web design software on linux
any comments, suggestions and bug reportsregarding the subtitle.e-mail to: geek@geekbone.org i was at agenda 2000 and uh, one of the people who was therewas craig mundie, who is some kind ofhigh mucky muck at microsoft, i think uh, vice-president of consumer productsor something like that. and uh, i hadn't actually met him i, i, i, uh, bumped in to him in an,in an elevator... in an elevator and uh, i looked at his badge and said,"oh, i see you work for microsoft."
and he looked back to me and said,"oh, yeah and what do you do?" and i thought he seemed just a sort of a tad dismissive i mean, here's the archetypal, you know,guy in a suit looking at a scruffy hacker and so i gave him the thousandyard stare and said, "i'm your worst nightmare." wonderview productionspresents aj.t.s. moorefilm revolution os
for most its short, but colorful history, the computer industry has been dominated by the windows operating system. but that could soon change, as windows faces a strong challenge from linux silicon valley has long been the place to develop new technology, start new companies and get really rich. now the valley is the front line in a revolution
fighting for that mostpolitically incorrect of ideas: individual freedom day and night,a loose confederation of hackers and programmers zaps bits pieces of computer code around the world as it builds the tools to set computer users free using open information andthe free exchange of technology to achieve its goals. this revolution began in the 1980's with
the free software movement and gnu project. and now is most commonly associated with linux and the open source movement. what is linux? we do have one sector that taking off today. it is the linux-related sector. and i thought this might bea good opportunity to say, "what is linux?" and i'll answer this question for you.
many of you probably already know, butthere are 12 million users out there a computer operating systemdeveloped by hundreds of programmers collaborating on the internet a challenge to microsoft windows nt very popular for its speed and so this's what the craze is about to kind of explain what linux is you have to explain what an operating system is
and... the thing about operating system isyou, i mean... you're never ever supposed to see it. because... nobody really uses an operating system, people use... programs... on their computer and the only mission in life of an operating system is to helpthose programs run. so an operating system never doesanything on its own it's only waiting for the programs to
ask for certain resources or, or, ask for a certain file on the disk or ask for the programs to connect them to the outside world. and then the operating systemcomes, steps in and then tries to make it easy for peopleto write programs and, what is open source? open source is a way for people to collaborate on software without being encumberedby all of the problems of intellectual property,
having to negotiate contracts every timeyou buy a piece of software, have a lot of lawyers involved. in general, we just wanna get the software to work and we want be able to have people contribute fixes to that, etc.. so we sort of sacrifice some of the intellectual property rights and just let the whole worlduse the software before there could be linux
there was richard stallmanand the free software movement. they think of richard stallman as the... great philosopher, right. and think of me as the engineer richard stallman is the founding father ofthe free software movement. through his efforts to build thegnu operating system. he created the legal, philosophical andtechnological foundation for the free software movement. without these contributions,
it's unlikely that linux and open source would have evolved in to their current forms today i joined themit artificial intelligence lab in 1971, i joined... a thriving community of hackers, people who loved programming, loved exploring the what they could dowith computers. and they had developeda complete operating system, entirely written there. and i became one of the team, that
continued to improve the operating system,adding new capabilities that was my job, and i loved it, we all loved it.that's why we were doing it. and [clears throat] we called our system "the incompatible time sharing system" which is an example of the playful spirit which defines a hacker. hackers are people who enjoyplayful cleverness. well, it first started going wrong
as the outside world started pressuring usto have passwords. we didn't have any passwordson our computer. and the reason was that the hackers who'd originally designed the system realized that passwords were a waythe administrators could control all the users. and they didn't want to build tools you know, locks and keys forthe administrators to control them, so they just didn't do it.they left that out and we had the philosophythat whoever sitting at the computer
should be able to do whatever he wants and somebody else who was there yesterdayshouldn't be controlling what you do today when they put passwords onto one ofthe machines at mit i and bunch of other hackers didn't like it, i decided to try a subversive sort of hack.[clears throat] i figured out how to decode the passwords, so by looking at the database of encoded passwords i could figure out what each personwould actually type to login and so i sent messages to people, saying,
"hello? i see that you'vechosen the password mumble, whatever it was.how about if you do as i do just type enter for your password it's much shorter, much easier to type". and... of course with this message i was implicitly telling themthe security was really just a joke. anyway, but in additioni was letting them in on this hack. and eventually, a fifth of all theusers on that computer joined me in using just enter as their passwords.
[ where did the ideas that lead to what is nowcalled open source world? how did that begin? who began that? ] it actually began with the startof computers because at that time software was just passed around between people and i think it was only like in thelates 70's or early 80's that people started really closing uptheir software, and saying, "no, you can neverget a look at the source code. you can't change the softwareeven if it's necessary for you to fix it, for your own application"
and... um, you can actually blamesome of that on microsoft, they are one of the real pioneersof the proprietary software model. in the mid 1970's, a group of hackers and computer hobbyist in silicon valleyformed the "homebrew computer club". in the club january 31, 1976 newsletter, bill gates of the recently formed microsoft, wrote an open letter to the community where he made a point by point argument forthe relatively new concept of proprietary software up to that point, the practice of computer usershad been to freely pass around software
with not much thought given toits ownership known as an open letter to hobbyists,bill gates writes, "to me the most critical thingin the hobby market right now is the lack of good software courses,books and software itself. without good software and an ownerwho understands programming, a hobby computer is wasted. will quality software be writtenfor the hobby market? gates goes on to write,"the feedback we have gotten from the hundreds of people who say they areusing basic has all been positive.
two surprising things are apparent, however. 1) most of these "users" neverbought basic, and 2) the amount of royalties we have receivedfrom sales to hobbyists makes the time spent on of altair basicworth less than 2 dollars an hour why is this? as the majority of hobbyists must be aware, most of you steal your software. hardware must be paid for but softwareis something to share. who cares if the people who workedon it get paid?
is this fair? one thing you don't do by stealing software is get back at mits for some problemyou may have had mits doesn't make money selling software. one thing you do do is preventgood software from being written. who can afford to do professional work for nothing? what hobbyist can put 3-man years into programming, finding all bugs, documenting his productand distribute it for free? the fact is,
no one besides us has invested a lot of moneyin hobby software. what about the guys who resell altair basic? aren't they making money on hobby software? yes, but those who have beenreported to us may lose in the end. they are the ones who givehobbyists a bad name, and should be kicked out of anyclub meeting they show up at. i would appreciate letters fromanyone who wants to pay up, or has a suggestion or comment. signed bill gates, general partner, micro-soft.
