are copying would be too expensive in the US. I guess
  645. they didn't think of hiring people from India to do it.  [Laughter] . so
  646. they are using robot guards. Programs that go in your computer and are
  647. designed to check what you are going to copy and stop you. Its a crime
  648. to bypass these robot guards. Second, Harsh Punisments: Well, ten years
  649. ago if you made copies of something and handed out them to your friends
  650. just to be nice, that was not a crime in the US. it had never been a
  651. crime. and then they made it a felony. You could be put in prison for
  652. years for helping your neighbours. In britain now they are proposing a
  653. '10 year' punishment. 10 year prison for sharing with your neighbours.
  654. That shows how far the repression can go when you try to stop people
  655. from sharing with their neighbours. When you try to prohibit their
  656. natural tendency.  And Third: Asking for Informers, In the US, there
  657. have been Ads on Television there were ads in the subways in Boston
  658. asking people to ???  on their co-workers to the information police.
  659. which there is called the "Business Software alliance". A Terror
  660. Organisation. and I say that after careful thought. in Argentina the
  661. business software organisation sent people letters threatening them with
  662. being raped in prison, if they shared with their neighbours.  Fourth -
  663. Collective Responsibilty: In the US this was done by Constricting
  664. Internet Service Providers to keep track of, they've been made legally
  665. responsible for everything their customers post.  and the onlyway they
  666. can escape being punished is if they have an automatic procedure of
  667. taking down anything within two weeks of a complaint. so nowadays if
  668. somebody accuses, you don't even get your day in court. your psych just
  669. gets unplugged. and thats it.
  671. Fifth - propaganda starting in childhood: Thats what the word Pirate is
  672. for. When I went to school, the teachers tried to teach us the habbit of
  673. sharing. Today according to the US govt, teachers are supposed to teach,
  674. quote, Say Yes to Licensing, Unquote. So instead of saying "Oh you
  675. brought candy to school, well you have to share it with the other kids".
  676. they say, "Oh you brought software to school, well, don't share it. oh
  677. no! sharing is wrong. sharing means you are a pirate." the US laws don't
  678. apply in other countries. But the US is trying to push the same kind of
  679. laws in every other country. I hope you will spread the word that the
  680. India should _not_ adopt a law like the DMCA in the US. Thats a
  681. Tyrannical law. Its an Oppressive law and _you_ should save yourselves.
  682. Even if you can't save us. So that is freedom 2: The Freedom to help
  683. your neighbour by distributing copies of the program.
  685. Freedom3 - is the freedom to help build your community by publishing an
  686. improved version so others can benefit from your work. now people used
  687. to tell me if the software is free that means no body will get paid to
  688. work on it, so nobody will work on it. they were confused by the two
  689. meanings of the word 'free'. because they thought it meant gratis which
  690. is not the case but none the less that was their theory. Today we can
  691. compare that theory with observed fact. and we see that hundreds of
  692. people or maybe thousands are being paid to develop free software and
  693. tens of thousands are developing free softwares as volunteers. and in
  694. fact we are developing large amounts of free softwares. What could
  695. possibly motivate these people? well, I tell you I'll tell you some of
  696. the motivations that I've heard people tell me. One of them is political
  697. idealism. and the desire to contribute to a good decent society where
  698. people can help each other instead of to divide people and to keep them
  699. helpless. Thats the important motivation for me but not everybody in our
  700. community has that motivation. second - Another motivation is fun.
  701. programming is great fun. not for everybody. but for some people and
  702. especially for some of the best programmers, programming is fun. thats
  703. why so many people after they do their job, which is software
  704. development want to develop some free software in their spare time.
  705. because its especially fun, where you are your own master and no one can
  706. tell you what to do. another reason is to get appreciation. if you
  707. develop a free program that a hundred thousand people use you can feel
  708. really good. a lot of people will be appreciating you. Another is
  709. profession reputation. if a hundred thousand people are using your free
  710. program thats gonna impress anybody who might want to hire you another
  711. reason is gratitude. if you've been using the community's free software
  712. for years and appreciating how useful it is, then when you write a
  713. program, that could be an opportunity for you to contribute something
  714. back to that community to express your appreciation. and there may be
  715. other motivations that I haven't thought of and don't know about. Any
  716. given person might feel a combination of several other motivations.
  717. Because human nature and human motivation are complex and money can also
  718. be a part of the motivation for some people those who are getting paid.
  719. When I released GNU Emacs after I while I got a message saying "I think
  720. I found a bug and here's a fix". and then I got a msg saying " I thought
  721. this feature was missing so I wrote it and here it is" and then I got
  722. another bug fix and another new feature and another and another and
  723. another until they were pouring in on me so fast that just using all of
  724. this help was a big job. MicroSoft doesn't have this problem. [Laughter]
  725. [Applause]. You know I've never understood why that is so funny but
  726. people always do find it funny. So, after a while people began noting
  727. the phenomenon that when a free program becomes so popular you often get
  728. a community of developers helping to improve it and free software
  729. started to get a reputation for being powerful, reliable software from
  730. the reports of the people who used it. and this was both good and bad.
  731. it led a lot more people to start using the free software especially in
  732. the 1990s. but at the same time we got lots of people coming into our
  733. community purely because the software was practically advantageous. it
  734. was expedient.and they didn't appreciate the freedom. They didn't care.
  735. and when they talked about the software to other people they didn't even
  736. mention freedom as an advantage. they didn't say the fact that you are
  737. free to co-operate with other people is an advantage. they just
  738. mentioned that it was powerful, reliable software and you could get it
  739. cheap. so then we got millions of more people coming into our community
  740. who never even heard anyone say that there is an ethical issue here. and
  741. eventually they formed a different movement called the "open source
  742. movement". Now I've been telling you the phil of the free software
  743. movement and as you can hear we sight both practical benefits and
  744. ethical benefits of having these freedoms in your use of software. the
  745. open source movement has practices close to ours, not identical. But the
  746. big difference is that they only sight the practical benefits. They
  747. don't say that morally speaking this is the way it should be. They go to
  748. companies and say we think it would be advantageous for you if you do
  749. things this way. well that is useful, they've persuaded some companies
  750. to release important pieces of software as free software. so practically
  751. speaking they contribute to our community but at the same time in the
  752. fundamental questions of the community we disagree completely. what your
  753. views are thats for you to decide. You might agree with the FSF, you
  754. might agree with the open source movement. you might disagree with us
  755. both. its up to you. but I'd like to invite you to support the Free
  756. software movement. After you've had a chance to think about the issues.
  757. and if you _do_ support us, please wave our banner. our banner is the
  758. term "free software" if you support the movement free software, say so
  759. by using the term free software. Its one of the ways to keep us in the
  760. public awareness so that our views and principles will be visible to
  761. people and they will have the chance to think whether they agree or not.
  762. because the open source movement tends to get more support and
  763. businesses.  and those businesses tend to use their terminology and bcos
  764. they don't criticise the practice of proprietary software their views
  765. are easier. They are less challenging ethically. So they get a lot of
  766. supporters who are not prepared to consider the ethical issues that we
  767. in the free software movement raise. so the result is that their name is
  768. heard more and we often get forgotten. If we were to raise this ethical
  769. issues so that people can think about it we need to get heard and you
  770. can help us with that by raising our banner the term "Free software". so
  771. if you don't have this freedom, the Freedom 3, the freedom to publish an
  772. improved version. that causes practical material harm because this
  773. phenomenon of community improvement if we don't get powerful, reliable
  774. software. but it also causes psycho-social harm. which affects the
  775. spirit of scientific co-operation.  the idea that we have to work
  776. together to advance human knowledge if we are going to do it
  777. effectively. so that is freedom 3. the freedom to help build a community
  778. by publishing an improved version so that others can benefit from your
  779. work.
