Some Notes on the "Who wrote Linux" Kerfuffle, Release 1.5
The history of UNIX and its various children and grandchildren has been in the news recently as a result of a book from the Alexis de Tocqueville Institution. Since I was involved in part of this history, I feel I have an obligation to set the record straight and correct some extremely serious errors. But first some background information.
Ken Brown, President of the Alexis de Tocqueville Institution, contacted me in early March. He said he was writing a book on the history of UNIX and would like to interview me. Since I have written 15 books and have been involved in the history of UNIX in several ways, I said I was willing to help out. I have been interviewed by many people for many reasons over the years, and have been on Dutch and US TV and radio and in various newspapers and magazines, so I didn't think too much about it.
Brown flew over to Amsterdam to interview me on 23 March 2004. Apparently I was the only reason for his coming to
Europe. The interview got off to a shaky start, roughly paraphrased as follows:
AST: "What's the Alexis de Tocqueville Institution?"
KB: We do public policy work
AST: A think tank, like the Rand Corporation?
KB: Sort of
AST: What does it do?
KB: Issue reports and books
AST: Who funds it?
KB: We have multiple funding sources
AST: Is SCO one of them? Is this about the SCO lawsuit?
KB: We have multiple funding sources
AST: Is Microsoft one of them?
KB: We have multiple funding sources
He was extremely evasive about why he was there and who was funding him. He just kept saying he was just writing a book about the history of UNIX. I asked him what he thought of Peter Salus' book, A Quarter Century of UNIX. He'd never heard of it! I mean, if you are writing a book on the history of UNIX and flying 3000 miles to interview some guy about the subject, wouldn't it make sense to at least go to amazon.com and type "history unix" in the search box, in which case Salus' book is the first hit? For $28 (and free shipping if you play your cards right) you could learn an awful lot about the material and not get any jet lag. As I soon learned, Brown is not the sharpest knife in the drawer, but I was already suspicious. As a long-time author, I know it makes sense to at least be aware of what the competition is. He didn't bother.
UNIX and Me
I didn't think it odd that Brown would want to interview me about the history of UNIX. There are worse people to ask. In the late 1970s and early 1980s, I spent several summers in the UNIX group (Dept. 1127) at Bell Labs. I knew Ken Thompson, Dennis Ritchie, and the rest of the people involved in the development of UNIX. I have stayed at Rob Pike's house and Al Aho's house for extended periods of time. Dennis Ritchie, Steve Johnson, and Peter Weinberger, among others have stayed at my house in Amsterdam. Three of my Ph.D. students have worked in the UNIX group at Bell Labs and one of them is a permanent staff member now.
Oddly enough, when I was at Bell Labs, my interest was not operating systems, although I had written one and published a paper about it (see "Software - Practice & Experience," vol. 2, pp. 109-119, 1973). My interest then was compilers, since I was the chief designer of the Amsterdam Compiler Kit (see Commun. of the ACM, vol. 26, pp. 654-660, Sept. 1983.). I spent some time there discussing compilers with Steve Johnson, networking with Greg Chesson, writing tools with Lorinda Cherry, and book authoring with Brian Kernighan, among many others. I also became friends with the other "foreigner," there, Bjarne Stroustrup, who would later go on to design and implement C++.
In short, although I had nothing to do with the development of the original UNIX, I knew all the people involved and much of the history quite well. Furthermore, my contact with the UNIX group at Bell Labs was not a secret; I even thanked them all for having me as a summer visitor in the preface to the first edition of my book Computer Networks. Amazingly, Brown knew nothing about any of this. He didn't do his homework before embarking on his little project
MINIX and Me
Years later, I was teaching a course on operating systems and using John Lions' book on UNIX Version 6. When AT&T decided to forbid the teaching of the UNIX internals, I decided to write my own version of UNIX, free of all AT&T code and restrictions, so I could teach from it. My inspiration was not my time at Bell Labs, although the knowledge that one person could write a UNIX-like operating system (Ken Thompson wrote UNICS on a PDP-7) told me it could be done. My real inspiration was an off-hand remark by Butler Lampson in an operating systems course I took from him when I was a Ph.D. student at Berkeley. Lampson had just finished describing the pioneering CTSS operating system and said, in his inimitable way: "Is there anybody here who couldn't write CTSS in a month?" Nobody raised his hand. I concluded that you'd have to be real dumb not to be able to write an operating system in a month. The paper cited above is an operating system I wrote at Berkeley with the help of Bill Benson. It took a lot more than a month, but I am not as smart as Butler. Nobody is.