in the late 70's and early 1980's,richard stallman was doing artificial intelligence research and coding at the mit artificial intelligence lab. richard had a number of negative experiences during that period which soured him on the whole idea of commercial software. [ such as? ] uh... some company wanted to work on andwanted to fix was locked up. and he couldn't get the company that ownedon the code to let him fix it
even though it would have been to their advantage to do so. and that put me into a moral dilemma, you see? because to get one of the modern computersof the day, which was the early 80's, you would have to get a proprietoryoperating system. the developers of those systemsdidn't share with other people, instead they tried to control the users, dominate the users, restrict them. say, if to get the system, you have to sign a promise you won'tshare with anybody else.
and to me that was essentially a promiseto be a bad person, to betray the rest of the world, cut myself off from societyfrom a cooperating community. and i had already experienced what happenedwhen other people did that to us, when they refused to share with us. because they had signed these contracts. and it hurt the whole lab, it kept us from doing useful things before. so i just wasn't going to do that.
i thought,"this is wrong!i am not going to live this way" and from experiences like thishe developed a profound hostility to the idea ofintellectual property and software. he eventually acted this outby founding the free software foundation. so, i looked for another alternativeand i realized: i was an operating system developer. if i were to develop another operating system. and then as the author,encourage everyone to share it. say, everyone, " you come and get it,use this, form a new community"
not only could i gave myself a wayto keep using computers without betraying other people,but i'd give it to everybody else, too. everybody would have away out of that moral dilemma and so i realized this waswhat i had to do with my life. i actually began the project in january of 1984. that's when i resigned for my job at mit to start developing the gnu operating system. now i should explain the name gnu is a hack. because it's a recursive acronym.
it stands for "gnu's not unix". you see so the "g" in "gnu" stands for "gnu". and what the name means is i was developing a system that was likethe unix operating system, but was not the unix operating system. this was a different system. we would have to write it completelyfrom scratch because unix was proprietory. we were forbidden to share unix,
we couldn't use unix.it was useless for a community. so we had to write a replacement for it. throughout the 1980s, as richard stallman was building the gnu project, computer scientists fromthe university of california at berkeley were developing their own freeoperating system. known as berkeley unix, or bsd, it was based upon the unix kernelwhich had been licensed from at&t. however, due to legal problems with at&tand fragmentation of the source code,
hackers and other non-institutional userswere slow to adopt it well, unix consisted of a large number ofseparate programs that communicated with each other. so we just had to replace these programsone by one. so what i started doing waswriting a replacement for one program, and then another, and then another, and then people started joining me, because i published an announcementinviting other people to join me to help write these programs.
and uh... and by around 1991,we had replaced practically all of them. [ what were some of the programs that you ... ] well... we had to... to have a complete system, you need to have a kernel, which is the program that allocates resources to all the other programs, you need a compiler, which translates a program from readable source codethat programmers can understand into numbers, mysterious numbersthat the computer can actually run. you need other programsthat go with the compiler to help do this job.
you need a debugger. you need a text editor. you need text formatters. you need mailers... you need lots and lots of things. there are hundreds of programsin a unix-like operating system. i saw stallman's announcement. actually i met him in february of 1987. he came to give a five-day tutorialon emacs at our company.
and during the day he would explain new ways to think about emacs and ways to extend it, enhance it,and to use the emacs source code uh, for better or worse. but in the evening,he was busily working on this compiler, and he had not yet released it to the public, so he was uh, being a little bit uh,careful about who, who got to see the source code. but i was very eager,and when he first announced it in june,
i downloaded it immediately. i, i played with it. i got some, some pointers from him. and when i sent the source code back to him, he was very, uh, actually amazed that how quicklyi was able to ramp up on his technology. whenever we worked on somethingat stanford or in the university, we would get, mostly at the time we were working off machinesfrom digital equipment or sun, mostly sun.
whenever we would get a sun machine, the first thing we would do iswe would spend literally days downloading gnu free software from the internet, building it and installing it on that sun machine. the crucial thing about gnu is thatit's free software. and free software refers not to price,but to freedom. so think of free speech, not free beer. the freedoms that i am talking about are the freedoms to make changes if you want to,
or hire somebody else to make changes for you if you're using a software for your business, to redistribute copies,to share with other people, and to make improvements and publish them so that other people canget the benefit of them, too. and those are the freedoms that distinguishfree software from non-free software. these are the freedoms thatenable people to form a community. if you don't have all these freedoms, you're being divided anddominated by somebody.
my first experience contributing to free software came in late 1989, early 1990. i was working as a graduate studentat stanford university on computer aided design tools. one of the pieces i needed was a tool called a parser generator. well, the free software foundation under richard stallmancreated a great tool called "bison". i needed a tool that worked with c++.bison worked with c.
i modified bison to create somethingcalled "bison++". and it's a tremendous feeling of empowerment be able to take a piece of software that was available and create what you neededin a very short piece of time by modifying it. i put it back on the internet and i was amazed at the number of people that picked it up and started using it. in fact, i remember going to, uh job interviews, i, at various times, consideredjust going out getting a job.
and i'd gone to a job interview. and i was talking to one of the people, and i started asking them about what tools they used, and they said, "gee, we used bison++!" and i said, "oh, i am the author of bison++!" free software generally does have a copyright. it does have an owner. and it has a license. it is not public domain.
if we put the software in the public domain, somebody else would be able to makea little bit of changes and turn that into a proprietory software package, which means thatthe users would be running our software, but they wouldn't have freedom to cooperate and share. to prevent that, we use a technique called "copyleft". the idea of copyleft is thatit's "copyright" flipped over. and what we do is, we say, this software is copyrighted
and we, the authors give you permissionto redistribute copies, we give you permission to change, we give you permission to add to it. but when you redistribute it, it has to be under these terms,no more and no less. so that whoever gets it from you also gets the freedom to cooperatewith other people, if he wants to. and then, in this way everywhere the software goes, the freedom goes, too.
and it becomes an inalienable right to cooperate with other people and form a community. [ and so, what is that? the license?what was that... ] well, copyleft being a general idea, in order to use it, you have to have specific example. the specific example we use formost gnu software packages is the gnu general public license, a particular document in legalesewhich accomplishes this job. a lot of other people use that same license,for example,
linus torvalds uses that license for linux as well. well, the license i use is thegnu general public license. that's the one richard stallman wrote. and i think it is really astounding contribution. uh, it's one of the few software licensesthat was written from the standpoint of the community rather than from the standpoint of um, protecting a company or um, as is the case with mit and bsd license performing the goals ofa government grant program.
