 Nancy equipment. Okay. Well, hello everybody. Thanks for taking the time to come this morning. It is a distinct pleasure and an honor for me to have the opportunity to start off the Debian related talks on Debian Day and to hopefully in the process provide a little bit of background and context about the Debian project that will help the rest of the talks during the rest of the day, maybe make a little bit more sense to you. Before I start, though, I have a couple of questions for you. How many of you are Debian developers? Yeah, so you all know all of this. I'm not sure why you're here, but that's cool. That's cool. I like an audience. How many of you are users of Debian but not developers? Excellent. I can't tell you how thrilled I am to see you here. And how many of you don't use Debian? You've just heard about it. You saw there was a conference here. You thought maybe it'd be some fun to come see what's going on. It's okay. Excellent. There are a few of you too. So what I want to try and do today is provide a little bit of an introduction to what Debian is, what it's all about, how it works, what it is that makes it so interesting and special. But before I do that, let me take a moment or two to introduce myself because I don't know how many of you actually know me and maybe what I have to say will make a little more sense or itself be put in better context if you know a little bit about who I am. I tell people that I made my first personal contribution of source code to this thing we now call free software in about 1979. I've had the pleasure of working for the Hewlett Packard company or one of its subsidiaries since 1986 and I now serve as the chief technologist for open source and Linux at HP. But more important for this week, I've been involved in the Debian project for a very long time. Among the people who are still involved in the project, I guess I'm now one of the longest serving participants. There are a few other people who are here or will be here this week who've been involved in Debian almost as long or as long as I have. But there aren't many of us because at that time the project was much much smaller. And even if everyone from back then were still involved, we would be a very small minority of the total today. Over the years I've done many different things in the Debian project. Back in 2002, I was elected Debian project leader and served in that role for a while. I was trying to count earlier this morning I was having trouble but we have at least three other current or former Debian project leaders here this week that I know of and you know maybe we can figure that out later. I continue to serve today as chairman of this thing called the Debian technical committee. I will explain a little bit later in my talk that that's maybe not as exciting as it sounds but it's certainly something that I have continued to do for a while. Over the years I've won some awards this is probably the most interesting one. In 2008 a French national organization voted me there for an open source software personality of the year. I'm not really sure exactly what that means but the award looks really cool and it's nice to have you know physical things to to point to from time to time. In recent years I've actually spent at least as much time on things that are involved in the organization of these groups that provide services to free software projects almost as much as I have actually working on software myself. This is sometimes a little bit frustrating because I am a technologist I like to work on technical things but when you've been around for a while and you spend a lot of time thinking about how the free software process works you realize that sometimes the most important thing you can do is to help with some of these organizations that provide services that allow other people to spend all of their time working on the actual free software. So I serve as president of software in the public interest which is a nonprofit corporation in the United States that provides services not only to Debian but to many other very well known free software projects as well and I'm on the board of directors the Linux Foundation and this year joined the board of directors of the Freedom Box Foundation. Freedom Box is something I'll be talking about on Thursday this week many of you have probably already heard about it particularly those of you who were at Debian Conf in New York last summer. The foundation has been formed to provide some context and legal structure for the work that we're doing there and I'm very pleased that in the last few days here we've made some significant technical progress working on actual software for Freedom Box and as I said I'll have a talk more about that on Thursday this week and finally those of you who know me know that it's not possible for me to give a talk without throwing in at least one gratuitous rocket picture and so there's your gratuitous rocket picture for today this is a rocket that these are toys that I play with and you know big boys big toys in this case that's my son this was about two and a half or three years ago that airframe was of particular significance to me because I built it to achieve a certain high level of certification within the rocketry hobby and in fact that is a photo of that particular rocket headed upwards on its way to about Mach 1.3 and about 17,000 feet or what is that a little over a little over five kilometers above ground unfortunately the only bits of that rocket that I ever saw again were the red parts the rest of it disappeared that's a long sad story but it led me to begin work with Keith Packard who many of you will know his name from the free desktop in X world he and I are now partners producing and developing avionics that are completely open hardware and open source that we use ourselves and sell to other people to use in the high power rocketry hobby and this is just proof as I've said many times before that my primary hobby is