 Good afternoon, thank you all for coming I'd like to introduce you to a talk that I'm personally very excited to see for Mr. John Maddog Hall enjoy a little bit high perhaps we'll tune that down I know but they said to me that they will take care of that so I trust in them okay hello everybody it's nice to see you all here my name is John Maddog Hall or as they say in China Ding Gao and I am the board chair of the Linux Professional Institute we do certification of open-source professionals and we try and move forward the open source model we've been doing that for 20 years as of last year so this is our 21st year we have 150,000 certified professionals in over 180 countries around the world my talk is called Fossage and Fostem back and forth from the brink of disaster and I think as I go along I'll explain a little bit about that title and why it's important first of all for the people that don't know me yes I am old I've had three heart attacks I only work with about 50% of my heart capacity the rest is dead but that's okay because I am now the old man moving very slowly in front of you the person I used to get mad at you know get out of where you old man I am now him if you see me going up and downstairs please don't hurry me because that's as fast as I go but I've had a lot of different jobs in the last 50 years that I've been in computers and one of the things I'm most proud of is that I am pragmatic a couple of talks ago we had some people stand up to say that it is wrong to belittle people who are using proprietary software and I am happy to say I have never done that to my knowledge I recommend to people that they should use free software in fact I specify free software instead of open source and you'll see a little bit about that later thank you but I'll give you a warning that this is a highly opinionated talk and if you wish to disagree with me we can do it later over beer so why the brink of disaster I have seen over the last 50 years things that are going along and seemingly at the very last minute and typically from some very stubborn people we have come back from disaster and that's what a little bit of this talk is about now originally this talk was supposed to be about the last 20 years and specifically about what Faustum had contributed to the free and open source marketplace and I was very happy to do that and then I received this email from Steven Goodwin who is very nice guy whose talk is following mine and he says hi I think we're talking about the same thing so then I met with Steve we worked back and forth and we came to a conclusion of what I could talk about and what he could talk about and I hardly recommend if you stay for his talk and and listen to that too and the second thing that happened was reality because if I started to talk about all the things that Faustum had done to the level that I wanted to talk about and we would be here until about midnight and we wouldn't have time for beer so that's the reality and talking about just 20 years of stuff that's like you know trying to leap as fast as far as you can without taking a run you know it's a standing broad jump as opposed to a running broad jump I'm not going to demonstrate this because I probably hurt myself but that's what that's the way I feel about it so we're going to take a look in the past with the knowledge of the presence and we're going to do what we should be doing a lot of times but sometimes we're just too busy and I'm going to show you how we have to be very careful in the future because we may not come back from the brink of disaster so let's go back into the long past I'll try and make this brief because once upon a time copyright and patents did not apply to software when people put out software there was no license to protect it we did this under contract law we would talk to our customers and we would meet with their lawyers and our lawyers who meet with their lawyers and we would have a contract drawn up that said where you could put the software and how many machines you put the software how many people could use it so forth and so on and a lot of times the software was distributed in source code form because it wasn't enough of one type of computer to justify making it into a binary distribution and the other thing that happened back in those days is when you bought say an IBM mainframe well it came with the operating system and there was no real contract to said that it just did because you couldn't use the computer really without the operating system and everybody seemed to be happy with that until one day a gentleman by the name of George and all created a computer that was completely compatible with the IBM 360 and nobody ever believed that that could happen and people could take their IBM operating system and put her on the end all computer and it was perfectly happy except for IBM they were not happy and actually the end all is a great system because the IBM systems were still water cooled and so you needed to put pipes and water and stuff and the M doll was air cooled you didn't need all of that the IBM had lots of lights and switches and stuff like that on it to control it and the M doll just had this very large CRT screen that if you toggle the switches right it would actually put a picture of the lights up on the screen and it was really cool and so and it was cheaper too and so everybody started buying the M doll but not everybody but a lot of people started buying the M doll and putting the