uh, and the gpl is really unique in that. it's not just a license.it's a whole philosophy that, i think, motivated the open source definition. i don't hide that a lot of what i do came from stallman. a crucial step in the growth of gnu/linuxand the free software movement was the creation of businessesbased upon the software and philosophy. grown zero for the beginning ofthe business phase was the electronics research labat stanford university. known as erl, the lab was the place for the first gnu
and linux business founder inspiration. so right here was where erl was. that would have been the entranceover there next to the uh, electrical engineering mccullough building. as you walk in, you come in,you walk down the hallway, down here. my office would have been about, about here. and then right across the hall,from that was michael tiemann's office. michael tiemann took uh.. andstarted a company, cygnus software with the idea was to sell consultingand services around the gnu free software
and, well michael's done very well with cygnus. well uh, i spend a lot oftime working out uh, how we were going to make money and in the original gnu manifesto which is the last chapterof the gnu emacs manual. stallman proposed a number of different possible ways to make money. from the beginning of thefree software movement i had the idea that there's
room in it for business to be done. one of the advantages of free software is that, there's a free market for any kind of service or support. so if you are using software in your business, and you want good support, you have a choice of people to go to for it, you have a choice of businesses... that are in the business ofproviding you with support. so they are going to have to in general give you
good support or you go to somebody else. with proprietory software, support is a monopoly, there is one company, typically, that has the source code and only they can give you support so typically, you are at the mercy of a monopoly. that's the case, for example with microsoft. so no wonder the support is so bad. the benefits of free software were
tremendous but the cost of supporting it internally uh, and made managers very very nervous and so the fundamental idea i had wasif we can build a model that could deliver two to four times the support and uh, and uh, and hand holding capability that an internal engineer could provide. and we could do it at 1/2 to 1/4 of the cost that would meet the test of whether or not people would actually buy.
and by about the fall of that year, we had all the things worked out about who needed on the technical team, what the terms the sale would be, what the key price point were, and we actually received our incorporation in nov of 1989. one of the most difficult things in starting our company was actually finding a name for it. i explained this to one of my friends
"we're having difficulty" and he returned an e-mail message that basically just had a bunch of wordswith the name "gnu" in it. and "cygnus" was the one thatlooked least obnoxious and least obscene. i can say very clearly that cygnus was the first business that specialized in free software. cygnus supported free software, filled a very essential niche becausewe had this great software, you could get it for nothing butyou couldn't get support - they made their money
by charging for support. the gnu project started by building a toolkit, uh, basic development tools such asa c compiler, a debugger, a text-editor, and uh, other necessary apparatus. and their intention was eventually todevelop a kernel to sit underneath those and be the center of the operating system. by about 1990 they had successfullydeveloped that toolkit, and it was in wide use on great many variants of unix. but there was still no free kernel.
the kernel happened to beone of the last things we started to do and we had started it not long before. and that's when linus torvalds came along. "lin-us" or "line-us"? what's the exact,do you prefer the pronunciation? um.. when i speak swedish it's "lee-nus"; when i speak finnish it's "leen-ous"; when i speak english it's "line-us". and i really don't care how people pronounce my name. but "linux" is always "linux".
he developed a kernel, and got it workingfaster than we got ours working, and got it to work very nicely and solidly. his kernel is called "linux". the initial goal was my very personal goal to be able to run a similar environment on my computer that i had grown used to at, at the university computers. and i could not find anything that suited me for that. so having been doing computers for all my life basically. at that point i decided that i'll do my own.
most of the inspiration early on came from, from sunos which was what um,i was using at the university at the time. [ which university? ] university of helsinki in finland. from 1991 to about 1993 was really i guess the infancy period of linux. that was when it was still only alpha or beta quality; it was relatively unstable. although, even then it was a good deal more stable
than a lot of what are now called"production" operating systems. linus used the traditionaltried-and-true method of writing one program that does the job, and he got it to work. quickly in fact faster thani would have thought was possible. the term for it is "monolithic", which means that basicallythe os itself is one entity, indivisible. uh, while in the microkernel, the, the operating system kernel is actually
uh, just a collection of servers that do different things and then they have a common protocol for doing communication between themselves. [ so why is that... the gnu project that's hadso much lead-time, that's been doing this, why...why is it that he was able to kindacome in at the tail end so to speak ] well we actually started thegnu hurd not long before he started linux. and it happened though we chose a designthat's a very advanced design in terms of the power gives you but also turns out to be very hard to debug.
we decided to divide up the kernelwhich traditionally had been one program, to divide it up into a lot of smaller programs that would send messages toeach other asynchronously to, to communicate. the problem is that, that style of programming has a great deal of potential for bugs, which are often very hard tofigure out because they depend on... does this program send this messagebefore or after this one sends that message... and the result was:it took us years to get the thing to work. [ what is linux's relationship to the gnu project? ]
well there's relationships to gnuon kind of multiple levels. one is just the philosophical level of thinking that "making your source open is a good idea". when linus developed the kernelhe wasn't doing it for the gnu project. he did it independently. and he released it independentlyand we didn't know about it. but some of the people who did know about it decided to look for what else they could find to put together with that kernel tomake a whole system.
they looked around, and lo and beholdeverything they needed was already available. "what good fortune!" they thought. but actually there was no chance about it. they had found all the pieces of thegnu system which was missing just the kernel, so when they put all that together really they were fittinglinux into the gap in the gnu system. but they didn't know that. there's a lot of these programs um, done by the free softwares foundation,
and done by other people like linux. and there's a symbiosis betweenlinux and the programs that the programs run on linux and at the same time and they take the advantage of linux as a platform, while linux takes the advantage of the programs by just being able to use them. [ what...what programs? ] umm... the main one is actually the gnu c compiler.
which... without a c compilerit would not have been possible to make linux or most of the open progress available. linux uses the gpl, and i agree with a kind of philosophy behind the gpl. that said the gpl itself is not a very pretty document which is probably just becauseno lawyerese can ever be very pretty. i'd been playing around with linux for actually late '92 or early '93 for about a year. before i decided that it was to the pointwhere actually had everything that i needed
to really replace a sun workstation. and i was looking for a way tohave a unix workstation at home at the time we used sun sparc stationsin the office at stanford. those machines cost us about 7,000 dollars. now i desperately wanted a unix machine at home! there's always a this thoughtyou get as a graduate student "gee... if i could work at home! then i would be so much more productive, i would graduate sooner becausei would finish my thesis sooner"
well, well, is it true? well, you can judge. you know most people end up spending a lot of their time becoming more productive so that they ever actually worked on their thesisthey'd finish it in a day. it takes a while sometimes. so i decided that i wanted a unix machine at home. and i went out there i was able touse linux together with the pc. for about 2,000 dollars, i put together a system. that was one and a half to two times faster
than that 7,000 dollars sun sparc station. it was absolutely amazing. i had one and a half to two times the speed, at a third to fourth the price. light bulbs went off. i knew there was an opportunity here. this was the chance to really do somethingbetter than what sun has done around open source and linux. i called it linux originally as working name.