turning all of my other hobbies into free software projects okay enough about me so that's a little bit about who I am I think the real message there is B Dale's been around a long time has been doing this a lot and as a result he's had a chance to think about this whole process quite a bit so what is Debian this is the official definition that we use for Debian a lot it is an association of individuals who have made common cause to free create a free operating system I think it's worth spending just a moment to think about what this means and in order to do that I actually want to step back just one step from Debian and talk a little bit about this whole idea of the community development model or the collaborative development model that we use for producing free software a key attribute of Linux and many other free software applications is that they are developed and supported by the community what does that mean what it really means is there's no single company that's in charge this is very different from the traditional or typical proprietary software product where there's a company and maybe it has partnerships with other companies but in the end it delivers a piece of software that only that company and its employees can work on and maintain and everyone else is sort of in a downstream user ecosystem in the free software world that's really not how it works there are free software projects that are directly associated with a company in the past you know a good example might have been the MySQL database which had a company associated with it that did a lot of things but they weren't the only people working on it over time we've seen that the projects that have a richer ecosystem of contributors are the projects that tend to succeed and last for a long time and do very well the other thing that's interesting about this process is that you end up with a range of contributors with very varied interests coming from different backgrounds different geographies different social and cultural contexts and they each bring different things to this process meaning that the software that results from this effort is better greater capable of handling a broader range of applications whatever way you want to think about it it is the definitely the case that the sum is more than the whole of the parts and it is because of this diversity of participation and input that we're able to do so many exciting things and part of what fuels all of this is that the licenses that we have designed and that we use for licensing free software have some very interesting attributes the most significant of which is that every user can if they want become a developer of the software or pay someone else to be a developer on their behalf so what does that do that breaks down the traditional barriers between the producers and the consumers of software it says that the kinds of situations that some of us found ourselves in at the end of the last millennium you know the y2k bug need never happen again i worked for a company where in preparing for the year 2000 we discovered that we had some design software it was very important used for designing printed circuit boards that if the clock on the system rolled past january 1st 2000 um the design database was corrupted now why was this the simple problem of you know we are sure we don't know because we didn't have the source code but in the places we've seen things like this it's because somebody only reserved two digits for the year and when it rolled past 99 it blew up it overwrote something else and corrupted the database what was the problem well the company we bought the software from didn't exist anymore it was not possible to get the source code it was not possible to get an update to the binary what do you do in that case well in the short term we set up a system that thought it was still 1970 and we continued to use the software to finish designs that had been begun in it but in the longer term we had to make an investment at a time not of our own choosing but a time imposed on us by somebody else we had to make a change in the software that we were using and spend money to acquire something new with free software that can't happen because if there's a problem you've always got access to the source code you've always got the tools that you need to investigate diagnose and repair and if you yourself are not a competent software developer it's okay you can find someone else and pay them to do it for you because you have that right and this is a fundamental distinction between free software and what I like to call secret source software or that other kind the kind that comes in shrink wrapped packages and is all shiny but doesn't carry these same sorts of freedoms abilities and capabilities and the ultimate expression of this is this notion that if a particular development community that's upstream of you whether that's the original developers or the distribution or whoever if they ever behave sufficiently unacceptably it's possible for a set of developers to make a copy of that code it's called creating a fork and go work on that copy and in effect move the center of development the focal point for future development somewhere else there's a huge social cost to this kind of forking and so we don't do it lightly but there are many examples now three or four really important ones where the ability to fork has taken a project that's become completely stagnant and allowed it to have new life a really good example was in the x-windows system several years ago more recently after the acquisition of some micro systems by oracle and the resulting changes and behaviors towards certain projects we saw the open office dot org code base fork to form Libra office within the document foundation and I think as we go forward that freedom and that power to fork is the lever that we have to ensure that the developers who produce and maintain the software continue to behave in a reasonable way or that