IBM operating system on it and they were IBM was suitably distressed and so out of that came a kind of a legal issue of you need to unbundle your system software from your hardware and allow people to buy it as a separate line item and this kind of started this whole thing of here's the here's the software here's the hardware and never the twain should meet now another thing that came out of the long past and this really this really amused me a few years ago when everybody started talking about virtual machines and VMware came out and virtualization came every place I had to tell you this folks but I was using this all the way back in 1968 because there was a software called CP 67 and it was a virtual machine the van on top of the IBM system and you could just stop your program right on any instruction you wanted to you could kind of like stir it away and come back the next day and start it up from that instruction again it was sweet and it was called CP 67 originally but then IBM's marketing people got to it and they just renamed it into VM and the little operating system that ran on top of this virtual machine was called the Cambridge monitor system because it came out of IBM's Cambridge research facilities and the marketing people renamed that the conversational monitor system and that became CMS so VM CMS was really used as long ago as 1968 likewise when people talked about you know containers and stuff like that I said I seem to remember this thing called change route and it was I was using it at 1979 you know now granted you know containers and stuff are much advanced and everything and it's not the same functionality but you know there's a whole bunch of us that just went around going okay and soft to write software copyright and patents really started to be applied in the early 1980s and the thing that started to apply that was actually games and you could build a game and then the game became this little rom that would actually hold the logic of the game and people were making clones of those games and they were going to clone the hardware fairly well but they didn't want to do all the work of creating the actual software so they would just copy the ROM the ones and zeros inside the ROM and that was you know that was the way they did it and of course the game manufacturers to put all the work and sweat into generating those ROMs were upset about that and they applied to say we should be able to copyright those ROMs and then later on patent other ideas in software and stuff and the courts agreed with that and kind of out of that type of licensing and the fact that these systems were being manufactured in larger and larger quantities came the concept of the shrink wrap software license and the fact that when you opened up the package you accepted whatever abominable terms were in those tiny little smudge like writings of them and of course there's a whole bunch of people who objected to this one of those was Richard Stallman created the free software movement but in a lot of ways the free software movement already existed there were people who exchange software you know worked together with source code back and forth this was nothing new in fact it was proprietary software that was the new thing on the block the free software was just a continuation of something that already existed and you know of course you're aware of the fact that Richard Stallman wanted to create the entire operating system that he called GNU and in 1985 he created a free software foundation now here's the first brick of disaster thing what if he had said simply it's not the free software foundation but it's the free dumb software foundation just think about how much easier that would have made it on all of us that tiny little slip that tiny little mistake made it so oh no no software Libra software Libra you know and all this stuff and oh it's first free software I don't have to pay for it no it's freedom software idiot that's the most important thing and then he invented the GPL which people laughingly called copy left now Richard and I disagree on a couple things what yeah we do one of the things we differ disagree on is the concept of is there intellectual property and I personally believe that there is intellectual property I believe that when you write something or you create something or you do something that you should have the right to say what happens to that and that you should be able to protect that intellectual property either through copyright or patent or some other legal means and you know and because without copyright then every single license you generate actually has no meaning whatsoever because without copyright your software is in the public domain and so Richard would continually badmouth the concept of copyright and I would say well Richard if you didn't have copyright then your license wouldn't make any difference but I consider that a small thing really it did create some friction between us and then of course we also had all the non-free software that was coming up and CPM and people never talk about poor CPM anymore but CPM was my real exposure to the bizarre world of shrink wrap software and MS-DOS and that Apple company and other proprietary systems and we don't talk about those not much anymore either MVS VMS HPU X I mean all the proprietary operating systems and Unix systems some of which did work around standards now I want to say one thing about standards and and Ken Olson who was the president