  781. If you have all of these freedoms and the program is free software for
  782. you now why do I formulate the definition in this complicated way? Why
  783. do I not just say the program is freesoftware if it comes with all these
  784. freedoms? The reason is that sometimes the same code can be free for
  785. some people and non-free for others. Now that might seem strange so let
  786. me give an example to explain how that happens. The biggest example of
  787. this that I know of is the X Window System which was developed at the
  788. MIT in, the Other MIT, in the 1980s and released as Free Software. so if
  789. you got their version you had all these freedoms. it was free software
  790. for you. but among those who got copies were various computer
  791. manufacturers who distributed Unix systems. So they took X Windows and
  792. they made the necessary changes to get X to run on their platform and
  793. then they compiled it. They made binaries and they put the binaries in
  794. their Unix systems and distributed just the binaries under the same
  795. non-disclosure agreement as all the rest of Unix. and then millions of
  796. users got copies of these binaries with no freedom at all. This created
  797. a paradoxical situation. If you asked "Is X Windows Free Software or
  798. Not?" the answer depended on where you made the measurement. If you made
  799. the measurement coming out of the developer's group, you'd say "Here I
  800. observe all these freedoms. Its a free Software" If you made the
  801. measurement among the users you'd say "umm...most of these users donot
  802. have these freedoms.  Its _not_ free software.". Well, the developers of
  803. X Windows did _not_ consider this a problem. because they were not
  804. aiming to give the users freedom. they were aiming to have a big
  805. professional success. It was a big success. It set the de-facto
  806. standard. but in the GNU project. Our goal was to give users freedom. To
  807. give you freedom. If the same thing that happened to X had happened to
  808. GNU, GNU would be a failure. so I looked for a way to stop that from
  809. happening and the method that I developed was called "copyleft". You can
  810. think of this a taking copyright and flipping it over. Opposite results.
  811. you see, copy left is based on copyright. we use copyright law in order
  812. to get these opposite results. bcos normally copyright is used to stop
  813. people from sharing. to deny them the freedom to share. whereas we when
  814. we use copyleft, we guarantee everybody the freedom to share in order to
  815. make sure that all of you get the freedom to share we have to make sure
  816. that those middle men cannot strip the freedom away. so here's how we do
  817. it. First we put on a Copyright notice which says "This program is
  818. copyrighted" and then and by default you are not allowed to change or
  819. share the program. But then we say you are explicitly authorised to
  820. modify this program. You are explicitly authorised to redistribute
  821. copies of this program. You are explicitly authorised to publish a
  822. modified or extended version. But there is a condition. This condition
  823. is the reason why we go to all this trouble. The condition says: "Any
  824. modified or extended version or any version of this program you
  825. distribute must as a whole carry with it the same freedom that you got
  826. from us". and so the result is that everywhere the code goes the freedom
  827. goes with it. Even if the program changes it still carries with it the
  828. freedom. so all the users get the freedom. The X Windows Problem did not
  829. happen for us. In effect these crucial freedoms that I explained to you
  830. becomes inalienable rights of users of our software. so copyleft is a
  831. general idea. You can't use copyleft. Like you can't use the concept of
  832. a text editor. you have that specific text editor, then you can run it.
  833. and likewise to use the idea of a copyleft you have to have a specific
  834. license that you use. The license we use for most GNU software is called
  835. the GNU General Public License or the GNU GPL for short. Note that the
  836. 'G' in 'GPL' stands for General. Not GNU.  we also have a couple of
  837. other copyleft licenses that are other more permissive in special
  838. situations. But mostly we use the GNU GPL. and infact about 2/3rd of all
  839. free softwares use the GNU GPL. We also have a kind of copyleft license
  840. for manuals and text books called the GNU Free Documentation License. So
  841. if are writing a text book on any subject at all, I hope you'll release
  842. it as Free Documentation under the GNU Free Documentation license. This
  843. license was designed to make it possible for commercial publishers to
  844. profitably publish free manuals and infact there are some six free books
  845. that have been published commercially under this license. Thats not
  846. counting the manuals that we publish in the free software foundation. so
  847. there is copylefted free software and there is non-copylefted free
  848. software. Both of them are free . The developers of the non copylefted
  849. free softwares like the X windows have respected your freedom. they are
  850. not trying to deny you any important freedom. The difference is with
  851. copyleft we go even further and we actively try and stop anybody from
  852. taking away your freedom. with a non-copylefted free software they are
  853. not actively defending you your freedom but they donot attack your
  854. freedom. so they are respecting your freedom. They are not doing
  855. anything wrong. they are just not doing as much right as they could do.
  856. its a very big difference. We don't say that they are doing wrong. but
  857. we just say that they could do better. so that non-copylefted
  858. freesoftware can be used in a free operating system like GNU and infact
  859. I did decide to use X Windows in the late 1980s. I felt we should have a
  860. window system in GNU but before we wrote one there was one X. it was
  861. becoming popular, it did the job, it was technically suitable. so I said
  862. alright we'll save trouble. We won't write a window system of our own.