we can do something about it if they don't so these are all freedoms that come from the choices we have made about the licenses that we put our software under and what the terms of those licenses guarantee to us so in that context of this larger free software ecosystem why is it that debbie and matters what is it about debbie and that makes it interesting well first of all from the beginning debbie and has been about freedom about freedom in lots of different ways it is a stable and very functional development community people have tried to argue over time that oh you know debbie is not relevant anymore there's something else that's more exciting and more shiny or they said gee those debbie and guys they just argue with each other all the time and don't ever seem to get anything done the reality is there's this concept called the tyranny of the vocal minority and we suffer from this sometimes every once in a while I think external perception is flavored or colored by the fact that a small group of people in debbie really like to spend a lot of time yelling at each other in person and on mailing lists and sometimes this makes this community seem a little harsh or a little frustrating but the reality is there is a non-vocal majority who just keep sitting there working doing great stuff inventing new software coming up with new ways of managing maintaining developing and releasing software and now for you know a decade and a half or more this is an engine that has just kept running and as long as we keep putting a little bit of oil in the right places every once in a while I predict that it will continue that way for a very long time to come there is a really large number of architectures that debbie and supports and packages that we support I've given up trying to keep up with the numbers it's just difficult to wrap your brain around one of the reasons for this is that debbie knows always been completely open to contributions there is no real distinction between sort of a core and things that are around the core some other distributions make a big distinction between the core of the operating system and the things which are contributed by other people in debbie and we have this open society where we say if you want to help if there's something you want to do if there's something you want our operating system to do that it doesn't already do come join us come work on it help us make it better make it a more universal operating system than it already is and if you demonstrate your willingness and abilities we will officially make you part of our project and the software that you work on become part of our project and as a result we have a very inclusive approach towards the rest of the world and I also think that part of the reason we're able to sustain this large distribution is that we've learned the benefits of allowing policies to be evolved which help to drive a uniform experience what I mean by that is that in debbie and we take things that we have learned our best practices we turn those into written policies and then we expect everyone to conform to those policies and we have tools for finding out on the software doesn't meet the policies and we have good tools for tracking where things need to be changed and actually getting the work done and then the other thing that I think makes debbie and really important in the world today is that there's a really huge downstream dependency chain for example you know many of you've probably heard of Ubuntu this is a huge distribution that owes much of its development base to having started with and continuing to work with debbie and there many other distributions that are derivatives from things like app to sid which you know tries to provide a a fresher distribution based on sid as well to many distributions put together by regional governments in different parts of the world and then the other thing that's really important I think is the frequency with which debbie and is used as the basis for creating embedded systems inside HP I serve on something that we call our open source review board and that board looks at all of the proposed interactions between free software and HP proprietary software content and I am surprised at how often when a project comes to our review board and I look at the packages of free software that they're using and I look at the version strings I recognize it they went and got that source package from debbie and why do they do that well everyone understands that if you take packages from the main debbie and distribution there's a certain set of commitments that we're making about the fact that they comply with our free software guidelines and you can have certain expectations about your ability to use and reuse that software and that's a pretty powerful force the other thing that makes debbie and significant I believe is just how incredibly geographically distributed the project is this is a plot unfortunately it's probably hard to see on this little screen but with the little spots all over here those are places where people in debbie and who are registered developers have been willing to identify where they're located and if you look at this most of the places where there is a developed information technology community there are also debbie and developers I of course when I look at this plot have a certain personal sense of accomplishment because two or three of those spots represent individuals that I could name who I met and helped sign their key so that they could become a debbie and developer or went to lunch and answered a few questions about how to package shared libraries or something like this and over the years it's been my great pleasure to help to personally through the willingness of my employer to send me to interesting places around the world I've had the opportunity to meet and talk about debbie and with lots of people some of whom have become very significant contributors so if we go back and look at sort