of digital equipment corporation famously said one time standards are as interesting as a Russian truck and the press dragged them over the coals for that and I thought that that was rather bizarre and he would say that because digital spent a huge amount of money in standards work in working with people to create standards and when I finally caught Ken one day and I asked him about that I said Ken why did you say that he goes oh he says that's just a personal thing I'm an engineer he says and you know I have to work on a standards body and those nitpicky little standards people they pick on this and they pick on that and I hate that stuff but once the standard is created it's my job to implement software to that standard to make it fast to make it small to make it robust to make it secure that's what I like doing and I said thank you sir thank you for clarifying that and that's what standards are really about because if we didn't have standards the pipes and your buildings wouldn't fit together and the electrical components and your computers wouldn't work and standards are very important open source and free software start to come about and projects that no longer exist like MIT's project Athena which gave us the X Windows system and Kerberos for network security BSD which fortunately does still exist but it's not just BSD the operating system itself it's interesting that the term BSD is not just Unix it's the Berkeley software distribution and they created a lot of software other than just Unix that they distributed I had a PIDP 11 with me at the LPI booth this week if people came along and start talking about it and then we got to talk about BSD and and all of the work that went into that in the history of it it was wonderful people think it's weird you say hey mad dog you're you're you're a Linux person and I am a Linux person but before that I was a Unix person and I worked with BSD and other Unix systems and the other software that we hardly ever talk about software like Send Mail, EXVI and software that came from the BSD distribution and passed out on to other distributions of other operating systems in fact I point out to people that if Richard Stallman when he started the Free Software Foundation had concentrated on the kernel first he might have worked with that kernel for two or three years and at the end of that development he would have no applications for it by the time the applications came around that kernel would be useless so instead he concentrated on applications that run across different operating systems and compilers that went across different hardware platforms to make it so that people like you could develop applications that were widely available to the greatest number of people and out of that creation of that software there was lots of little companies that all of a sudden came into existence companies we may not remember anymore but the Ingress company that supported the Ingress database engine that started at the University of California Berkeley by Michael Stonebreaker and Michael left the University and went and formed a company that gave support to the Ingress system and Ingress still state is free you can still pull the University Ingress system down and use it but if you wanted to get support you could get it from Stonebreaker's company out of that eventually was Postgres Michael went back to the University again started the Postgres program and the same thing happened very good free software mechanism Ingress who took the canoe compilers and created a support system Ingress was great I loved the company because the board of directors would meet at a hot tub out in Palo Alto you know to have board meetings I'd like that type of a company Matt we have to do that sometimes and it's companies like Primetime Software and Walnut Creek and another little company called Young Minds who distributed software now here's another edge of disaster story because once upon a time software was distributed on magnetic tape the tape drives are pretty expensive the tapes were expensive I remember a TK 50 tape from Digital Equipment Corporation was 100 US dollars back in 1989 100 US dollars and it held 95 megabytes of data and because it was a tape and because it was linear you couldn't really have something like a live file system on it and I at Digital wanted to see software distributed on CD-ROM and I had a lot of my management laugh at me because CD-ROMs oh those are for the music stuff yeah music and so I said no no I think that we should put digital data on there too your computer data so the standards bodies were working towards a standard and that was standard for the placement of the file system on a CD-ROM and one day I got a call from a guy at a company called Young Minds he said mad dog we have a problem I said what's the problem he says with a standard for CD-ROMs is going to allow for four levels of directory structure and file names that would have eight characters before the dot and three characters after the dot and no symbolic links I said really that's kind of bad for Unix isn't it he says yes he says the reason that standard is coming though is because the two major drivers of that standard are Dex VMS system which by the way used file file names of eight characters and three and Microsoft with MS-DOS I said I see the problem I said he says we have a fix for this it's called the Rockridge standards but they're not going to put those there are optional standards they're not