I set out to write a minimal UNIX clone, MINIX, and did it alone. The code was 100% free of AT&T's intellectual property. The full source code was published in 1987 as the appendix to a book, Operating Systems: Design and Implementation, which later went into a second edition co-authored with Al Woodhull. MINIX 2.0 was even POSIX-conformant. Both editions contained hundreds of pages of text describing the code in great detail. A box of 10 floppy disks containing all the binaries and source code was available separately from Prentice Hall for $69.
While this was not free software in the sense of "free beer" it was free software in the sense of "free speech" since all the source code was available for only slightly more than the manufacturing cost. But even "free speech" is not completely "free"--think about slander, yelling "fire" in a crowded theater, etc. Also Remember (if you are old enough) that by 1987, a university educational license for UNIX cost $300, a commercial license for a university cost $28,000, and a commercial license for a company cost a lot more. For the first time, MINIX brought the cost of "UNIX-like" source code down to something a student could afford. Prentice Hall wasn't really interested in selling software. They were interested in selling books, so there was a fairly liberal policy on copying MINIX, but if a company wanted to sell it to make big bucks, PH wanted a royalty. Hence the PH lawyers equipped MINIX with a lot of boilerplate, but there was never any intention of really enforcing this against universities or students. Using the Internet for distributing that much code was not feasible in 1987, even for people with a high-speed (i.e., 1200 bps) modem. When distribution via the Internet became feasible, I convinced Prentice Hall to drop its (extremely modest) commercial ambitions and they gave me permission to put the source on my website for free downloading, where it still is.
Within a couple of months of its release, MINIX became something of a cult item, with its own USENET newsgroup, comp.os.minix, with 40,000 subscribers. Many people added new utility programs and improved the kernel in numerous ways, but the original kernel was just the work of one person--me. Many people started pestering me about improving it. In addition to the many messages in the USENET newsgroup, I was getting 200 e-mails a day (at a time when only the chosen few had e-mail at all) saying things like: "I need pseudoterminals and I need them by Friday." My answer was generally quick and to the point: "No."
The reason for my frequent "no" was that everyone was trying to turn MINIX into a production-quality UNIX system and I didn't want it to get so complicated that it would become useless for my purpose, namely, teaching it to students. I also expected that the niche for a free production-quality UNIX system would be filled by either GNU or Berkeley UNIX shortly, so I wasn't really aiming at that. As it turned out, the GNU OS sort of went nowhere (although many UNIX utilities were written) and Berkeley UNIX got tied up in a lawsuit when its designers formed a company, BSDI, to sell it and they chose 1-800-ITS UNIX as their phone number. AT&T felt this constituted copyright infringement and sued them. It took a couple of years for this to get resolved. This delay in getting free BSD out there gave Linux the breathing space it needed to catch on. If it hadn't been for the lawsuit, undoubtedly BSD would have filled the niche for a powerful, free UNIX clone as it was already a stable, mature system with a large following.
Ken Brown and MeNow Ken Brown shows up and begins asking questions. I quickly determined that he didn't know a thing about the history of UNIX, had never heard of the Salus book, and knew nothing about BSD and the AT&T lawsuit. I started to tell him the history, but he stopped me and said he was more interested in the legal aspects. I said: "Oh you mean about Dennis Ritchie's patent number 4135240 on the setuid bit?" Then I added:"That's not a problem. Bell Labs dedicated the patent." That's when I discovered that (1) he had never heard of the patent, (2) did not know what it meant to dedicate a patent (i.e., put it in the public domain), and (3) really did not know a thing about intellectual property law. He was confused about patents, copyrights, and trademarks. Gratuitously, I asked if he was a lawyer, but it was obvious he was not and he admitted it. At this point I was still thinking he might be a spy from SCO, but if he was, SCO was not getting its money's worth.
He wanted to go on about the ownership issue, but he was also trying to avoid telling me what his real purpose was, so he didn't phrase his questions very well. Finally he asked me if I thought Linus wrote Linux. I said that to the best of my knowledge, Linus wrote the whole kernel himself, but after it was released, other people began improving the kernel, which was very primitive initially, and adding new software to the system--essentially the same development model as MINIX. Then he began to focus on this, with questions like: "Didn't he steal pieces of MINIX without permission." I told him that MINIX had clearly had a huge influence on Linux in many ways, from the layout of the file system to the names in the source tree, but I didn't think Linus had used any of my code. Linus also used MINIX as his development platform initially, but there was nothing wrong with that. He asked if I objected to that and I said no, I didn't, people were free to use it as they wished for noncommercial purposes. Later MINIX was released under the Berkeley license, which freed it up for all purposes. It is still in surprisingly wide use, both for education and in the Third World, where millions of people are happy as a clam to have an old castoff 1-MB 386, on which MINIX runs just fine. The MINIX home page cited above still gets more than 1000 hits a week.