of the early part of history in debbie and I think it's interesting to see how the progression of things that happened in the project led us to create certain core documents and certain expectations about how everyone in the project interacts with each other and with the rest of the world that have gone on to form a very fundamental part of the way we think about debbie and today you know if we look back at when things were first beginning I mentioned that in January of 1994 when Ian Murdoch first put out a quote-unquote release of debbie and one of the things that came with it was a document called the debbie and Linux manifesto I'll show you a few key elements from this in just a moment but for those of us who got involved in the project in 1994 that document was really important because that explained what Ian's philosophy was it explained what it was he was trying to do and why he was trying to do it and that acted as something that attracted people like me who saw some commonality with of thought or some had some desire to participate in that kind of a project it's very interesting to me that you know very early long before other distributions came along and started to do this debbie and started to make technical innovations like the introduction of the packaging system that we use even today I tell people often that debbie and was the first Linux distribution that had a real modern style packaging system it was the first distribution that had a packaging system that recognized dependencies between different packages there's been a number of technical contributions like this and innovations over the years that continue to form the fundamental basis of how our distribution works by the time the clock had rolled forward to June of 1997 we were introducing this concept called the debbie and free software guidelines as part of the debbie and social contract I'll talk about that just briefly in a couple minutes too because I think that was a really significant milestone in the history of the project and as far back as July of 1998 we had support for more than just one processor architecture and today I've lost track we're up to 11 or 13 or 13 yeah I knew it was an odd number up there somewhere and I take some pride that over the years I personally have had the opportunity to either start or be an early contributor to five different ports of debbie and over the years alpha spark arm pa risk and itanium not all of which are still active but some of which have ended up being very significant so I mentioned the debbie and Linux manifesto it had really these four concepts that it embodied and these were the things that Ian Murdock who founded the project wrote down as his explanation of what he was trying to accomplish and as I said they served as an attractor for people like me to understand what this was about and did we want to participate the first thing he said was this was a brand new kind of Linux distribution developed openly in the spirit of Linux and GNU what he really meant by that is that Ian was very impressed with how the early Linux kernel development process was working with lots of people contributing things and a person in the center sort of coordinating and bringing those together and he wanted to combine that with the development of a larger distribution the second point is that it was being carefully and conscientiously put together and would be maintained and supported with similar care this is the basis of this notion that debbie and is largely about quality I believe that that element of the original manifesto is the thing that causes us to believe that it's more important to release debbie and when it's right than on a particular date and this is a you know still today a very fundamental element of how we in the project think about what we're doing on ourselves the third point that the design process would set process would be open to ensure that the system was of the highest quality and reflected the needs of the user community in other words we don't go hide in a corner somewhere unless upstairs at Bensky Dior and Banja Luca qualifies is hiding somewhere and that we actually communicate what we're working on with each other openly on mailing lists and through the bug tracking system and so forth none of which is hidden from the rest of the world and then finally probably the most controversial point at the time in the manifesto was this notion that Linux was not a commercial product and it shouldn't be but that didn't mean it couldn't compete commercially I'll leave you to think a little bit about what that means but it's very interesting to me to see that the huge growth and expansion of what's happened with Linux is largely because of that thing I talked about before there's no single company behind Linux there's no single company behind Debian it's this collaborative inclusive process that has allowed so many people to find their connection with the software and to be able to take it make interesting use of it so as the clock rolled forward and more and more people became involved in the project an interesting thing happened in the early days all of the people who joined Debian people like me had a pretty strong understanding of the terms of that manifesto and we understood the context in the software industry that Ian was trying to react to and trying to change a portion of and we didn't have any trouble understanding what it meant to collaborate and contribute with this project but as the project started to grow and we brought more and more people in what we realized was some of the new people coming in didn't quite understand the philosophical basis of what we were trying to do quite as directly and so one of the things that happened is we realized that as we brought new people into the project we needed to have some kind of process for ensuring that the new people coming in understood what it was we were trying to do and what commitments and promises we were trying to