going to put those into the ECMA standard and I said oh well well who's working on this who can I contact he says there's a guy at deck who's the editor of the standard I said really what's his name and he told me the person's name and I says that guy only sits a cubicle away from me I'll go have a talk with him and to make a long story short I blackmailed him into putting those standards in as a side note and a somewhat sadder note an engineer a young engineer was working with me to implement those those extensions in one of our distributions of Unix and we had to trick our management into letting us put that in one of our releases and that engineer was a very great engineer who unfortunately died at a very early age the age of 30 his name is Paul Shaughnessy and it's because of Paul Shaughnessy that digital got an early distribution of the ECMA standard ISO 9660 with the Rockridge extensions so you know and then as time went on there was lots of almost their solutions the Berkeley software distribution created us a different distribution than AT&T now why did a lot of the different companies like Sun Microsystems digital Hewlett Packard all decide to use the Berkeley distribution instead of the AT&T distribution well for one thing the Berkeley distribution had demand page virtual memory and the AT&T distribution of that time only had a swapping system the Berkeley system had a fast file system from Berkeley and the system five systems still had a rather weak and miserable file system there I mean things changed it got better but at that time at the time of the decision was being made these are the decision factors so a lot of the companies went with a Berkeley distribution at that point and then sometime later on as examples on Microsystems went from SunOS that was based on Berkeley to slow Larris that was based on system five now to be fair to be fair the system five code at that point was mostly a research type of system and therefore was never really tuned to be efficient versus the Sun engineers has spent a huge amount of time tuning SunOS to be even more efficient than it was from the Berkeley people and when they got the source code for system five they went oh my god this is terrible but their management pushed them to put it out and so I called that release the SunOS to slow Larris I think it was slow Larris 3 I called that the all-pain no-gain port another thing that happened about that time was a suit against BSDI because there's this little company that said hey we can give you an entire Unix system with the source code for $1,000 per machine and this was a really great deal because if you went to source code from AT&T the license at that time was $160,000 per machine well that wasn't the worst part the worst part was that you had to tell them the serial number of the machine that you were going to put the code on now how many of you know the serial number of your laptop oh one person raised their hand that's good the rest of you are slackers and if that machine broke you had to find the person at AT&T who had he was giving out the licenses and you had to tell them all I'm taking it off of serial number and I'm putting it on serial number before you could do that that was a nasty thing to have to do but BSDI didn't care about that you paid them a thousand dollars here's the source code put it on your machine unfortunately AT&T took a little bit of you know being nasty about this they sued BSDI and this became the first really horrible suit of AT&T versus everybody else in the world with UNIX and that dragged on forever and finally it was settled in 1993 and when that was settled and with some work by Berkeley because Berkeley as soon as this started to happen they said we're gonna rewrite all the code there's not gonna be a single line of AT&T code left in our distribution and Keith Bostic you know he should be sated okay Keith Bostic led this effort and all the people and I think it eventually came down and there was only like 17 files that had anything that looked like AT&T code in it he said oh that's easy we're gonna rewrite the 17 files and and out of that came BSD light which generated net BSD free BSD and then later on open BSD but even with that UNIX was losing because all the vendors all the major vendors had seated the desktop the Microsoft the vendor said oh we'll sell server systems and then people could just use MS-DOS on the desktop and then it came along Windows NT and then all of a sudden the vendor said oh my goodness they don't want just the desktop they want everything and that's when they started to panic and even Tim O'Reilly has stopped selling UNIX books well maybe he's kept up selling them but he didn't develop any new ones he started developing Windows NT books and programming books and stuff like that because he said oh you know I have to keep my business alive I employ people I have to keep them alive and I don't blame them for that because that's if you go bankrupt that's the opposite of making profit you know a lot of people in the free so I'm sorry a lot of people in the free software space they say oh they're charging money for that as if that's a crime even Richard Stallman says you should be able to make money with free software because the opposite of making money is going bankrupt and then everybody loses out so that was a really bad part and then along came this other thing the Linux kernel because the GNU people had everything well not everything they had a