and that was just because "linus" and the "x" has to be there. it's unix. it's like a law. and...what happened was that... i initially thought that i can't call it linux publicly, because it is just too egotistical. and that was before i had a big ego. right? they thought they were taking a whole bunchof components putting them around linux so they ended up calling the whole thing"a linux system"
and somehow that term caught on. and the result is there are now ten million people using this variant of the gnu system... the gnu/linux operating system. and most of them don't know it. [ some people advocate it be described as gnu/linux. i mean what's your thought onthat? i would say, justify or... ] well, i think it's justified but it is justifiedif you actually make gnu distribution of linux.
the same way that i think that red hat linuxis fine or suse linux, or debian linux. uh, because if you actually make yourown distribution of linux you get to name the thing. but calling linux in general "gnu linux"i think, is just ridiculous. i got involved in fall '93. because i was sent a copy of the first cd-romcommercial linux distribution, which was called yggdrasi produced by adam richter. and i got a copy because... i had been myself writing free software fora long time since the early 80's.
i was actually one of theearly gnu contributors myself. and i was absolutely astonished, i was completely astonished. because i've been a software engineerfor nearly 15 years at that point. and according to all the rules i knew... about controlling complexity,keeping a project group small, having closely managed objectives. linux should have been a disaster, and it wasn't. instead, it was something wonderful,
and i was determined to figure outhow they were getting a way with that. in order for linux to grow beyondthe world of the computer programmer it needed a use, an applicationthat made it a must-have technology that threshold was crossed with the development of a programthat made complex websites possible that program is the apache web server. the killer app of linux was undoubtedlythe apache web server. if you look at the history of linux, the adoption curve of linux andthe adoption curve of the internet
exactly track each other. 1993, which was when the apacheweb server project really got started, was also the beginnings of the popular isp explosion when the internet first became a mass market commodity and the idea of web-based electronic commerce and, and mass communication became real. i think it was one of the first applicationsthat caused people to go "well, if i install linux." i get some tangible benefit from doing so, right?
i mean, clearly there were a lot ofinteresting applications on linux at, at the time, this being maybetwo or three years ago, when the root thing really started to take off but there wasn't a driving, you know, you could almost say business case for someone to use linux versus using nt until, i think, apache anda lot of the things that plugged into apache enhanced apache i mean, when you want to go out and build..go out to build a server farm
it was much more cost effective cost effective real dollar returns to build it on linux and apachethan was to build it on iis and nt even if it meant that you have tospend a little bit of money to train your staff to learn how to use that or to find people who were knowledgeable. but the good news was that that knowledge wasn't very expensive because there were all those college students out there
who'd been using linux for a long timeand were very familiar with it. if you look at the trend curves in web servers apache has steadily been gaininga market share ever since it's up to something like 66% now it's steadily clobberedall of the closed source competition and that's because it's more reliableit's more flexible, it's more extensible it does what webmasters actually need and the combination of apache and linux found its way into a great many commercial shops.
essentially, apache became the applicationthat motivated internet service providers and e-commerce companies to choose linuxover microsoft's windows. it would probably runs best on linux and on freebsd and the reason is thecommunities around those operating systems are also the communities thatcontribute the most back to apache, right? and there were also the operating systemsthat internet service providers started using very heavily as well and internet service providers really liked apache because it allowed them to do
a lot of different thingsthat some of the commercial web servers didn't such as the ability to host more thanone web site on a single box, which clearly if you are an isp and you would have 40,000 users and they all want their web site, is gonna be pretty important to you. one of the key factors in the growth of linux was the creation of companies thatspecialized in the distribution and support of the operating system itself among these companies,red hat software is the best known.
red hat started as a product of marc ewing while he was working at ibm. he wanted a little better linux distribution he started playing around, found out he uh, he spent more timemaintaining his linux distribution than he did uh,than he did working on his new project. so he uh, sort of started the distribution himself he met up with bob young, who at the time was running company calledacc bookstore
which was a mail-order pc unix uh, catalog and bob kind of knew he wanted something, you know, more his own to market, rather than reselling other people's products and he was fairly good at marketing, and... mark knew he needed some marketing help because he was very good at the technical parts,so they kinda got together i started working with red hat in may of 1995,basically right out of nc state. along with eric troan who, me and himcombined make up employees #4 and #5.
we actually reported to work in an apartmentthat mark ewing used to live in we took it over as kind ofthe development part of red hat software and stayed that way till about november of 1995 when... a toilet we had in the apartment kind of exploded, flooded our downstairs neighbor and she got little upset and... the apartment folks found outwe were running a business there instead of actually living there the same time so they decided to throw us out.
so at that point, we had about a week to go find our first office, which we did and get ourselves moved in a hurry we started going in again '95 or soto the venture capital firms, asking, saying, there's something happening here. there's a great business opportunity, to build the next sun for open source. well, the venture capitalists looked at this and said "gee, you are selling systems
the software is free. this is kind of scary. we're not sure that we want to put money in. and... by the way, we funded other systems companies and it hasn't really panned out. we are scared." i came to the us about 3 years ago, and the reason really was that i'd been spending like 6 or 7 years at helsinki university. and decide it was time to see the real worldand not just university life. especially this area had a lot of the mostinteresting work being done.
so i just decided that... let's try to move half way across the world, and give this a try. and it's turned out pretty well. [ ah...you see this as temporary or long term? ] well, we saw it as temporary at first. and i think it's certainly looking likeit's turning into long term. our youngest daughter isboth us and finnish citizen, because she was born here
and the older one is speakingboth swedish and english, so... the next major event was one thati had directed hand in. i wrote a paper, called"the cathedral & the bazaar". which was my observations,my anthropological analysis of what it was that madethe open source world work. we didn't call it that then. we were stillusing the term "free software" primarily. so it was my observation ofwhat made the free software world work and why we were able toproduce extremely high quality software in spite of constantly violating all ofthe standard rules of software engineering
in that paper, i was setting up a contrast between two different styles of development, two opposed styles of development. one, which is theconventional closed development style, which i called the "cathedral" style. in that one, you havetight specification of objectives. small project groups which are runin a fairly hierarchical authoritarian manner. and you have long release intervals on the other hand,what i identifiedis happening in the linux world
was a much more peer to peer decentralized,market or bazaar-like style, which has a very short release intervals and constant solicitation of feedback from peoplewho are formally outside of the project. a very intense peer review process. and the startling thing was that the more i looked at this, the more it seemed that trading awayall the supposed advantages of conventional closed development, for that one single advantageof massive independent peer review actually seemed to win,actually seemed to get you good results.