make to each other in that process and that led to the creation of this thing called the Debian social contract what is a social contract it's very interesting at the time we wrote it I'm not sure any of us really knew what a social contract was either but we were the first free software project that I'm aware of that created a social contract and the terms of the social contract were meant to be a promise that we made to each other within the project and the promise that the project collectively made to our users and when you look at it it is today stuff that you know at least in the Debian community we look at with this and we go oh yeah well that's obvious but at the time ensuring that we wrote this down was a way of capturing our core values as a project and ensuring that all the new people who came to the project could understand and explicitly agree to participate in our community in accordance with that core set of values many of the things here are things we've already talked about the most interesting thing though is this last point we kept talking in the social contract about things being free and free at least in the English language is a word that has too many meanings and so in order to explain what we meant by free we created something called the Debian free software guidelines and this was a set of criteria that if the software met that criteria we considered it free and if it didn't meet that criteria it was non-free and even today in Debian there's a distinction between the software that's in main and the software that gets relegated off to the side to the thing called non-free and it's based entirely upon whether that software meets or fails to meet the terms of the Debian free software guidelines the interesting thing is that at that point in history nobody had really thought about that a whole lot before everybody kind of thought they knew what it meant to be free software certainly Richard Stallman and his cohorts at the free software foundation had a definition of freedom that many people thought was a good one but the moment the Debian free software guidelines were issued several interesting things happened at that time there was a large archive site on the internet called Sunsite and Sunsite immediately accepted the Debian free software guidelines as their test for whether to accept new software or not into their free software archives a little while later when someone decided that free software was too confusing a term in English and they needed to create this new term open source the way they created the open source definition was they made a copy of the Debian free software guidelines and did a little bit of edit and replace and even today if you take the Debian free software guidelines and you put them here and you put the open source definition beside it you can see it's almost exactly the same the changes are small and they were attempts to clarify things that some of us don't think needed to be clarified but that's okay in the end my point is that Debian as a project had a need to create a social contract and we did that and in that process we ended up creating a definition of what it means to be free software that ended up being the basis for almost everything that anybody else now believes about the difference between free and open source software and things that don't fit that criteria another thing that I think it's important if you're trying to understand Debian is to understand how it is a couple of elements of how it is that the the project has been able to maintain so much software and to bring so much software into the ability to usefully work with each other over the years and the two things that I think are the most important in this sense are the bug tracking system in the policy manual and I believe that they actually work together in a very complementary way since 1994 Debian has had a public bug tracking system on the internet anyone can go and read the history of any of the bugs against any of the packages in Debian in fact anyone on the internet can comment on any of the bugs in Debian anyone on the internet can change clothes modify or work on any of the bugs in Debian that's usually a feature but this the key is this is a completely open system it's used primarily with email in and a web interface out but over the years other kinds of bug tracking systems have come and gone the Debian bug tracking system has evolved and had new features added and some would say it's become more complicated I like to think it's become more capable but who knows but in the end the power of having all of the bug reports against all of the software open and visible to everyone other developers users people who would argue about whether we're doing things right or not it doesn't matter everyone can see what's going on and see what we're trying to work on the other big thing is the policy manual and the reason I talk about the policy manual is that I think it's that the way it's been created is interesting we tend to discover best practices by seeing what people do that works and what people do that doesn't work so well from among those we try to pick the things that work and write those into our policy manual and then we expect everyone to try and implement those best practices in the packages that they're creating and over time we've ended up creating software that will automatically check for many elements of the policy to find out whether they're being complied with or not in the different packages and so it's the point now where someone who's packaging software for Debian can know before they ever even upload the software into the archive whether that package does most of the things right to comply with policy or not this is no substitute for a human brain but what we've discovered is by having tools that help us to construct packages correctly to begin with and tools that help us understand whether the resulting packages are compliant with our policies or not it