Finally, Brown began to focus sharply. He kept asking, in different forms, how one person could write an operating system all by himself. He simply didn't believe that was possible. So I had to give him more history, sigh. To start with, Ken Thompson wrote UNICS for the PDP-7 all by himself. When it was later moved to the PDP-11 and rewritten in C, Dennis Ritchie joined the team, but primarily focused on designing the C language, writing the C compiler, and writing the I/O system and device drivers. Ken wrote nearly all of the kernel himself.
In 1983, a now-defunct company named the Mark Williams company produced and sold a very good UNIX clone called Coherent. Most of the work was done by three ex-students from the University of Waterloo: Dave Conroy, Randall Howard, and Johann George. It took them two years. But they produced not only the kernel, but the C compiler, shell, and ALL the UNIX utilities. This is far more work than just making a kernel. It is likely that the kernel took less than a man-year.
In 1983, Ric Holt published a book, now out of print, on the TUNIS system, a UNIX-like system. This was certainly a rewrite since TUNIS was written in a completely new language, concurrent Euclid.
Then Doug Comer wrote XINU. While also not a UNIX clone, it was a comparable system.
In addition, Gary Kildall wrote CP/M by himself and Tim Paterson wrote MS-DOS. While these systems from the early 1980s were not even close to being UNIX-clones, they were substantial and popular operating systems written by individuals.
By the time Linus started, five people or small teams had independently implemented the UNIX kernel or something approximating it, namely, Thompson, Coherent, Holt, Comer, and me. All of this was perfectly legal and nobody stole anything. Given this history, it is pretty hard to make a case that one person can't implement a system of the complexity of Linux, whose original size was about the same as V1.0 of MINIX.
Of course it is always true in science that people build upon the work of their predecessors. Even Ken Thompson wasn't the first. Before writing UNIX, Ken had worked on the MIT MULTICS (MULTiplexed Information and Computing Service) system. In fact, the original name of UNIX was UNICS, a joke made by Brian Kernighan standing for the UNIplexed Information and Computing Service, since the PDP-7 version could support only one user--Ken. After too many bad puns about EUNUCHS being a castrated MULTICS, the name was changed to UNIX. But even MULTICS wasn't first. Before it was the above-mentioned CTSS, designed by the same team at MIT.
Thus, of course, Linus didn't sit down in a vacuum and suddenly type in the Linux source code. He had my book, was running MINIX, and undoubtedly knew the history (since it is in my book). But the code was his. The proof of this is that he messed the design up. MINIX is a nice, modular microkernel system, with the memory manager and file system running as user-space processes. This makes the system cleaner and more reliable than a big monolithic kernel and easier to debug and maintain, at a small price in performance, although even on a 4.77 MHz 8088 it booted in maybe 5 seconds (vs. a minute for Windows on hardware 500 times faster). An example of commercially successful microkernel is QNX. Instead of writing a new file system and a new memory manager, which would have been easy, Linus rewrote the whole thing as a big monolithic kernel, complete with inline assembly code :-( . The first version of Linux was like a time machine. It went back to a system worse than what he already had on his desk. Of course, he was just a kid and didn't know better (although if he had paid better attention in class he should have), but producing a system that was fundamentally different from the base he started with seems pretty good proof that it was a redesign. I don't think he could have copied UNIX because he didn't have access to the UNIX source code, except maybe John Lions' book, which is about an earlier version of UNIX that does not resemble Linux so much.
My conclusion is that Ken Brown doesn't have a clue what he is talking about. I also have grave questions about his methodology. After he talked to me, he prowled the university halls buttonholing random students and asking them questions. Not exactly primary sources.
The six people I know of who (re)wrote UNIX all did it independently and nobody stole anything from anyone. Brown's remark that people have tried and failed for 30 years to build UNIX-like systems is patent nonsense. Six different people did it independently of one another. In science it is considered important to credit people for their ideas, and I think Linus has done this far less than he should have. Ken and Dennis are the real heros here. But Linus' sloppiness about attribution is no reason to assert that Linus didn't write Linux. He didn't write CTSS and he didn't write MULTICS and didn't write UNIX and he didn't write MINIX, but he did write Linux. I think Brown owes a number of us an apology.