lot of stuff and we took stuff from Berkeley we took stuff from just a free software space in general and we put it match it all together to create these distributions unfortunately some that are no longer with us soft landing systems Iggresil that's how you say it by the way Debian fortunately Debian still with us red hat the slack wear and others and a lot of other distributions but the really thing that that they really got to me was the media love of Linus Torvalds and the free software of movement and I know that for a long time the BSD people were really incensed about this because they said hey we have a better TCP IP stack we have better this and better that I would say yeah you probably do you know but you guys have a demon as an emblem we have a cute little penguin you guys you guys and I'm not I'm not you know I I've always looked scruffy so I know I'm not a person to talk about this but hey quite frankly a lot of the BSD people at that time had big beards and long hair and stuff like that and here we had this nice clean cut young man with sandy brown hair wire rim glasses that were were were wool socks and sandals and spoke with this lilting spoke perfect English with this lilting European accent hello I am Linus Torvalds and I pronounce Linux as Linux Linus eventually left Helsinki Finland and came to the United States to work for transmitter and I called up transmitter one day looking for him and the voice on the other end of the phone says hello this is Linus I said Linus that isn't even your name because I know but nobody in California can say Linus so I'm Linus and that just endeared him to me because there would be so many people that would just spend the next hour on you I'm Linus I'm Linus you know he didn't care you know he famously one time says I don't care what you call the operating system as long as you use it so here's your chance BSD people you should just call Linux BSD so Linux version 1.0 came out in 1994 that's when I first saw really saw Linux and understood about it and I met him and gave him an Alpha system so he could make a 64 bit and also get the Intoisms out of it make it more portable but from a Unix standpoint it was still a weak miserable system because it didn't have symmetrical multi processing it didn't have failover it didn't have it had a relatively fragile file system you know all these different things and so you know and but we had lots of distributions that kept going on all the time but the other thing it didn't have was applications I mean sure it had all the little commands and stuff like that yes there were some you know other free software that ran on top of it but the type of commercial applications that sells a distribution or a computer system did not exist so I've bought out about five or six different operating system hardware combinations in my life commercial systems and the first thing you do is you release your system and then you go out and you try and get application people to put their applications on it and you go up to them and you say hi I'd like a mentor graphics I'll just use them as an example mentor graphics would you port your code to our system well I'm sorry you don't have the compilers we need you don't have the debuggers we need you don't have this you don't have that but then you say to them well but we're shipping 10,000 a day oh I'm just kidding it's running out of the lab we can have it next week because the only thing that gets an application vendor to port to your system is volume they'll tell you all sorts of other reasons but it's volume and if you can say that we're gonna have millions of systems out there they'll do the port no matter how difficult it is because they do the little calculation of mine they're shipping a million systems a month I'll get five percent of the systems oh my god oh my heart be still that's lots of systems so applications or lack of them was one of the problem but we had this great thing happening because there were these wonderful people called ISPs internet service providers and they were already providing machines and stuff for people to log into with shell accounts and stuff and these machines were almost invariably based on spark and Solaris and they found out very quickly that for about 30% of the price they could do the same thing with Intel and Linux and so a big market of Sun just went pew alpha underneath of them and then along the way there was this wonderful concept of lamp Linux Apache my SQL and Pearl or Python whichever you wanted I could have even thrown in Pascal there if you wanted it but and it was a nice algorithm you know lamp everybody loves a lamp you go into a dark room you turn the lamp you know you can't miss with that it's marketing it's all marketing right and so thank you Tim Burner Lee wherever you are for coming up with World Wide Web because it was just great and we found out that we could reuse old systems you know for more than just a doorstop we could have it we could use them for firewalls DNS servers all these things that we didn't want to dedicate a newer more costly machine to and so we you know when we take bringing in 46s we could use the old 3d 6s we're bringing in 586s we you know do the old 486s and people liked this they didn't like a certain company in Redmond Washington coming along and saying we realize that 5 million people are still using XP but we're not going to give you any patches anymore and if you have to you know and you have to upgrade the