has been possible to take the best practices that we discovered and put into our policy and to spread those behaviors through the entire archive and so one of the things that I think so immensely cool about Debian is that you can install almost any software from Debian's archive and expect it to be well behaved on your system it's not going to smash something else if if there's a conflict the packaging system will let you know this and you have to figure out you know which way you want to resolve it if there's software you need to have installed to use that which isn't installed yet the packaging system can discover that and let you know these are all best practices that we've learned over the years and coded into our policies and now we expect everyone packaging software for Debian to comply with those policies so this is sort of a combination of a social or administrative tool set with technical tools the combination of which leads us to have the ability to deliver thousands and thousands of software packages they really do all work together they can be installed in the system and uninstalled cleanly and all of these sorts of things and there's really no other software operating system in the world that does this as well for as many packages on as many architectures and all of that sort of thing so another important thing to understand about Debian is that as the project continued to grow you know we started with a few people who were attracted by the terms of the Debian manifesto people like me as we grew we realized we needed a social contract that everyone would explicitly agree to as part of becoming part of the Debian family and then as the project continued to grow we realized at some point that you know there are few times when individuals have a hard time resolving things amongst themselves we have the need to be able to do things like pick someone to be our leader to represent us to the rest of the world to make certain decisions within the project and so this document called the Debian constitution was created and it sets out very explicitly a process by which the project can make formal decisions it does not try to describe the goals of the project it does not try to talk about how the project is going to achieve its goals instead it talks about the minimal division of powers and the minimum number of sort of official positions that need to exist in order for the project to be able to operate as it got larger and larger one of the things that came out of this was a decision to use a voting system that at the time at least was a little different and many people had not seen which is a preference based voting system one of the problems with the way many of our political elections in the real world work is that you get to vote for one person and if that person gets enough votes great they win but if there are lots of other candidates and the votes are split they then begin complex processes by which things we as the voters don't have much direct control over happen to decide what the government's going to look like in a place like the united states it's just sort of you know a winner take all and parliamentary kinds of system sometimes other things happen but in debian we have this really cool voting system where you basically are ranking your preference for all of the possible alternatives or if it's an election for the debian project leader maybe your preference for the the candidates who are involved in the election and the consequence of this is that it's many of the behaviors that we see in simpler voting systems just don't happen in many of the failure modes don't happen the same way if you have three or four candidates who have very similar views and one who's radically different in a system like the one we use in the united states unfortunately the one who's radically different maybe the one who ends up winning because the other candidates divide you know the population amongst themselves and debian that doesn't happen if those four or five candidates with similar views are all more desirable to the voters than the one lunatic over here then the right things end up happening in the end as a consequence this voting process has gained a great deal of respect and in fact many other free software projects many other non-profit organizations around the world and even some government entities in various places have looked at these processes and the success that we have had using them in debian and have tried to duplicate those elsewhere so there are a couple things that I have learned over the years by being involved in debian and the first one I like to talk about is never underestimate the value of values and that's meant to be a little bit of a strange statement but I learned a long time ago through a process of management training thing that my employer put me through many many years ago that when you want a team of people to be able to effectively come together and work towards a goal you don't start with a vision everyone thinks many people think the right thing to do is start with a vision but there's a step you have to go through before that and that's a step of establishing a set of commonly agreed to core values in debian those are generally agreed to be the things that are written out in our social contract those are the core values that we agree to operate by as a project once you have that then you can create a vision of where it is you're trying to get to understanding that you all have a common set of things that you think are important and from that you can build strategies and objectives but when things get complicated when things go wrong when there are problems that you need to resolve knowing that you have a common set of core values always gives you something to go back to it gives you a place to fall back to and start over from if you need to and any project that's run as long as debian has has had the opportunity to do that once or twice another thing that i have