Linus and Me
Some of you may find it odd that I am defending Linus here. After all, he and I had a fairly public "debate" some years back. My primary concern here is trying to get the truth out and not blame everything on some teenage girl from the back hills of West Virginia. Also, Linus and I are not "enemies" or anything like that. I met him once and he seemed like a nice friendly, smart guy. My only regret is that he didn't develop Linux based on the microkernel technology of MINIX. With all the security problems Windows has now, it is increasingly obvious to everyone that tiny microkernels, like that of MINIX, are a better base for operating systems than huge monolithic systems. Linux has been the victim of fewer attacks than Windows because (1) it actually is more secure, but also (2) most attackers think hitting Windows offers a bigger bang for the buck so Windows simply gets attacked more. As I did 20 years ago, I still fervently believe that the only way to make software secure, reliable, and fast is to make it small. Fight Features.
If you have made it this far, thank you for your time. Permission is hereby granted to mirror this web page provided that the original, unmodified version is used.
Andy Tanenbaum, 20 May 2004
Ken Brown's Motivation, Release 1.2
On 20 May 2004, I posted a statement refuting the claim of Ken Brown, President of the Alexis de Tocqueville Institution, that Linus Torvalds didn't write Linux. My statement was mentioned on Slashdot, Groklaw, and many other Internet news sites. This attention resulted in over 150,000 requests to our server in less than a day, which is still standing despite yesterday being a national holiday with no one there to stand next to it saying "You can do it. You can do it." Kudos to Sun Microsystems and the folks who built Apache. My statement was mirrored all over the Internet, so the number of true hits to it is probably a substantial multiple of that. There were also quite a few comments at Slashdot, Groklaw, and other sites, many of them about me. I had never engaged in remote multishrink psychoanalysis on this scale before, so it was a fascinating experience.
The Brown Book
I got an advance copy of Ken Brown's book. I think it is still under embargo, so I won't comment on it. Although I am not an investigative reporter, even I know it is unethical to discuss publications still under embargo. Some of us take ethics more seriously than others. So I won't even reveal the title. Let's call it The Brown Book. There is some precedent for nicknaming books after colors: The International Standard for the audio CD (IS 10149) is usually called The Red Book. The CD-ROM was described in the Yellow Book.
Suffice it to say, there is a great deal to criticize in the book. I am sure that will happen when it is published. I may even help out.
What prompted me to write this note today is an email I got yesterday. Actually, I got quite a few :-) , most of them thanking me for the historical material. One of yesterday's emails was from Linus, in response to an email from me apologizing for not letting him see my statement in advance. As a matter of courtesy, I did try but I was using his old transmeta.com address and didn't know his new one until I got a very kind email from Linus' father, a Finnish journalist.
In his email, Linus said that Brown never contacted him. No email, no phone call, no personal interview. Nothing. Considering the fact that Brown was writing an explosive book in which he accused Linus of not being the author of Linux, you would think a serious author would at least confront the subject with the accusation and give him a chance to respond. What kind of a reporter talks to people on the periphery of the subject but fails to talk to the main player?
Why did Brown fly all the way to Europe to interview me and (and according to an email I got from his seat-mate on the plane) one other person in Scandinavia, at considerable expense, and not at least call Linus? Even if he made a really bad choice of phone company, how much could that cost? Maybe a dollar? I call the U.S. all the time from Amsterdam. It is less than 5 cents a minute. How much could it cost to call California from D.C.?
From reading all the comments posted yesterday, I am now beginning to get the picture. Apparently a lot of people (still) think that I 'hate' Linus for stealing all my glory (see below for more on this). I didn't realize this view was so widespread. I now suspect that Brown believed this, too, and thought that I would be happy to dump all over Linus to get 'revenge.' By flying to Amsterdam he thought he could dig up dirt on Linus and get me to speak evil of him. He thought I would back up his crazy claim that Linus stole Linux from me. Brown was wrong on two counts. First, I bear no 'grudge' against Linus at all. He wrote Linux himself and deserves the credit. Second, I am really not a mean person. Even if I were still angry with him after all these years, I wouldn't choose some sleazy author with a hidden agenda as my vehicle. My home page gets 2500 hits a week. If I had something to say, I could put it there.