  863. we use X. so we started the other making the other pieces of the GNU
  864. system work together with X. and we put into the coding standards that
  865. if your program is graphical it should work with X. Through out the
  866. 1980s our mission was to come up with all the pieces we needed to make a
  867. complete Operating System. now sometimes we were lucky and somebody else
  868. wrote the piece we needed or something that could do the job like X
  869. Windows. and when that happened we said good we don't have to write this
  870. piece. We'd just use the piece that we found. but that happenned or
  871. didn't happen by chance. bcos those other projects, they were not aiming
  872. to make a free Operating System. They had various different goals of
  873. their own. so it was just accident whether their software was useful for
  874. us and whether they made it free. well, we were looking for
  875. opportunities to save the effort by using some existing software that we
  876. could find. because the job of developing a whole unix like system was
  877. very big. many people said it was so big we'd never finish it. well, I
  878. thought we'd finish it but clearly we had to look for shortcuts. But
  879. when we didn't find this existing software to use then we had to develop
  880. those programs or recruit people to develop them. And thats what we did
  881. during the 1980s. In october 1985 we founded the Free software
  882. foundation. which is a charity whose purpose is to raise money to
  883. promote Free software and in the 80s a lots of its activity was hiring
  884. people to work on developing GNU. some important pieces of this system
  885. like the Shell and the C library were by the staff of the Free software
  886. foundation. but most of the work was done by volunteers. For instance
  887. I'm a full time volunteer for the free software foundation. bcoz the
  888. foundation doesn't pay me. now there's a reason for that when the
  889. foundation first had enough money to hire one person I as the president
  890. had to decide who to hire. and it was my responsibility to spend the
  891. money effectively. So I realised that paying st(.*)nous[?] salary would
  892. be like throwing the money away bcos we could get st\\1n to work for
  893. nothing. So I decided to hire someone else instead. and now a days
  894. though there are many full time volunteers for the GNU project, most of
  895. them are getting paid by somebody else. so for the FSF they are
  896. volunteers. There are also thousands of part time volunteers who are
  897. contributing their spare time and don't get paid for it but they still
  898. get a large amount of work done. In the 80s and in the early 90s it was
  899. possible to doubt whether free software could actually develop the full
  900. spectrum of software to fill the public's needs. but nowadays we are
  901. pretty close to doing the whole thing already. so there's no long any
  902. possible doubt that we can't do the job. the only question is whether we
  903. will be allowed to do the job. but thats getting ahead of things. by
  904. 1991 the job was almost done. we had almost all the pieces necessary but
  905. there was one major gap still. The Kernel. Now we started developing our
  906. kernel in 1990. Again we were looking for a way to save time by finding
  907. something we could start with that was already working and we found an
  908. already working micro kernel called 'MACH'. developed at Carnegie Mellon
  909. University. so mach did the lower level part of the Unix kernel's job.
  910. So we had to write the upper half and we were going to do that by
  911. writing a collection of servers that communicate through message
  912. passing. and this provides all sorts of technical advantages, greater
  913. power. And we thought that bcos these were effectively user programs it
  914. would be easier to debug them unfortunately thats not so. That wasn't
  915. so. The Debugging facilities were lousy and these programs were
  916. asynchronous so they can have timing errors that were not reproducible.
  917. it took many many years to get this collection of servers which is
  918. called the GNU/Hurd to run. we call it the GNU/Hurd btw, because GNUs in
  919. Africa live in Herds. So this is a herd of GNU servers or GNU/Hurd.
  920. Fortunately we didn't have to wait that long because in 91, 92 a finnish
  921. college student called Linus Torvalds wrote another free kernel, well he
  922. wrote a kernel and at the end he decided to make it free software and he
  923. released it under the name "Linux". He used to monolithic approach that
  924. had been used before. well we didn't know about linux. because he never
  925. contacted us to tell us about it. But he announced it on the network
  926. somewhere and people who knew about it said "Lets see if we can find all
  927. the other parts of an operating system so that we can make a complete
  928. system." So they looked around and lo and behold, everything they needed
  929. was already there. What good fortune, they said its already available.
  930. but there was no rock about it. What they had found were all the pieces
  931. that were going to be the pieces of GNU! so infact what they were doing
  932. was fitting linux into that gap in that GNU system to make the
  933. combination of GNU + Linux. The GNU/Linux system. But they didn't
  934. realise that. They didn't that they were finding all the pieces of the
  935. GNU system. Therefore they were starting with Linux and finding these
  936. other pieces and putting them on top of linux. so they call that a Linux
  937. system which they really shouldn't have done. They had no business
  938. calling this version of our operating system by someother name. but
  939. thats what they did. and the misnormer got immitated by other people.
  940. and thats how it happened. that we developed an Operating System thats
  941. used by some 20 million people and most of them don't know its our
  942. system. They think its Linux. They think it was all started in 1991 by
  943. Linus Torvalds. Well, the development of Linux, the kernel was a major
  944. contribution to our Community because that was the step that took us
  945. across the finish line.  because before that we had 95% of the system
  946. but it wasn't capable of running by itself. so you take parts of that
  947. system and install them on top of another operating system but you had
  948. to get another operating system to start with once that last gap was
  949. filled that made it an entire operating system. so you could put it on a
  950. bare PC. You didn't have to have some other operating system first. And
  951. infact that meant that the goal we had set out for in 1984 had been
  952. reached. It was a Free Operating System that you could run on a modern
  953. computer. So you could actually live in Freedom. you could install this
  954. free software you could refuse to use any proprietary software and then
  955. you would have freedom and you would have the freedom to form a
  956. community with other people the freedom  to cooperate with others. so
  957. you could live an upright life. A happy life as a computer user. but,
  958. the error of calling the entire system Linux, was a major blow to the
  959. Free software movement. bcos it broke the connection from our software
  960. to our philisophy. before that people who used these pieces of GNU on
  961. other system they knew they were using pieces of GNU. so they thought of
  962. themselves as GNU users. They became the fans of GNU. and so when they
  963. saw the things we had put in there to describe the philosophy they would
  964. think about it seriously at least sometimes they would. because they
  965. realised that this was the philosophy behind the software that they
  966. liked. so that was the reason to atleast give it a serious
  967. consideration. And that philosophy I had been telling you today. Well,
  968. after thinking about it, some of them would agree. And if they agreed
  969. they then feel a motivation to develop more software for GNU. so the
  970. software helped to spread the philosophy and the philosophy helped to
  971. extend the software. but when people started calling this entire GNU
  972. system Linux, this connection was broken and instead the software lead
  973. people to a different philosophy. The philosophy associated with the
  974. name "Linux". Which was the apolitical philosophy of Linus Torvalds. He
  975. didn't like considering these things as ethical or political issues. he
  976. just put that it in terms of whats practical. He didn't say that
  977. software should be free. He sometimes develops and uses non-free
  978. software. so the result was here was this system thats basically GNU and
  979. it was attracting people over to the other philosophy and not to ours
  980. anymore. it was a real problem. btw the other philosophy was the one
  981. that later on became the philosophy of the open source movement. so the
  982. result is that if you look around on our community, most of the users of
  983. this version of this GNU system have never even come across the
  984. philosophy that motivated us to do all this work. sometimes people say
  985. to me when they hear me making efforts to ask people "please call the
  986. system GNU/Linux", they say to me it looks bad to ask for credit. it
  987. looks like you're just being selfish. so it would be much wiser to let
  988. it drop and when people call the system linux, smile to yourself and
  989. take pride in the job well done for this would be a very wise advise
  990. except for one mistaken premise. the idea that the job is done. We have
  991. a lot of work to do. we've made a great beginning but we haven't
  992. finished the job. you see we've developed free systems that are used on
  993. some 20 million or so computers but thats a fraction of all the
  994. computers that there are maybe 5% I think I heard. and that means we
  995. have a lot more to go. we have a large range of free software now, but
  996. there's still other programs that users would like. we have to develop
  997. those. we have to make sure that you have free software for every job.
  998. and there are laws being passed in some countries that prohibit us from
  999. developing free software for certain jobs. we have to do what it takes
  1000. to overcome those obstacles, repeal those laws whatever is needed. Its
  1001. going to take determination. some of these jobs can be done just by
  1002. writing software and people will do that because its fun in many cases.
  1003. But overcoming these obstacles takes more than just fun, it takes the
  1004. kind of determination people show when they know that they are fighting
  1005. for their freedom and for their communities. So we have to teach people
  1006. that. but if you look around in your community today you'll see every
  1007. where you look most of the institutions in our community are calling
  1008. these systems linux and not presenting freedom as the goal and the
  1009. result is when I talk about the importance of freedom I get responses
  1010. like "This idealism is bad for the success of Linux". They say "what
  1011. does this got to do with me? I'm a linux user". People call themselves
  1012. Linux users that means people use the GNU/Linux system but when they
  1013. hear about the philosophy behind GNU they say "What does this have to do
  1014. with me?" because they don't see any connection between themselves and
  1015. GNU. If they knew the origin of the system that they're using they would
  1016. see the connection and the name GNU/Linux shows people that connection.
  1017. and then there's the others that say "This Idealism must be impractical
  1018. and its bad for the success of Linux". There's so many ironies in that.