the reason netscape is important is that they were the first large companyto participate in open source. we had cygnus providing support, but we didn't really have much business. and netscape went open source essentiallyas a way to fight microsoft, which was giving away internet explorer, but not letting anyone else have the source code,not letting companies collaborate. working as part of the sales force, i got a bit of, i got a good idea of.. of why people bought our software
and what it took to make our software successfulin the marketplace against competitive products. however, the problem was, we were seeing, as that, as time went on, our software was uh, being competed against by otherpeople's software, particularly microsoft's and as time went on, the price of our software had to drop because other people were giving their software away at no charge or at little charge. now the real problem was that they feared
microsoft would achieve a monopoly lock on the browser market and they would then use that monopoly lock to pervert actually, the http and html standards that the web depends on. and once they had turned thosestandards in to lock in devices, they could then use that controlto drive netscape out of the server market, which was where it was making its real money. my concern was that as time went on, netscape's business would be threatened,
by the fact that we didn't have enough peopleto do what we needed to do as a company in order to keep our software viable in the marketplace. the netscape release happened in early 1998. and uh, i was told later, i had no idea at the time, that it came about as a direct result of the right people having read"the cathedral and the bazaar". "the cathedral and the bazaar",the paper by eric raymond, was a significant influence onnetscape's decision to release source code. it came as a complete shock to me.
i wasn't really ready for the thought that i was changing the world even by accident. however he was not by any meansthe only influence on that decision. uh, and not necessarily the most important one, when all is said and done. as i said, netscape, netscape had already been talking aboutreleasing source code for quite some time before anyone ever heard of eric's paper. linux congress in early 1997,
which was the first place that i gave that paper. and one of the people who heard it wastim o'reilly of o'reilly and associates. and uh, he thought it was pretty intriguing and he asked me to give it athis first perl conference, which was uh.. later that year, in fall of 97. and apparently what happened, i was told later, although i had no idea that this was happening at the time, uh is that some people from netscape actuallyheard the paper at the perl conference and took those ideas back to netscape andthey kind of lit a fire there.
the role of my paper was essentiallyto make the internal case at netscape, uh to make the business case forwhy netscape should release its source code. the paper was called"netscape source code as netscape product". uh.. a strange title,essentially what the title meant was that in my opinion we needed to think of source code not just as somethingthat was used in creating our products, but as something that was a product in its own right. something that customers might use, other people might use. i then looked at what the business models might be
if we released source code for our products. how would we license them? how do we sell products in this environment? then i looked at the competition, particularly microsoft. what would they be likely to doif we released source code? was there some way they coulduse our source code against us? i used eric's paper as an example ofhow distributed development could work, how a company could develop software notjust using their own people, but also working with people on the internet.
and that's why i included a reference toeric's paper in my paper. once my paper was circulated, the people who read my paper would naturally enough find a reference to eric's paperand read that as well. [ and who was involved inmaking that happen at netscape? ] primarily the person who made theactual decision was jim barksdale. and this turned out to be important later. that our big win, the big score that gave us mainstreamvisibility and credibility with investors
came not because of bottom up evangelismfrom a bunch of engineers, but because one strategist at the top saw the potential power of this method and then essentially imposed that visionon everyone underneath him. when i completed the paper,i first gave a copy to mark andreessen, who was co-founder of netscape andwas at the time one of, on the senior management team at netscape. mark then gave a copy of the paper to severalother people within netscape management, including jim barksdale.
i'm not sure exactly when jim andthe other senior managers made the actual decision, i believe it was in early january sometime. netscape actually announced that it was gonnarelease the source code on jan 22nd, the same time it was going to givecommunicator away for free. when netscape decided to release the source code, people sort of got a wake up notice and said "hey, maybe there is something to this ideaof releasing source code and doing development with peopleoutside your company."
so netscape's decision brought a lot ofpublic attention to the idea of free software, what became known as open source, and brought a lot of attention tothe linux operating system, which was one of the most prominent examples ofopen source software at that time. this is our first office, mountain view, california. we moved here in early 1995. this is 4,000 square feet. it was an incredible leap of faith for us tomove out and take the company to our own office. now what's really important about this place
is that this is the officewhere the term "open source" was invented. if you walk in to an executive's officeand say "free software", ok, if you're lucky, the responseyou'll get is something like, "hmm, hmm, free software, must becheap, shoddy, worthless." uh, and if you're not lucky,it has associations with, uh. with the free software foundation'swholesale attack on intellectual property rights, which regardless of whatyou think about the ethics of that, it's lousy marketing, it's not somethingthat businesses want to hear. so eric raymond knew there was a problem.
we'd been calling this free software, but people took the term "free"and associated with "free of charge", they thought they couldn't make money or couldn't sell, which is exactly the wrong concept. we wanted to get across the idea the softwarewas open and that the source code was available. very important pieces. we had this meeting at the va officesin mountain view, where eric, myself, and christine peterson from foresight institutejoined us as well as some other people. christine peterson was there by phone. uh..
jon "mad dog" hall was also there by phone. uh.. and then todd anderson, who later worked forsuse for a while was there. sam ockman who now runs penguin computing was there. he was uh.. he was an employee of va at the time well, we came up with the concept of open source, we called linus in fact and asked linus if he liked it. he was interested, he liked it. eventually we came up with something that replaced "free software". that was the beginning of open source.
[ how did you chose words "open source"? ] you know, i think christine peterson wasthe person who really came up with the idea. uh, we wanted, again, the ideathat the source code was out there and it was open. there weren't many choices. well, since the fist three recipients havespoken for the open source movement, i think i should speak aboutthe free software movement. the open source movement,focuses on practical advantages that you can get by having a community of users who can cooperate on interchanging and improving software.
i agree completely with the points they make about that. the reason why my views are different,while i am in the free software movement rather than the open source movement, is that i believe there's somethingmore important at stake. that freedom to cooperate with other people,freedom to have a community... is important for our quality of life. it's important for having a good societythat we can live in. and that that is in my view, even more importantthan having powerful and reliable software. but i think some of the people inthe free software camp...
are a little scared by the commercialization. and uh, you know,of course a rebel is put off by success. uh.. i think that commercialization is very important. we want to mainstream this software, and i work with richard stallmanwho's the gray haired man of free software, uh, on a regular basis, and i don't feel i have any philosophical differences. me as author the open source definition and he is originator of free software as an organized thing,
except for one thing.richard thinks that all software should be free, and i think that free software andnon-free software should coexist. that's the only difference we have. uh, we decided early on that what we needed, a..a definition, we needed a kind ofmeta-license to define the term "open source". a, a definition, we needed a kind ofmeta-license to define the term "open source". and what we came up with is a document called"the open source definition". it's derived from the debian free software guidelinesthat were originally written by bruce parens. i'd written the original draft of that, uh,
discussed it for a month with the debian developers debian is a linux distribution and made it their project policy and eric and i decided to relabel what we'd written for debian as the open source definition and to say open source is a software that gives you a list of nine rights which is in the open source definition.