When The Brown Book comes out, there will no doubt be a lot of publicity in the mainstream media. Any of you with contacts in the media are actively encouraged to point reporters to this page and my original statement to provide some balance. I really think Brown's motivation should come under scrutiny. I don't believe for a nanosecond that Brown was trying to do a legitimate study of IP and open source or anything like that. I think he was trying to make the case the people funding him (which he refused to disclose to me despite my asking point blank) wanted to have made. Having an institution with an illustrious-sounding name make the case looks better than having an interested party make the case.
Clearing Up Some Misconceptions
I would like to close by clearing up a few misconceptions and also correcting a couple of errors. First, I REALLY am not angry with Linus. HONEST. He's not angry with me either. I am not some kind of "sore loser" who feels he has been eclipsed by Linus. MINIX was only a kind of fun hobby for me. I am a professor. I teach and do research and write books and go to conferences and do things professors do. I like my job and my students and my university. If you want to get a masters there, see my home page for information. I wrote MINIX because I wanted my students to have hands-on experience playing with an operating system. After AT&T forbade teaching from John Lions book, I decided to write a UNIX-like system for my students to play with. Since I had already written two books at this point, one on computer architecture and one on computer networks, it seemed reasonable to describe the system in a new book on operating systems, which is what I did. I was not trying to replace GNU/HURD or Berkeley UNIX. Heaven knows, I have said this enough times. I just wanted to show my students and other students how you could write a UNIX-like system using modern technology. A lot of other people wanted a free production UNIX with lots of bells and whistles and wanted to convert MINIX into that. I was dragged along in the maelstrom for a while, but when Linux came along, I was actually relieved that I could go back to professoring. I never really applied for the position of King of the Hackers and didn't want the job when it was offered. Linus seems to be doing excellent work and I wish him much success in the future.
While writing MINIX was fun, I don't really regard it as the most important thing I have ever done. It was more of a distraction than anything else. The most important thing I have done is produce a number of incredibly good students, especially Ph.D. students. See my home page for the list. They have done great things. I am as proud as a mother hen. To the extent that Linus can be counted as my student, I'm proud of him, too. Professors like it when their students go on to greater glory. I have also written over 100 published research papers and 14 books which have been translated into about 20 languages. As a result I have become a Fellow of the IEEE, a Fellow of the ACM, and won numerous other awards. For me, these are the things that really count. If MINIX had become a big 'commercial' success I wouldn't have had the time to do all this academic stuff that I am actually more interested in.
I can't resist saying a few words about microkernels. A microkernel is a very small kernel. If the file system runs inside the kernel, it is NOT a microkernel. The microkernel should handle low-level process management, scheduling, interprocess communication, interrupt handling, and the basics of memory management and little else. The core microkernel of MINIX 1.0 was under 1400 lines of C and assembler. To that you have to add the headers and device drivers, but the totality of everything that ran in kernel mode was under 5000 lines. Microsoft claimed that Windows NT 3.51 was a microkernel. It wasn't. It wasn't even close. Even they dropped the claim with NT 4.0. Some microkernels have been quite successful, such as QNX and L4. I can't for the life of me see why people object to the 20% performance hit a microkernel might give you when they program in languages like Java and Perl where you often get a factor 20x performance hit. What's the big deal about turning a 3.0 GHz PC into a 2.4 GHz PC due to a microkernel? Surely you once bought a machine appreciably slower than 2.4 GHz and were very happy with it. I would easily give up 20% in performance for a system that was robust, reliable, and wasn't susceptible to many of the ills we see in today's massive operating systems.
I would now like to correct an error in my original statement. One of the emails I got yesterday clarified the origins of Coherent. It was not written by Bob Swartz. He was CEO of the Mark Williams Company. Three ex-students from the University of Waterloo, Dave Conroy, Randall Howard, and Johann George, did most of the work. Waterloo is in Canada, where they also play baseball I am told, but only after the ice melts and they can't play hockey. It took the Waterloo students something like 6 man-years to produce Coherent, but this included the kernel, the C compiler, the shell, and ALL the utilities. The kernel is only a tiny fraction of the total code, so it may well be that the kernel itself took a man year. It took me three years to write MINIX, but I was only working at it only in the evenings, and I also wrote 400 pages of text describing the code in that time period (also in the evenings). I think a good programmer can write a 12,000 line kernel in a year.
If you have made it this far, thank you for your time. Permission is hereby granted to mirror this web page provided that the original, unmodified version is used.
Andy Tanenbaum, 21 May 2004