  1019. One of them is What do they mean by success of Linux?  Well they are
  1020. really talking about the success of the GNU/Linux system, but what does
  1021. it mean for that system to succeed? They think it means just to have a
  1022. lot of people running it. and never mind how. never mind if these people
  1023. run it together with non-free software. because they are thinking of
  1024. success as nothing but popularity. which is irrational.  Its actually an
  1025. example of mental inertia. you see, they copy that idea of success from
  1026. the proprietary software world. In the proprietary software world
  1027. atleast its rational. Why do they want their software to be popular?
  1028. because every user is supposed to pay them! so more users means more
  1029. money. its an evil system but atleast they're being rational in that
  1030. part of it.  When you copy this definition of success over to free
  1031. software it doesn't make any sense at all! You know why does it matter
  1032. how many people use that free kernel linux? Its a kernel it does its
  1033. job, but is the number of users of that kernel really important? I don't
  1034. think so... and why does it matter how many users are using the
  1035. GNU/Linux system or any variant of the GNU/Linux system? Now I might
  1036. feel something about it because I launched it. but thats just ego.
  1037. Really is it important how many users this system has? I don't think its
  1038. of any real importance. The important thing is to spread freedom to as
  1039. many people as possible and ultimately to everyone. Thats the goal thats
  1040. worth striving for. We have to remember that goal. because the most
  1041. important thing for reaching a distant goal is to remember the goal. if
  1042. you forget where you've headed you're going to end-up somewhere else!
  1043. and is idealism really impractical? Not at all. If you are a distant
  1044. goal, there are only two ways to reach it. one of them is to have a lot
  1045. of money. and the other is idealism. because idealism will enable you to
  1046. keep on going until you get there, otherwise you'd just give it up. so
  1047. we don't have a lot of money. so we have to have this idealism instead.
  1048. there's nothing more practical like idealism. The GNU system and the
  1049. GNU/Linux variants are our idealism made real. But, very few people are
  1050. saying this to the users of GNU/Linux. Most of the institutions in our
  1051. community call this system as Linux and they don't say these things.
  1052. Consider for instance the companies that package this system, that
  1053. package this GNU/Linux system, well, they all put some non-free software
  1054. on the CDs, Yes, its true that you can get a free Operating System to
  1055. run on your PC, but its not easy to find one. If you go to the store and
  1056. buy something that says a version of the GNU/Linux system which says
  1057. "Linux" on it, it will always have non-free softwares as well. you have
  1058. to be an expert to know what to get rid of.  and then what about the
  1059. magazines dedicated to the use of the GNU/Linux system. Typically these
  1060. magazines call themselves Linux Something.....
  1063. So I have a different term, I call them freedom subtracted packages.
  1064. because if you have installed a free GNU/Linux system if you are
  1065. enjoying the benefits of freedom that we've worked so many years to give
  1066. you, those packages give you the opportunity to buckle on a chain. and
  1067. what about the trade shows about the GNU/Linux system, they typically
  1068. call themselves Linux-something or the other and they host companies
  1069. advertising non-free software so when in effect the trade show gives a
  1070. seal of approval to non-free software. and what about the user groups
  1071. for the use of the GNU/Linux system most of them call themselves linux
  1072. user groups and they typically invite salesmen to come in and present
  1073. non-free software in effect giving the group's seal of approval to the
  1074. use of the non-free software. Now I would like any linux user groups to
  1075. become GNU/Linux user groups and to take a stand on freedom. to stand up
  1076. for freedom and give GNU the credit for launching this system. Now I
  1077. believe that there's a Linux users group here in chennai and I hope that
  1078. you will become a GNU/Linux User group. Please tell me if you do. we
  1079. have a page where we list GNU users groups and GNU/Linux user groups. So
  1080. if you become a GNU/Linux User group, we will list you there. but in any
  1081. case in our community therefore most of the institutions are talking
  1082. about Linux and they are not standing up for the philosophy of the GNU.
  1083. so the only place you see this free software philosophy generally is in
  1084. association with the GNU name. this is why it makes a difference when
  1085. you call the system GNU/Linux because it shows people a connect between
  1086. them and their system and GNU and the Free software philosophy. it wil
  1087. help people lead people to this philosophy so that they can think about
  1088. it and maybe get motivated to get help work for freedom. and we're gonna
  1089. need a lot of the US today there are too different kinds of laws
  1090. prohibiting free software. one is the Digital Millenium Copyright Act,
  1091. which has been used to prohibit free software for jobs like playing a
  1092. DVD, listening to an audio stream, reading an e-book. Its a narrow range
  1093. of jobs but they are very important jobs. if users can't do that with
  1094. free software, free software is in danger. and the other problem the
  1095. even bigger problem comes when patents are applied to software ideas.
  1096. when software ideas can be patented and this is dangerous for free
  1097. software development because its dangerous for all software
  1098. developments. program tend to be complicated these days. one program
  1099. will have a hundred ideas in it. well, if ideas can be patented if
  1100. software ideas can be patented then any one of those hundred ideas might
  1101. be patented by somebody. so even then you get sued if you wrote the
  1102. program yourself. without software patents you can be confident. if you
  1103. wrote the software program yourself, if you didn't copy the code then
  1104. its legal and you're not gonna get sued for writing it. but with
  1105. software patents you are quite likely to get sued for writing it or
  1106. atleast get threatened. you can't do this. you can't write such a
  1107. program. and the worst thing is if there are a 100 ideas 5 or 10 of them
  1108. might be patented by 5 or 10 different companies. so you get one of them
  1109. and when you're finished with doing it one another comes along. so its
  1110. constant trouble for any software developer and even for software users.
  1111. so all the companies in india that use software have to recognise that
  1112. software patents are a danger. all the companies in India that develop
  1113. custom software, bcos its typically custom software thats developed by
  1114. indian companies they have to fear software patents because those custom
  1115. programs sometimes use many ideas too. and they could get sued for
  1116. writing those programs. so software patents are a danger. fortunately
  1117. the Indian govt seems to be aware of this. so there's a good chance that
  1118. India will resist the danger of software patents. but don't leave it to
  1119. chance. Make sure to spread the word, especially if you know companies
  1120. that develop software or companies that use computers and software. Make
  1121. sure they understand how dangerous software patents can be. now why do
  1122. we in the free software movement particularly pay attention to this
  1123. goal? to this issue? bcos we want to write a full range of software to
  1124. do all the jobs that published software does. custom softwares are a
  1125. different area. we talking about published software. but we want to do
  1126. all the jobs that published software does. so that means we gotta be
  1127. allowed to do every job. if many of the jobs are prohibited for us, we
  1128. can never attain our goal. we can never attain our freedom all across
  1129. the globe. I should explain that the issue of free vs proprietary
  1130. software arises for published software. it doesn't arise for typically
  1131. custom software there you know if you write the program for one company
  1132. and they use it in-house. well, they do have, presumably, they do have
  1133. full freedom. They are not obligated to ever release it. but they have
  1134. the freedom to do so. so its fine. The ethical issue I've been talking
  1135. about arises when software is published. when its avlbl for users to get
  1136. copy. and btw, one consequence of this is that most of the jobs in the
  1137. software industry are not really going to be affected by free software.