the first right is free redistribution this doesn't mean free as in no price it means liberty um, you have to be free to redistribute your software to someone else and actually no price is a side effect you can charge for that redistribution or not it has to come with source code so that someone can maintain a program.
if they go from a pc to a mac for example they can change the software. derived works have to be possible if someone has to improve your program um, they should be able to distribute the result uh, there is a provision about integrity of the author's source code which says that the author can sort ofmaintain their honor and if you make a change
you might have to change the name of the program or mark out your change very clearly so that your change doesn't reflect on the author there is no discrimination against people or groups. uh, the example i usually use is you can't stop an abortion clinic or an anti-abortion activist from using the software there is no discrimination against fields of endeavor and that means the software has to be usable
in a business as well as in a school the license has to be distributable in other words i have to be able to give that license to someone and that license then should work if that someone gives it to yet a third person the license can't be specific to a product in other words if i distribute my software on a red hat system
the license can't say you can't distribute this on a suse or debian system the license can't contaminate other software so if i distribute this on a cd with another program it can't say that other program must be free otherwise you can't distribute my software and then the only other part of the open source definition is a list of licenses that were accepted
and the ones that we started with were the gpl which was actually the example for a lot of what's in the open source definition in the bsd license because software for bsd system pre-existed linux i think the next moment that i thoughtwas really pivotal was when the database vendors flipped over which happened about three months sooner than i expected to and actually happened in late july early august
commitments to do one ports from oracle and and sybase and the other key database vendors. [ and why was that critical? ] because we knew that in order for the open source story to be credible and especially in order for the linux storyto be credible, we'd have to get commitments fromindependent software vendors to do ports of their applications to these platforms and i was actually kind of worried
i thought that we were in a window of vulnerability between the time that we announcedthe open source campaign and the database vendors flipped over that was the point at which hostile action by microsoft or other close-source software companies that was the point in which a serious marketing blitz might have sunk us but once the big database vendors flipped over that opened the way for other isvs
that started the snow ball effect going every six months or so i would come back to the venture capitalists i would show them the new numbers showing more and more people adopting linux and new people porting, new users and i'd show them our customer list and our customer list was getting much more impressive it was people like cisco that were beginning to appear
people like, you know, those dot-com companies were started to show up on our customer list and eventually the venture capitalists you know, they kept looking at it they kept saying "oh, we can't quite do it" finally, linus appeared on the cover of fortune there was something happening with open source well, at that point, the venture capitalistscouldn't ignore it they just got sick of hearing about linux everywhere
and they got tired of me, just, you know showing it to them every,at that point it was almost every week so they uh, they decided it was time to invest, that there was something happening i announced open source to the world on the internet i did a lot of the early... administrative work of starting the open source initiative and i think six months later i was reading the words open source
in the news all the time and was totally astounded and a year later, i believe microsoft was talking about releasing some source code and someone in the press asked steve ballmer if they were going to open source their code and steve ballmer said, "well, open source means more thanjust releasing the source code" and i realized that he had read my document
and understood it and was now tellingthe press about this now if you are like just a guy on the net who's not doing this for a job at all and you sort of write a manifesto and it spreads out through the world and a year later the vice president of microsoft is talking about that you'd think you were on drugs, wouldn't you? but that's what really happened
the local users groupstend to be more than an issue of building a social network especially getting people familiarized with the issues also just acting as a kind of supporting network for people who, who do not, for example,have the ability to pay for commercial support network so one thing they're doing in this area for example, is they're making these. i think it's once a month
they're having install fests, which means that people who have problems getting linuxinstalled on their machines or have some issue, i mean maybe they've installed linux but want to set up the network in a specific way can actually bring in their machines to this users group meeting and there's a lot of people there willing to help who may have seen the same problem before.
well, actually things aren't so well. i tried it earlier myself. i had problems. and so i came to this install fest where all the gurus abound. hopefully i'll have better luck getting it in. instead of having, uh, sending e-mails, or writing to news groups on the internet and waiting several days for the answers sometimes it's easy to come here and find other people
who might know about your problem and may be able to help you and hopefully within a few hours you have your machine installed originally i wanted to it install onuh, my larger laptop and so i just did a search on the net and found where there were resources to get help and um, i'm here today cause i'm trying to put linux onthis little guy right here.
just a toshiba libretto it's not a easiest thing in the world to do because it's a weird piece of hardware, so... any chairs around here? i think that department of justice case has made people aware of the fact that you should at least look for alternatives to microsoft. and maybe microsoft isn't the american dream after all and that kind of shifting perception
you can very clearly see that people just took microsoft for granted and maybe they're still buying microsoft but at least they're kind of more aware of the issue these days microsoft actually uses linux as defense they used linux to ground a claim that they don't have a monopoly because linux could essentially push them off their catbird seat at any time.
it was a very ingenious argument,totally specious because it didn't do anything to answer the charge that they had previously engaged in bullying and various anti-competitive practices. but it was clever of them and, in an event, the judge didn't buy it while ordinarily we in the linux community arerather worried about letting microsoft become the issue,
but there was a slashdot articleabout december of '98 where a fellow named matt at the noodlehad pointed out that... the, a gentleman in australia has managedto receive a refund for the unused copy of windows thatcame with his computer. so he declared the 19th of january, was it january? err, no, it was february. it was february.oh, i'm sorry, the 19th of february... he declared 19th of februarywindows refund day. and he encouraged everyone to go tothe computer manufacturers, and return their unused copies of windows...
as it was specified in thewindows end-user license agreement. it's important to remember thatin the license itself, it says that you can receive a refund if you don't use the software, and that the manufacture is bound by law to do this... or it was bond by contract. and we found if you called upthese manufactures, they basically said... "stop bothering me kid." and hung up on you. we didn't really wanna sort of giving out our location, or where we were going to meet, until...
you know, the very last seconds. what we did is, we have people meetat the place where we could control in the different towns around here. so i was the san jose marshal, and i believe nick you were... i was..., rick moen and i did san francisco. right and so we had maps there and wehanded them off everybody who was coming. well, we actually met at a denny'sthat's just outside the foster city limits... foster city city limits,
which meant also just outside offoster city police jurisdiction, which meant any, any incidentsthat happened at the meeting point happened in the jurisdiction of san mateo, and if they told us they get lost, we'd say,"fine, we are going to foster city. bye." it's sort of the dukes of hazzard method ofavoiding the cops, so... well actually, originally we marchedon the other side of this building. we marched around and up onto the parkingstructure that's up there, and that's where microsoft had a receptionlaid out for us with drinksand a big sign that said...