Windows 7 or Windows 10 or Windows 2000 or Windows me or Windows you or something like that in order to keep using your old hardware well gee was it doesn't want on that either and so you have a doorstop but we can use this new Linux system to make use of those old systems and speaking of reusing in 1995 a very interesting thing happened the supercomputer market was going out of business Cray ECL cyber CDC cybers were all going out of business because they would spend millions of dollars in making these designing these new machine machines and then they would sell five of them one of them to the agencies we did not say their name and you know who they are and then 40 universities that couldn't pay for them anyway and these companies were going out of business and two people Dr. Thomas Sterling and Donald Becker of NASA implemented this concept that became known as Beowulf supercomputers now is somewhat distressed today that breakfast I was talking about this and there was somebody that didn't know what a Beowulf supercomputer was be still my heart because these were great they could use commodity systems to implement a distributed cluster that could solve very important problems and probably the most important problem and that includes today it's even more important today that it was in 1995 is this problem called fluid dynamics because fluid dynamics is every place even with things that we don't think of as being fluid I mean this is we think of this as being solid and it is but the heat's passing through this is fluid and we need to be able to track that heat water air the environment everything is fluid dynamics and we need to solve these problems and it gave them the ability to solve these problems 40 times cheaper or looking another way but the same amount of money you could buy 40 times the computing power but the great thing about this for Linux was that most of these applications were not something you bought in shrink wrap that at the store how many of you gone down to your PC store and seen the box of supercomputer software shrink wrapped up 2995 those are actually it did exist red hat one time made a CD rom they called it the extreme Linux CD rom for rocket scientists and they and they thought they were only going to sell like a hundred of these and they sold thousands of them most of them were never taken out of the shrink wrap people just wanted to buy it and stick it on the shelves so they could say they had supercomputer software on their shelf and today there are still people trying to make supercomputers out of the raspberry pi zero never mind the fact that the poor raspberry pi zero is going to use all the CPU power just sending the data from one to the other and have nothing left over it actually solving the real problem in 1997 to 1999 it was a series of things that started to spring up slash dot source forge thinking now here's here's a another almost disaster happened I had a friend of mine I was working for VA Systems and he told me he was just an I can intern there and he told me that he was thinking about going to West Point oh dear I just sell 10 minutes left I'm one slide 19 of 157 he said he said I'm thinking about going to West Point assistant was that's very nice but that is the thing in my mind this is going to go to West Point it's going to end up in some military no it actually it was the Naval Academy it's going to end up in some ship and he's going to be sunk and we're going to just lose this great programmer and I said Drew I don't think you should I think you give back what were your country by being a good free software developer then going to the Naval Academy and he thought about that for a little bit he decided not to go and it was Drew's tribe who came up with a concept for source forge and implemented it so I am guilty of taking away a great naval officer on the other hand I am guilty of helping to give source forge in 1995 Patrick DeCruz from Australia decided we needed a organization to try and promote the business side of Linux and so he started Linux international and the first president of that was Alan Federer after about six months Alan said I need a job that pays real money so he left and I took over I was being paid by deck and so I could do the job for free to help the Linux community and one of the first things we had to do was to protect the Linux trademark from a guy named Bill Delacrochie in Boston who had trademarked the term Linux and was holding it for ransom and so we jerked it away from him and gave it to Linus and created the Linux mark institute to protect it for all time we also started the Linux standard base project that was started by Bruce Perens after about a month people realized that Bruce wasn't the person to lead that and Jim Zemlin actually took over and started to took the Linux standard base project forward we realized we needed certification and we helped to promote two different certification methods one of them was SARE and one of them was LPI SARE was a for-profit that combined teaching with a certification LPI said we feel that people should be able to get sort of a teaching or education any way they want to we only do certification SARE eventually failed LPI went forward and is still going strong and we supported early Linux marketing events in 1998 database support came in with Informix being the first database to actually support Linux and Oracle coming about nine months later but they found out about the Informix announcement