  1138. its true that some jobs that involve developing proprietary published
  1139. software may go away.  there would be a certain number of free software
  1140. jobs replacing them  and we don't know how many there's going to be in
  1141. the future. but then there are going to be all the custom software jobs
  1142. and they may remain or a large fraction of them are likely to remain
  1143. unchanged. and this is, btw, an answer to the question of how a
  1144. programmer is going to make a living when all published software is
  1145. free. well one way is if worse came to worse, they could have jobs
  1146. writing custom software and they could write the free software in their
  1147. spare time for fun. but in fact there are jobs writing free software and
  1148. companies support these jobs in various ways. we don't really know
  1149. whether a free software will have more jobs or fewer jobs developing
  1150. published software than today's world. there's no way to tell except to
  1151. try it. but I believe, because there's an ethical issue here, software
  1152. must be free in order to treat the users decently. that we are better
  1153. off having free software even if its less software. a non-free program
  1154. is not a contribution to society. its a pit for users to fall into with
  1155. bay at the bottom.  because it looks attractive. if you are thinking
  1156. short term, if you don't value your freedom, you might be attracted to
  1157. that bay and then you might get into the trap. and then you might
  1158. starting other people to get into the trap. and when you are in the trap
  1159. you're divided from every body else. you know its not a trap where you
  1160. are all in there together and you can work to get out. its a trap where
  1161. you are divided.  and you are kept helpless. thats what makes it a trap.
  1162. so we have to recognise just the fact that you can make money, there's
  1163. nothing wrong with making money in enough itself. but just making money
  1164. doesn't justify mistreating other people. developing non-free software
  1165. is mistreating other people. so don't make a living that way. find
  1166. someother way to make a living. A way that involves ethical behaviour.
  1167. so, somtimes people have criticised me for my 'holier than thou'
  1168. attitude. well they are right.  I'm supposed to be holy. oh, oh, I'm
  1169. attached to this. its my job because I'm a saint (wears the black robe
  1170. and the halo hat). I'm not going to be able to keep this on for a very
  1171. long time in this hot weather.
  1173.         I bless your computer my child  [Laughter]
  1174.         I'm saint iGNUtius of the church of Emacs.
  1176.         Emacs started out as a text editor but it became a way of life for
  1177.         many computer users and then a religion.  
  1179.         infact there was an alt.religion.emacs newsgroup back around 1990.  I
  1180.         don't know what it was used for because I never read net news.  
  1182.         but so,emacs became a religion. we even have a great sism[?] between
  1183.         two different versions of emacs.
  1185.         and now we have saints. fortunately no gods yet.[Laughter]
  1187.         To join the church of emacs you must repeat the confession of the
  1188.         faith three times. you must say, "There's no system other than the
  1189.         GNU. and Linux is one of its kernels." [Laughter]
  1191.         The church of Emacs has some advantages compared with some other
  1192.         churches because to be a saint in the church of Emacs does not
  1193.         require celibacy [Laughter]
  1195.         however it does require making a moral commitment to live a life of
  1196.         purity and then living by it.  
  1198.         You must Exorcise the evil proprietary operating systems from all the
  1199.         computers you control.  and then install a wholly free operating
  1200.         system instead.  because wholly can be spelled in more than one ways.  
  1202.         and then only install free software on top of that.  
  1204.         if you make this commitments and live by it, then you too would be a
  1205.         saint and you too may eventually have a halo, if you can find one
  1206.         because we don't make them anymore [Laughter]
  1208.         Now sometimes people ask me "In the Church of Emacs, is it
  1209.         a sin to use vi?"
  1210.         The answer is that "Using a free version of vi is not a sin its a
  1211.         penance."
  1213.         and people sometimes ask me if my halo is really an old computer
  1214.         disk. [Laughter]  
  1215.         this is no computer disk.  this is my halo. But it
  1216.         was a computer disk in a previous life.  [Laughter]  
  1218. so, 'happy hacking' every one, and now I'll answer, oh btw, I have some
  1219. stickers here to give out. so let me put the stickers somewhere over at
  1220. this side and then you can take some as you're going or whenever and
  1221. take as many as you can make good use of. who wants to take these and
  1222. hand out giving them out. ok you can do it [srini comes up]
  1224. basically go stand somewhere in the back and as people are leaving, if
  1225. you have to leave now, you can get some stickers. There are a few
  1226. different kinds. The best use of them is to put them in a place where
  1227. they will stay permanently, so people will keep on seeing them and keep
  1228. on doing some good.  so lets see if this will reach here. so how are we
  1229. going to do this? is there are spare microphone for the audience?
  1231. What we should do is everybody should come down and stand here. get in a
  1232. line to ask questions. in that way whenever we're done with one question
  1233. the next person would be right there ready to ask. so please if you want
  1234. to ask a question form a line over here in the middle. and then we'll
  1235. have the microphone right there for you.
  1237. Or you can leave. you're free. and please speak as loud and as clear as
  1238. you can. looks like we're having technical difficulties.  [Laughter] .
  1239. lets see if he can fix them.
  1241. [rms yawns] MIC-TESTER: Hello... testing... testing...
  1243. questioner: RMS, I'm just curious to know what is powering your laptop?
  1245. RMS: I'm sorry I couldn't understand what you said. please try to speak
  1246. as clearly and slowly as you can. try to make every sound be heard.
  1248. questioner: I was just curious to know right now what is powering your
  1249. laptop.
  1251. RMS: What is running on my laptop?
  1253. questioner: exactly.
  1255. RMS: I'm running Debian GNU/Linux
  1257. questioner: Thats great!
  1259. audience: Applause. whistles.
  1261. questioner: I sometimes wonder whether these strong division between GNU
  1262. on one hand as a philosophy, moral standing and Linux on the other hand
  1263. really lets say puts the whole community thinking forward or whether it
  1264. hinders it on the other hand. See I give you a question: We had right,
  1265. in Germany,[...][inaudible]
  1267. RMS: I'm having trouble understanding, could you speak a bit slower and
  1268. try to make sounds clear?
  1270. questioner: Okay. I wonder whether the attempt to show the strong
  1271. differences from your view point between GNU on one hand and the
  1272. [inaudible] Linux and the Open Source movement on the other hand whether
  1273. this way of seperation helps the community at all in the moral standing
  1274. you are trying to promote? We had in germany for example, the last week
  1275. the problem that our govt had to go for new operation system for its
  1276. complete setup and it was something MicroSoft versus the open source
  1277. linux community. the...the... moral undertuned[?] [inaudible] between
  1278. the whole thing ... uh... sounded almost the way like you tried to
  1279. explain... GNU here. so is it really fair to try to seperate these so
  1280. strongly or wouldn't it be better to try to work on the similarities
  1281. between both ideas.
  1283. RMS: oh. I'll tell you why its vital to express the differences. the
  1284. reason is that we who talk about freedom as a goal are a minority. and
  1285. we're easily forgotten we could be completely ignored if we don't work
  1286. to spread these ideas. and you can see that happenning constantly. there
  1287. are constantly articles being published in the press about the use of
  1288. this version of the GNU system. which treated just as a commercial
  1289. alternative and never mention freedom at all. there are many people
  1290. talking about it that way. and those people generally use the term
  1291. opensource thats the term they chose for their , for what they choose to
  1292. say. well, we who want to work towards the goal of freedom, we have to
  1293. make our ideas heard. if we use the term open source. if we call the
  1294. system linux, we're going to be confused with every body else. I've seen
  1295. this happen many many times.
  1297. questioner: You mentioned Freedom 0. Which is the right to use...