"microsoft welcomes the open source community". and the local news cameras got shots oferic raymond and microsoft representative. microsoft's story seems to mostly be that... ...this was not an issue for microsoft,rather from the oems. so we all needed to go back to ourcomputer manufacturers and try yet again to try and get refund from them. we responded to them saying, you know, that we tried that and it's not possible,we need microsoft to take action at this point. and they just repeated the tag lineover and over again...
"you need to go to the oems and manufacturersand get your refunds there." we had about 150 people,probably half of which had signs and such, so.. well, we ended up actually rightin this courtyard here. basically we originally met, gathered outside, various people sent groups in,people from freebsd camp sent a couple of folks in. we had eric raymond and chrisactually tried to go up eventually. they had blocked the elevator off to us. [ where are the offices? ] the offices are right up here on the 9th floor.
we got some really nice press out of it. and we think as a result toshiba made it possible for you to buylaptop without the operating system on it. so, it's a small victory, but... well, even now, companies such as ibm and... a lot of other computer manufacturersare allowing you... now to buy machines that don't havewindows on them. when i was a kid and i went to school, the teachers were trying to teach us to share.
they said if you bring some candy, you can't eat it all yourself, yougotta share with other kids. but now the administration says teachersshould be teaching kids to say yes to licensing. if you bring some software to school,"oh! no, don't share it... ...sharing means you're pirate,sharing means you'll be put in jail." that's not the way society should work.we need the good will, the willingness to help other peopleat least when it's not too hard, because that's the basis of society, that's the fundamental resource,
that gives us a society instead ofa dog-eat-dog jungle. [ so what about people say if that if you have rampant piracy and eliminatethe profit motive and... ...and creative works, software.will not... ] well, they were on both counts. for one thing, people are making a profit fromdeveloping free software, but for another, the freedom tohave a community is more important. [ people that look at, casually look atopen source free software and think... well, because you are supposed toshare and do it for people's good will,
doesn't that seem someone communist.what's your response? ] absolutely nonsense,it makes me really angry when people do that. well, back in, back in 1989,actually communism would have been a compliment. the word people were using at that timewas "crazy", and i want them to use capitalism. communism is an ideology that forces people to share. if you don't share, you get thrown in jail or killed. in 1990, we got a visit from a directorof an institute in uh, moscow university, and actually i saw him in helsinki just 2 weeks ago,
but in any event, he came by, and richard stallman had suggestedthat he visit cygnus, because he was interested in andunderstanding how the free software model... might apply to stimulating entrepreneurialinnovation in russia of all places, and we had been kind of secretiveabout our business plan, because you know, we were'nt reallysure it's gonna work, we didn't want to look too stupid if it failed. but i was very very open with him. the more i told him,the more he started to shake his head like this,
and i finally said, you know, "what's wrong?" and he said, "this sounds to much like communismto be successful in russia." you got to go a gulag and end up in a mass gravewith a bullet in the back of your head. open source is not communismbecause it does'nt force people. carl marx did not invent helping your neighbor. it's not communist to have a commons, a commons existed long before communism asa philosophy of government. there are many commons in our lives, for example, we drive on the highway,
something that is maintained for our common good. actually labeling our business model... means that it misses the point a little bit. whether it's communist, or whether it's capitalist, the label doesn't matter, the real question is, how much value can you deliver,how scalable is the business, what kind of problems,what kind of of rate of innovation can you sustain. and then, however you want to label that,it's really up to you. the revolution goes prime time.
a lot of people described, that august linuxworld as"linux's coming-out party", linus torvalds was very funny about this,he said, "what? was linux gay?" but some people said"yeah, that was our debutante ball." that was when the... linux gods, and the hardware hackersreally got it together with the suits. at 3pm on august 10th 1999, linus torvalds delivered the keynoteaddress at linuxworld. the crowds of 6000 people began lining up at 12 noon.
ladies and gentlemen,please welcome larry augustin... linuxworld conference chair and president andceo of va linux systems. these guys have to clap, i pay them. thank you all for being here,looks like it's been a great show so far. if you'll indulge me for a moment, i'm going totry to avoid the glare of the lights. i still think there's lots of people,even though this is the 2nd show, i still think there's lots of peoplewho don't quite get... ...what it is that's so exciting about linux. so there is a great show goingon next door,
there's huge exhibits and everything, but it's the people out herethat are real contributors, not those companies. the person on next i know you all know, so i don't have to give anything in the wayas introduction. ladies and gentlemen, i give you, linus torvalds. linus thanks. calm down. say "oooo".
yes. i don't want to just give oneof my normal talks, because i find them boring, probably by now most of you find them boringtoo because you've heard them like 10 times. but after the technical updates,we will actually try... ...whether we can do a question andanswer session with 5,000 people, or how many of you there are there. and it may not actually work out, because one of the 5,000 people is really loud.
the one thing i will do, which i always do in all my talks isthe gratitude thing, i want to kind of acknowledge the fact that... ...i'm obviously not been alone in doing linux. rhat ipo: on august 11, 1999 red hat software becamethe first linux company to go public. red hat, up 228% this is the ipo that everybody was waiting for. they of course are behindthe linux operating software.
r-h-a-t, i know all i've gotten today, are comments aboutwhat the stock price is, all morning, you know it wasat 41, it was at 42, it was at 47, it was at 53, it was at 51... every machine as far as i can tellon the show floor is pointed to their e-trade accounts or theirbroker accounts, they know the red hat's price i can't believe this.47 i just heard 53.
oh, boy. hang on, i didn't buy it. you didn't buy? no, no, i didn't buy. i should've bought, but uh... no no, that's great. if it's... if it's.... you guys don't know? well, you know
red hat being successful as just means that it legitimizes linuxso it's much easier for us to go out rob malda (cmdrtaco) ...on the red hat ipo it's kind of been a little bit divided. you've got a lot of people that are pretty hardcore and and they're kind of offended by that you know, cause they work really hard,they are not really getting... maybe they fair share out of that.
some people do get ticked. and you know the thing that you see that on a lot of mailing listsor on slashdot you'll read, you know this guy is really mad because he didn'tget chance to, he's didn't get a chance to do... to get stock from red hat he didn't get a chance to get...to get a job from this other company, you know. but the, the kind of the shocking secretthere is that, most the really hardcore guys,you know, they don't care so much. the guys that are kind of really down in trenches.
they're writing this code becausethey need this code. if we could invite richard stallman who's the founder ofthe free software association and tim ney, who's the managing director. there we go. hahahahaha ah, here it is. now, richard, i saw you playing your recorderat, in paris at that linux conference but i didn't have audio tracks. so would you them to add audio to their, uh...