and they hurried up their announcement two days before taking some of the wins out of Informix more side events Sun buys my SQL 2008 and then Oracle buys Sun yeah okay in 1998 the open source term was coined I am one of the people credited for the term open source it where he wasn't me I was in on the meeting but I left to go to the bathroom and by the time I came back it was too late but it did recognize that all these different licenses some of which were competing with each other and incompatible with each other we needed to cut down on that and some people didn't like open source and there was one particular person that didn't like the term free software because they considered the GPL a virus we'll see more about them later the dot-com boom really got underway in 1999 it burst the 2000 and a lot of the things that we love and dear were in danger of going away but they managed to survive right before the dot-com bust the IPO of Red Hat one of the most successful IPOs of all time and VA Linux systems that was definitely successful as an IPO even though the president of the company admitted to me didn't really have a business plan Sun microsystems became open and IBM invested a billion dollars in Linux and the next year they recovered that which was a big shot in the arm for the Linux systems open source development labs OSDL was created in 2000 they were started to help developers get access to hardware because even though developers could have a PC under desk or something when you're trying to put an IBM mainframe underneath your desk it's difficult and they hired Linus Torvalds and paid him a salary and he hired other people but unfortunately they were on the verge of bankruptcy in 2007 they were reorganized as the Linux Foundation under Jim Zemlin and have been fairly successful since then of course FOSDM was created in 2000 but it was actually OSDM at first and became FOSDM the next year we'll hear more about that later and free and open source developers came in so these are some of the interesting events from the year 2000 to year 2020 embedded systems in the year 2000 Linux became the most used operating system and embedded system design starts why because up until that point everybody wrote their own operating system for embedded systems but then a very strange thing happened in 2000 the customer started to say I want my embedded system to talk to the internet oh by the way I want to use all these little chips that are really neat and the embedded system people said TCP IP stacks are hard security is even harder and we don't support all those chips we only support one oh my god there's this operating system called Linux that does all this and so they went over to Linux canopics came out September 30th 2000 very interesting operating system is a live operating system Klaus is a very great guy really love them free BSD jails were in 2000 because they could and so people started using free BSD free BSD jails to be able to isolate various pieces Steve Ballmer said that Linux is a cancer June 1st 2001 in 2003 to 2007 there was a lawsuit of SCO against Linux and IBM very famous lawsuit how many people remember that who doesn't remember that I want to take the time right here and right now to tell you that that was the evil SCO that did that that was actually Caldera who had bought SCO from the good SCO guys and it was the evil SCO guys that did that the good SCO guys were Doug and Larry Michaels who started Santa Cruz operations created really nice little Unix to distribution and they had they had we were going to sell SCO and start a company called Terra Tella and the reason that I would tell you that these guys were good guys was because in 1997 Doug Michaels insisted that Linus Torvalds get a lifetime appreciation award in 1990 at the tender age of 27 I was 47 years old I hadn't got a lifetime appreciation award and here's this guy getting a lifetime appreciation award and Doug insisted on taking Linus to breakfast the morning of the award and talking with him and afterwards Linus came back and I said Linus what do you say to you and he says oh I was so embarrassed because he asked me how could SCO help Linux Ubuntu started in 2004 green great desktop distribution Android started in 2008 and then there was the VMs coming out of different types KVM Zen and different ones and then the cloud started coming and these are things that are just kind of interesting to me okay the Raspberry Pi shipped in 2011 and then we had containers of different types show up Docker and Kubernetes and others I got an interested FPGAs Linus has said if he if he wasn't going to start an operating system again he probably would start studying FPGAs and risk five I'm very interested in risk five about that time I had a friend working for a consulting company his name was Jonathan Eunice the company was called Illuminata unlike a lot of people who are analysts Jonathan was a technical analyst he looked at the technical abilities of different operating systems and he specialized in Unix so we compare SunOS to Solaris to digital Unix to Ultrix to all these and I asked him my first story working with Linux I said Jonathan what do you think of Linux he says Linux is a toy a couple years later I said what do you think of Linux he says Linux is getting interesting I asked him a couple years later I says he says I recommend Linux for non-critical applications to my customers and finally I was at a enterprise meeting down in Washington