  1299. RMS: Right to run it.
  1301. questioner: <shakes head> run for whatever you wish to. now we have
  1302. something called the SCCA. I could be wrong with the number of 'C's in
  1303. there. Which makes it mandatory for any program to build copy
  1304. protection...
  1306. RMS: [interrupts] basically the SSSCA is a proposed law in the US which
  1307. has a substantial amount of opposition and which we think will not pass.
  1308. but we're working on opposing it. Some big companies are also trying to
  1309. oppose it. we'll see what happens. But indeed that law would be an... a
  1310. vicious offense and it would prohibit free operating systems.  yeah,
  1311. basically there's no limit to how bad a law can be.
  1313. questioner: ok, I have this [inaudible] [...] freedom zero.
  1316. RMS: I'm having trouble hearing you at all because the sounds are not
  1317. clear. try to ennunciate the sounds clearly.
  1319. questioner: Again in conjunction with Freedom 0, there is for example,
  1320. we know about... the ASP hole in the GPL as of now.
  1322. RMS: We don't have a what?
  1324. questioner: The ASP.
  1326. RMS: you are talking about the ASP issue. let me explain what that is...
  1327. The issue is the GPL says that you can release the modified version to
  1328. the public but you are not required to. you can make a version and use
  1329. it privately yourself. and thats important. thats an important freedom.
  1330. but there are come programs which are designed [coughs] to provide
  1331. network services to the public. and there, [coughs] the original
  1332. developer can feel very bad if somebody else gets a copy modifies and
  1333. improves and sets up a network service himself. and then you see he's
  1334. never releasing his improved version. so the result is that his
  1335. improvements never become available to the initial developer. so we are
  1336. about to try out a clause that can help with this problem. which says
  1337. that if you are using the program to provide network services to the
  1338. public, then you must have a command in the server which allows the user
  1339. to download the source code from your server, because you see, we
  1340. believe that you have the right to make modifications for your private
  1341. use, but setting up a public network server is not really private use.
  1342. its a different kind of a publication. you can think of it as being
  1343. analogous to public performance of a piece of music. and so, it seems
  1344. legitimate in that case to say, if you are inviting the public to come
  1345. and use your server you must also let the same public get your modified
  1346. source code. Now this would be an option. This wouldn't apply to any
  1347. existing software. because in order to make it apply you have to release
  1348. your initial version so that it has this download command.
  1351. questioner: my...
  1353. RMS: [interrupts] so this were thinking of putting this version into GPL
  1354. version 3, and when we do it not change anything for any existing
  1355. software. But it would mean people can develop programs that activate
  1356. this clause and then it would apply, but only in that narrow situation.
  1358. questioner: my question was actually analogous to the ASP situation.
  1359. Actually what can you do... what can the... uhh... FSF do if somebody
  1360. takes a version of the GNU/Linux kernel and puts it in an embedded
  1361. system and adds various copy protection routines to that system and
  1362. distributed as... you know... effectively as ICs or even as devices
  1363. which people can buy like the playstation or [inaudible]
  1365. RMS: well, they are allowed to do that. and if you can't, you know the
  1366. hardware doesn't give you a way to install software well we don't insist
  1367. it has to give you a way to install a software. on the other hand they
  1368. will be required to publish a modified version of their source code. so
  1369. people could use it in making someother kind of device. I think its time
  1370. now to give the next person a chance to ask questions.
  1372. questioner: umm...It is public information...
  1374. RMS: [interrupts] I'm sorry I cannot understand you. please try to make
  1375. every sounds clear. I'm hard of hearing and there's a lot of noise
  1376. here...
  1378. questioner: ok. I'm sorry...
  1380. RMS: [interrupts] and [...][inaudible] accent is very strange to me.
  1381. with all those three things together it is almost impossible for me to
  1382. understand anyword you say.
  1384. questioner:  It is public information that you favour the Debian
  1385. Project.
  1387. RMS: I'm sorry I can't understand. Can you enunciate each sound clearly.
  1389. questioner: It is public information that you favour the Debian project
  1390. and you also run the Debian GNU/Linux operating system on your laptop
  1391. and so it is... it... the Debian project has been happening for a very
  1392. long time now. why is it that you filed for a develop... developer so
  1393. late in november last year as opposed to earlier say...
  1395. RMS: [interrupts] I'm having trouble. I heard the first half, but when
  1396. you started saying the actual question 'why is' I couldn't hear it.
  1397. could any one.. can you repeat to me what he asked [pointing to
  1398. Dr.P.Sriram]
  1400. DPS: [inaudible].. his question is basically why did wait so long to
  1401. become a debian developer.
  1403. RMS: It didn't occur to me that it was a good thing to do  [Laughter]
  1404. unfortunately it looks like my application has been put on hold because
  1405. I have not had the time to package a program and two overwhelmed with
  1406. works so I can't do it. Its unfortunate, I hope I'll be able to get to
  1407. it sometime soon.
  1409. questioner: What do you think has been the impact of the Free software
  1410. movement in India as such. What do you think needs to be addressed?
  1412. RMS: Did you say, what is the impact of free software movement in India?
  1414. questioner: yeah
  1416. RMS: well, some Indian govts are now starting to actively sponsor free
  1417. software use and development. In... for instance govt departments in
  1418. Kiosks, in schools. Its just beginning. But we have hoped that this will
  1419. spread a lot more. and it has the potentiality to help with the
  1420. elimination of the digital divide. because its hard to put very many
  1421. computers into the hands of villager's schools when you have to pay
  1422. Microsoft and Adobe and all those companies for licenses. you know you
  1423. could spend that same money on computers it would be much better and you
  1424. could get a cheaper computer because GNU/Linux is more efficient that
  1425. would be better too. and not only that, when you have free operating
  1426. systems in the schools in these villages that provides a great
  1427. opportunity for anyone who wants to learn to do system administration
  1428. or.. or programming because they can actually read the existing programs
  1429. and then they work on them. they can learn by doing. learn by tinkering.
  1431. questioner: therewasthisarticleyesterdayonslashdot...
  1433. RMS: Slow down please...
  1435. questioner: There was this article yesterday on slashdot where you were
  1436. quoted as saying that the HURD is of age. in the sense that the HURD is
  1437. going to be independant and the hurd kernel....
  1439. RMS: [interrupts] not that its going to be available, its likely to be
  1440. available. that it... it... its a... its a.. an estimate, not a promise.
  1441. I see so many people getting that wrong. the article said it right but
  1442. various people repeated to me leaving out that crucial point. I don't
  1443. know.  But what I'm saying is ... its close.. its so close now already.
  1444. but there's not much further to go, the GNU/Hurd system which is the GNU
  1445. system using the GNU Hurd as its kernel, its basically working now...
  1446. and Anand Babu one of the main Hurd Developers, he is in Bangalore, his
  1447. name is Anand Babu, he showed me his laptop with GNU/Hurd running on it.
  1448. There were a couple of missing features that I think are pretty
  1449. important but they... they... they wont be too hard. they should get
  1450. done this year I expect.
  1452. questioner: assumingthatthehurdsystemisready...
  1454. RMS:  S L O W E R ! P L E A S E...
  1456. questioner: Assuming that thehurdsystemisreadysoon...
  1458. RMS: I'm sorry could you repeat what he said? [pointing to Dr.P.Sriram]
  1460. questioner explains to DPS and the first few words are heard...
  1461. "Assuming that the hurd system is ready soon [inaudible]"
  1463. DPS: Assuming the Hurd system is ready soon do you see it competing with
  1464. the Linux?