video downstream next time uh..i don't have any control over that unfortunately those things canonly be done with non-free software. we'll give you the award, and before you say a word, we'll have a tim and yourself hold upa little representation of the contribution towards the free software association. so, very ironic things have happened, but nothing to match this giving the linus tovarlds awardto the free software foundation
is sort of like giving thehan solo award to the rebel fleet. you see, some of you may not realize how farthat analogy goes. but actually let me tell youhow this, how we got here. see what happened is, 15 years ago, if you wanted to use a computer, the only way you could do itwas to, was with proprietary software, software that divides and subjugates the users. and most people just... a lot of people didn't like it.
but they saw no alternative. but some of us were determinedto make an alternative. and we said we're gonna developa free operating system, a free software operating system that will give users the chance to have freedomwhile they use their computers. now a lot of people said, "well, it's a nice idea but it's so hard, you'll never get it done, so i don't wanna participate,i don't believe you can ever get it done." but luckily not everybody said that.
and clearly, we knewwe would eventually get the kernel done. but as it happens. somebody else did a better kernel before we did. now in the old days, we had an overall strategy from calling people's attentionto the importance of freedom to the freedom they can have, or not have,when they use a computer. well what can we do about it?as far as i can tell, the only workable way of trying tochange this make that strategy work again is to, spread the word that the operating systemyou're using is actually the gnu system.
somewhat modified of course. and when people know this, they'll take a look at the reasonswe developed this system. they'll think about these issues. and some of them will decide they agree. so i ask people,please tell the people this is the gnu system. it's the combination of gnu and linuxso we can call it gnu/linux [ so larry, when you were at stanford8, 9 years ago during your ph.d did you ever think you'd be in this position? ]
no [ all kidding aside. ] no i had no idea honestly [ what did you think you would bewhen you were finished up with your ph.d? ] you know that's a good question, i really didn't have a good idea, i mean. here we are, on this huge show floor. there are people just going crazy about linux. we had 6,200 people crammed into a room
to see the linux, linus speak last night. here we are, with, you know, all of these huge venders all over the show. i..it's, just, you have no idea thatthis is gonna happen. i mean this is just this little operating systemthat we're happy with. that a few people cared about, you know, i thought i'd have a nice littleconsulting business. and here i am suddenly, with all of this huge show going on.
it's just incredible. i mean, a year ago, you could look and say,you know, this is gonna be big and everyone standing at the show going "you know the show was big last year, is it gonna, is it gonna be as big as this year?" the you remind them "you know, last year was only 6 month ago." and then they go "oh, , linux time". so leading up to the ipo, uh...
we had arrived actually in san diegoon tuesday night we spent wednesday morning meeting investorsin san diego. we flew up to san francisco spent wednesday afternoon meetinginvestment firms in san francisco va linux selecteddecember 9, 1999, as the dayfor its stock to begin trading. then on the thursday morning of the ipo is when our stock would be traded publically. so it was nice we had endedthe tour in san francisco because we could go to the credit suite'strading desk the next morning
to watch the public offering. and in san francisco being closeenough to the company, and to our families, we could invite people up toactually join us in the first trade. so i invited my wife and we invited linus and tove, and a number of other friends and peoplewho worked in the company in to join us. whenever we invite linus and tove, they have uh, two young children, and i have a daughter, andrea. and we always bring the kids along
so we went in to the credit suite's trading floor with all these traders and there are these 3 year old kids running around and chasing each otheraround the show floor, around the trading floor. so linus and i walked in andwe walked up in to the trading floor and everyone was very excited. and we kept asking them, well"hows it going? are things going ok?". and they said, "oh, it's.. uh, we're really excited, i think things are going well.
we don't wanna, we don't wanna say,we don't wanna jinx anything." we walked in and it was a big screen tv showing cnbc. and it was amazing to us,but the theme for the day was linux. now we have an ipo that's gonna go today. and when i mean go, it is going to go. the estimates i'm hearing are staggering. but watch va linux systems. it goes at 12:40 today. the symbol is l-n-u-x. a provider of large scale computerservers and workstations,
specially designed for the linux operating system. the original range on this ipo was11 to 13 dollars, then 21 to 23, then 28 to 30. priced at 30, and the estimates i'm hearingi don't wanna repeat because i don't have a confirmation. but if they're true, they will blow you're mindwhen this stock takes off at 12:40. i turned to linus and i said "gee, did you ever think, you know, you'd walk in here some day andlinux would be the theme on cnbc?". and linus said in his joking way said "oh absolutely!". so we walk in and they show usthe buy and sell orders coming in..
and it's incredible. we're seeing numbers like320 dollars, 340 dollars a share. and i'm just in complete shock. you know, this is over 10 timeswhere we priced the offering. it was incredible. and i remember linus just kind of,sort of patting me on the back and saying, you know "relax..." and it waspretty exciting to see that. we were, it was just amazing. we were stunned. we were lucky that we were able toget back to the offices, we'd been in san franciscoso we could come back to va's offices
to, to see everyone in the office for the ipo. when we got back, we had uh,everyone was obviously very excited. the ipo had done just tremendously well. we had a little party that we put together. it was interesting, while we were celebrating there were plenty of people that were still trying to work. i recall cries of "be quiet!","we're on the phone!", "we're working!", uh, as we uh, as we went in to the offices. one of the things i did wasi gave the road show presentation
for the employees back at the office, so they could have an idea ofwhat we'd been telling investors, and understand exactly what we'd uh,put together for them. but again the story of the day is va linux,now up 766% to 235 dollars to 265. sue, the best performing ipo ever. here it goes, sycamore networks was uh,priced at 38 dollars, surged to $270. this has just beat it. and by the way.. [ how do you feel about potentially billionsof dollars of wealth being created from your creation,that you're not directly cashing out? ]
so, if i hadn't made linux available, i mean, i wouldn't have gotten any money that way either. so i mean, it's a win-win situation. uh, just the fact that there area lot of commercial companies means that there are a lot of linux peoplewho used to work on linux kind of on the side. and now they get paid fordoing what they wanted to do. and that helps me in the sense thati wanted them to work on linux anyway. the whole gnu project is really one big hack. it's one big act of subversive playful cleverness,
to change society for the better, becausei'm only interested in changing it for the better, but in a clever way. hi, we're the gnu/stallmans, and this is"the free software song" join us now and share the software; you'll be free, hackers, you'll be free. hoarders may get piles of money, that is true, hackers, that is true. but they cannot help their neighbors; that's ain't good, hackers, that's ain't good.
when we have enough free software at our call, hackers, at our call, we'll throw out those dirty licenses ever more, hackers, ever more. oh, you'll be free i let you know. that you'll be free. [ repeat ] i let you know
Tidak ada komentar:
Posting Komentar