DC with the government education business and Jonathan was a keynote speaker and I got up at the end of his talk I said Jonathan and I recited all these things he had said to me in the past I said Jonathan what do you think of Linux now he says I recommend it I recommend to all my customers that they at least consider it for every application that was a long way to come in just a short time but I when I was leaving digital in 1999 the day I was leaving my senior manager took me down to the cafeteria very quietly asked me to question when can I fire all my deck Unix engineers when will Linux have the functionality necessary to support our customers and I can migrate them off of digital Unix onto Linux I told him that I couldn't give him the exact date but it was going to happen so why open open gives all of you people the ability to get the code and know that you could put it into your products the things that you are making but most open licenses do not force you to give your changes to the end user or even back to the open system why freedom software because freedom software gives you the same capabilities of getting the software and knowing what you can do with it but it also allows the end user to have access to those changes I have an entire drawer full of calculators and small computers that I cannot use anymore because the software that went into them was not free software and people say well most users don't have the expertise to fix it and that is true but they don't have to they don't have the chance of fixing it they can't make the decision to make the investment to fix it they can't join with a group of people if the hundred thousand XP users had gotten together with a group and said we want to fund patches to XP if they had had the source code for XP they could have done that they might have been able to find one Microsoft engineer that could code and so what are the new challenges that we as a group have we have to decide who really owns free software do we own free software oh I'm sorry I went back do we own free software do the companies own free software does the community own free software to end users own free software or is it all of the above nobody owns free software because we all own free software and that is why I like free software over over open source and what worries me is that some companies who say they love open source are creating closed solutions with open source and there's been a lot of people in this track who have talked about that today not just me and the other thing that worries me is we have 95 percent of people who are now using open source and some they're using free software that don't value that freedom and they don't care where they have it or not and that's a problem so we have new challenges and opportunities these are the major ones I think security and privacy it's worse than most people understand one of the reasons I'm interested in this five is because I want to create a completely open ship with a completely open bios with completely open device drivers we completely open software we completely open applications change with the chain of trust up it and those processors you should be able to look at the mass for those processors and be able to say yes this is free of trap doors it's free of malware we should be able to have clouds in our own home from the cloud software that if you want to use clouds from a company you can but we should be able to use the same software in our own homes we need ease of use that is so intense that even my mother and father should be able to use it I wrote a white paper in 1986 talking about my mother and father and electronic things if it had more than two buttons on it they were lost they would not be able to survive without their tech guy it's funny to vage her vaults every time she talks about linear she calls them the IT guy yesterday the IT guys father showed up here literally and it was nice to see him again artificial intelligence all the people out there to talk about artificial intelligence please stop calling it that you should call it in organic intelligence because there's nothing that is going to limit artificial intelligence I believe in the Alan Turing model of artificial intelligence that says that our brains are made up of synapses and neurons it's a chemical electrochemical reaction and if we could duplicate that with silicon we can have systems as intelligent as us and there's nothing that's going to stop that be very afraid money versus the community where the problems with free software is we typically don't generate enough money with free software to put lots of ads on TV education with false we're still not doing the right education I think I'm out of time but the fight is not over we need to fight love is love I I got in a problem with that last year I'm not going to go into how you should act in the shower but we should love free software and we should be aware that some entities that we're using free software are now going back to proprietary software but one of the things that drives me forward and I really love this and it's happened here I hope it continues to happen it's when the people come up to me and say I listened to you 10 years ago 20 years ago 30 years ago 40 years ago and now I own my own company or employees 60 people or I've created this free software project that 10 million people are using is because I listen to you thank you and now I say thank you thank you thank people at foster