  1466. RMS: Well, the Hurd is not a _system_. the hurd is a _kernel_. The Hurd
  1467. and Mach together are a kernel. and linux is also a kernel and yes, they
  1468. _are_ alternatives. so it will in some ways compete. Both of these
  1469. kernels are used in the GNU system
  1471. questioner: ah, good evening Mr.Stallman. My question to you is, How do
  1472. I build a successful software model, business model based on free
  1473. software because it has become increasingly difficult to convince the PC
  1474. community about... because the first quote [...][inaudible] is do you
  1475. have a software which is proprietary, which is close so that it can
  1476. cause entry barriers for other people. so what would be your advise to
  1477. people like us who can convince the PC community.
  1479. RMS: can you repeat it for me? [pointing to Dr.P.Sriram]
  1481. DPS: What would be your advise to developers who are looking for
  1482. funding... who are looking for the... the... funding people... the..
  1483. the...  venture capitalists.
  1485. RMS: oh, give up on venture capitalists... thats the wrong approach...
  1487.  [Laughter]
  1489. RMS: But you don't want to start a company funded by a venture
  1490. capitalist anyway, because in a year or two they would take it away from
  1491. you and kick you out. I mean its really a stupid way to do anything. the
  1492. only reason anybody would do that, is money is the most important thing
  1493. to that person and he's desperate to get rich. which is stupid.
  1495. questioner: thanks
  1497. RMS: Now there are people who have found ways to make a living doing
  1498. free software for instance there are companies that configure and setup
  1499. machines for clients using free software and then writing additional
  1500. stuff to do what the client wants. and thats the way people can make a
  1501. living. and they support people developing free software so thats.. you
  1502. know, thats something to consider. it doesn't lead to getting rich. but
  1503. ofcourse most companies that people start are not successful anyway,
  1504. they may think I'm going to get this venture capital and I'm going to
  1505. make a big splash and I'd get very rich. and chances are it just goes
  1506. broke in a few years. you know, over 90% of the time it fails within a
  1507. few years, the cases that are successful are very few. so really, if you
  1508. think of that as a way to succeed you are deluding yourself anyway.
  1510. questioner: The
  1511. GNU/[inaudible]...[inaudible]...[inaudible]...[inaudible]...
  1513. RMS:  Okay could you please repeat the question? I didn't hear anything.
  1515. same questioner: The [inaudible]...[inaudible]...[inaudible]
  1517. RMS: No. Please, you repeat the quetion because you heard it. I know I
  1518. can hear you.
  1520. DPS: The GNU Hurd and GNU linux. They share a lot of things. Why don't
  1521. they work together? including ideas. they share a lot of ideas. why
  1522. don't you work together closely?
  1524. RMS: Now this time I couldn't really hear _you_. Its really hard...
  1526. MIC-MAN: check...check...
  1528. DPS: I'll more try. The GNU Hurd system and the GNU Linux
  1529. system share many things. including ideals. why aren't they working
  1530. closer together.
  1532. RMS: well, they are basically working pretty closer together. most of
  1533. them is the same. its just the kernel thats different. well the... well
  1534. the C library has to be fairly different as well.
  1536. questioner: [inaudible]... the kernel groups to join together.
  1538. RMS: You know the... the user space programs are basically the same.
  1539. The same source typically run with both kernels.
  1541. questioner: [inaudible].. the kernels... the kernel groups to join
  1542. together.
  1544. RMS: I can't hear you at all. I'm sorry.
  1546. questioner: The   KERNEL    GROUPS   to join together.
  1548. RMS: oh that wouldn't work at all. and the reason is that the two
  1549. kernels are totally different. they have so little in common its just
  1550. not useful for these kernel groups to join together. The design is
  1551. totally different. you know, ones a monolithic kernel and the other is a
  1552. micro kernel design. which linus torvalds doesn't like. also...
  1553. [Laughter]  so you can be sure he's not going to wanna work on the Hurd.
  1555. questioner: You were talking about patents for software ideas. which you
  1556. wanted to prevent. are such patents currently in force in the US?
  1558. RMS: Yes. Software patents are... exists. they're more than a hundred
  1559. thousand in the US. there may be several hundred thousand by now. and
  1560. its a big problem.there are a number of free programs that we don't
  1561. have. more atleast in the US you can't find them. because they've been
  1562. driven underground.
  1564. questioner: What do you exactly mean by patents for software ideas?
  1565. ideas in the sense algorithms? data structures? and such things..
  1567. RMS: [inaudible] yes. an algorithm could be patented, or something more
  1568. general than an algorithm could be patented. or a feature could be
  1569. patented. or an idea for data format can be patented. or an idea for a
  1570. protocol can be patented. uh... for instance the LZW data compression
  1571. algorithm used in GIF format is patented. so you can't write a program
  1572. to generate compressed GIFs in the US without getting a permission from
  1573. the patent holder. Thats just one example but there are many examples.
  1574. there's too many for me to keep track of.
  1576. questioner: When there are too many things to keep track of, naturally
  1577. there's a possibility that somebody develops something on his own which
  1578. by mistake, maybe similar to a patented thing, so how do I resolve such
  1579. issues?
  1581. RMS: how would what? how... I didn't hear the very end. I heard most of
  1582. it. the last sentence I didn't hear.
  1584. questioner: How do you resolve? like accidentally... because the
  1585. ideas...
  1587. RMS: [interrupts] ok. ok. now I heard you. What happens is the patent
  1588. holder will threaten you and either say, you just have to stop or he'll
  1589. say I'll let you continue if you pay me a lot of money. and at that
  1590. point you can either 'give in' you can either...[interrupts]  when he
  1591. just says stop, you have two choices. you either stop or you either go
  1592. to court.  if he says in order to do this you have to pay me a lot of
  1593. money, you have three choices. you can stop. you can pay him they money
  1594. or you can go to court. now if you are doing this in a company then
  1595. maybe you can afford to pay. but if you are doing this as a volunteer,
  1596. you'll probably don't have the money to pay. and usually they demand a
  1597. price per copy.  well, with free software its impossible to collect a
  1598. price per copy bcos we can't count the copies. Its literally impossible,
  1599. even if it were one thousand of a rupee per copy, we couldn't pay it
  1600. because we can't count the no. of copies.
  1602. questioner: I have two questions. the first one ...
  1604. RMS: [interrupts] Please speak loud and clear!
  1606. questioner: I have two questions with me. the first one is whenever I go
  1607. in for a new system why is it that it always carries the tag "Microsoft
  1608. Operating system Preloaded".
  1610. RMS: Could you repeat... I couldn't understand that. so could you repeat
  1611. it?
  1613. DPS: Its really a hardware question.. When you buy a new PC, why does it
  1614. say Microsoft Software Loaded.
  1616. RMS: Well, I mean... I don't know. I mean... you know much about this as
  1617. I do. I think the reason is that Microsoft effectively has a near
  1618. monopoly and because of that the PC companies think that all the users
  1619. are going to want Microsoft. I think its unfortunate, but I can't tell
  1620. them what to do. Your second question?
  1622. questioner: Take for example IBM, ok? he says that some of his models
  1623. are lienux compatible. but why is it that he does not favour a system
  1624. which is shipped with Lienux?
  1626. RMS: oh, they are equipped with GNU/Linux. But IBM calls it Linux which
  1627. is not right but I didn't hear the other word so maybe you could repeat
  1628. the question to me so I can answer that.
  1630. DPS: you know... IBM has both... Microsoft loaded and GNU Linux
  1631. compatible. Why aren't they loading their machines as loaded with GNU
  1632. Linux.
  1634. RMS: Why aren't they whatting their machines?
  1636. DPS: Labelling. putting a tag.
  1638. RMS: Labelling them. I don't know you have to ask them. But... you
  1639. know... I know that they wouldn't label them as GNU/Linux because they
  1640. don't call it that. They might label them as Linux if you ask them to
  1641. but they might have some reason why they don't. Who cares. I don't know.
  1642. you have to ask them.
  1644. questioner: Many people argue that ... the... the philosophy of the free
  1645. software can be arbitrarily applied to any other areas as well. like
  1646. production of music... writing books... or even digital logic circuit
  1647. design...  so...
  1650. RMS: Oh I couldn't hear that...
  1652. questioner: Chip design. chip design for example.
  1654. RMS: chip... oh Chip design. oh well, yes and no. Basically you could
  1655. try to apply it to any kind of information but its not equally important
  1656. for all kinds of information basically the area for which its important
  1657. is functional information. Information typic...basically that ordinary
  1658. people could use to get something done. so that includes programs and
  1659. recipes for cooking and dictionaries and encyclopedias and manuals and
  1660. text books... Chip designs because they can only be used by Chip
  1661. fabrication plants, I think they are a less important issue. but yeah
  1662. they could be free... and there are people starting to work to some
  1663. extent on free chip designs and on free hardware designs. I don't think
  1664. that free hardware designs are vital. However its nice. so we do say its
  1665. a good thing when people work on free hardware designs. because its a
  1666. lot of work to build hardware from a design. You can't just type 'build'
  1667. and has something build the hardware for you. well it doesn't raise the
  1668. same important ethical issue that the issue of free software raises. its
  1669. just nice if the hardware design is free.
  1671. questioner: and does that have anything to do with the fact that its a
  1672. lot of money to develop hardware.
  1674. RMS: I didn't hear any of that could you repeat it for me?
  1676. questioner: Is it also because one has to spend a lot of money to
  1677. develop hardware? to develop chips?
  1679. RMS: Not really. In that... that would not be a primary issue if that
  1680. were the only one I don't think it would convince me. but the fact that
  1681. you need a lot of money to _build_ the hardware that its not the kind of
  1682. things individuals typically can build. and if you do its a lot of work
  1683. anyway. because of that I think its a less important issue ethically.
  1685. questioner: thank you very much.
  1687. questioner: I have a question in the subject of kernels. The Linux
  1688. kernel is a monolithic kernel. on the other hand the Hurd for example is
  1689. a Micro Kernel. Now linux grew out of the Minix project. In 1991  when
  1690. Torvalds posted to comp.Operating System.minix newgroup about linux, he
  1691. was very severely criticised by Andrew Tanenbaum because he believed
  1692. that Monolithic kernels... they are technically inferior...
  1694. RMS: I... I have a trou... I think I know what you are talking about but
  1695. I can't hear you most of the words. If you could make it brief and then
  1696. he could repeat it then I could try to answer you.
  1698. questioner: I'll make it brief. A monolithic style was considered
  1699. technically inferior, but it has proved to be a more practical style for
  1700. kernels. Why is this so?
  1702. RMS: Could you repeat it now?
  1704. DPS: umm... Monolithic vs. Microkernels... the success points to you
  1705. know that Monolithic kernel can work.
  1707. RMS: And therefore... whats the question?
  1709. questioner: A monolithic kernel style was considered to be technically
  1710. far less inferior as early as 1991 but yet even 10 years later the most
  1711. practical kernel that we have today is a monolithic kernel. Why is it
  1712. that even inspite of being technically inferior it has proved to be
  1713. practically more feasible alternative.
  1715. RMS: k, I didn't hear it so you could ...[inaudible]
  1717. DPS: The monolithic kernel would be considered technically inferior. Yet
  1718. it is so popular why?
  1720. RMS: Well, I don't have any ...[inaudible] which one is superior. What I
  1721. see is that the micro kernel design with multiple servers offers some
  1722. advantages and power and flexibility. There are many things you can do
  1723. with it that are hard to do with a monolithic kernel. on the other hand,
  1724. Linux has had an advantage in speed and reliability. Now we have partly
  1725. closed the gap in reliability, not totally. As for speed its less
  1726. important now because the computers are so damn fast. so which one is
  1727. really going to be technically better? I don't know. But I'm glad that
  1728. people are working on the hurd and make it as good as they can.
  1730. questioner: I have one more question. You have said that people calling
  1731. the GNU/Linux system as a Linux System and forgetting about the GNU
  1732. philosophy is a very bad thing. but my question is how serious is this
  1733. problem: For example there was a student in my class  who believed that
  1734. the linux kernel itself was actually written by the Free software
  1735. foundation and he thought that all linux softwares are actually GNU
  1736. softwares. so my belief is that people are not at all ignorant of the
  1737. role of the GNU and of the philosophy of the GNU. So how serious is
  1738. this...
  1740. RMS: [interrputs]... very serious. and most of the users have heard of
  1741. the philosophy of the GNU and most of them have barely heard the name of
  1742. GNU. If they've learned about GNU the fraction who've learned about GNU
  1743. at all they think... most of them think, that GNU was a project to
  1744. develop some software tools. and that there was no particular overall
  1745. reason why we were doing that. We just felt like doing that. because you
  1746. see, if you think at it that way, the whole logic is gone! there was a
  1747. reason why we developed a free Operating System. Its because the
  1748. Operating System is the first thing you need to run a computer.  and
  1749. with an Operating System you could run your computer. so that has to
  1750. be... so everything we did was all arranged witha logic of making it
  1751. possible to have freedom in community. but the picture...  those
  1752. fraction who hear about GNU at all... the picture they get is wrong. and
  1753. it leaves out the most important logic of it which relates to the
  1754. ethical philosophy a year ago I gave a speech in the Netherlands and
  1755. someone came up to me afterward and said, I've been using this system
  1756. for 5 years and this is the first time anyone told me that the purpose
  1757. of it is for freedom. so if... he wasn't just a new comer. He is a
  1758. person who works in a free software company. and yet, he never heard of
  1759. this.  so I believe you that you heard a person who got the opposite
  1760. misunderstanding. all sorts of misunderstandings are possible. but I can
  1761. assure you that the most common misunderstanding is the idea that Linus
  1762. Torvalds started everything in 1991 and that the purpose of it all is
  1763. just to be powerful and reliable and that its fine if you use non-free
  1764. software with it to make it more popular. and that this is more common
  1765. than a correct understanding. the misunderstanding is more widespread
  1766. than the truth.
  1768. questioner: thank you
  1770. questioner: yeah...ahh... There was a mention in the Linux Gazette of
  1771. this month saying that the G N U... GNU will be stopping the manuals...
  1772. the publishing of manuals in the distributions... it basically said
  1773. something like...
  1775. RMS: we'll be doing what?...
  1777. [tape exhausts]
  1779. This Documentation is released under the GNU Free Documentation license.
  1780. Please take a look at for
  1781. more information on the Free Documentation License.
RAW Paste Data
Pastebin PRO Summer Special!
Get 60% OFF on Pastebin PRO accounts!