learned if i had it all to do over again i would be very inclined to try and come up with some kind of an internal social contract for the project there are other projects including some of debian's derivatives that have explicit codes of conduct for things like behaviors on mailing lists and so forth and i have personally become to believe over the years that it would have been nice if we had thought earlier about doing some of those things in debian i don't know how easy it is to fix those things after the fact we talk about this every year at debconf is there's something we can do to improve the way that we behave towards each other in the mailing list i'm personally very pleased that in the last year or so i've actually seen fewer of the kind of knockdown drag out screaming at each other kinds of discussions that we used to have two or three years ago so maybe this is getting better and we just don't need to worry about it too much but i always am concerned about this thing i mentioned earlier the tyranny of a vocal minority when a small group of people start yelling at each other and they make so much noise that nobody else can stand to stay in the room it's a problem for a project such as debian and if i had debian to do over again that is one of the things i would be inclined to change and finally i think if you want to understand debian there are a few things that you need to understand about the way that we package and deliver software there are many other talks today and through the rest of the week that will go into more details about how various things work here but the key is to realize that all of the software in debian is packaged that those packaged are maintained by individuals in small groups that as i've mentioned before there are policies captured mandate best practices for those package a lot of how things work in debian and the way we're able to provide so much support for so many architectures is because of the very capable auto building network that we operate that allows packages as they're uploaded to go through a process of being automatically rebuilt for all the different architectures and then humans review the results of those build logs and make decisions about whether everything is okay to go into the archive or not this combination of automation with human oversight is i believe the reason that all of this works as well as it does and for developers and others another element of the openness of the whole project is that the build status and logs from those auto builders are always publicly visible so you can go see what's building what's not building why did it fail all of those sorts of things and after software is uploaded i think that in order to understand debian you kind of have to understand this progression that happens from the time a package is uploaded which always happens or almost always happens into our unstable repository we call that the version of debian that gets a new release more than once a day and then packages going through a process of mostly mechanical criteria checking to be promoted into this thing we call testing which forms the basis for the next upcoming future stable release sort of in parallel with that there's our security process for making sure that the existing stable releases get updates immediately when when necessary and then you know every couple of years or so we make a stable release and then more frequently point releases that update that stable release with various security fixes so at any given time they're really sort of at least three different versions of debian that exist there's the one that people like me run called unstable and many developers track unstable and use it many others prefer to run something a little more stable even on their development machines testing is the thing which will become our next stable release and there's been lots of discussion in the last year about making it easier for people to use that as the thing they actually run because it provides a good balance between stability and freshness and then there are the stable releases which in some way were the original enterprise distributions in that they have the same sorts of life cycles and support commitments that you see from commercial Linux distributions with their enterprise versions and finally I'll wrap up just by talking briefly about how it is that and why it is that HP has worked with debian over the years as I mentioned when I started I personally have worked with HP for many years a large part of it is that inclusive and collaborative nature of the project that I described earlier it has enabled us to pursue certain markets that were not well served by commercial distributions a good example is in the carrier grade telecommunications space where we invented some hardware and we went and talked to the commercial Linux distributions at the time and they all said that's great but we're busy working on other things and we said okay fine we took debian and built a debian based telecommunications distribution and we have now shipped that to many many customers mostly people who have big names in the telecommunications industry who combine that with their own software and then sell it to people who do things like deploy mobile phones at one point in history I was told that something like one third of all mobile phone calls in the world touched an hp titanium system running our carrier grade telco stack based on debian somewhere in the process of the call I never knew how to verify that and I have no idea whether it's still true or not but this is an indication of the kinds of things that people in hp and elsewhere have been able to go and do in debian and with debian because of this inclusive and open nature I've already mentioned the way things are done and and and used how the software gets used for many products we ship thin client computers that run debian that most of them end up being used to run local windows sessions off of Citrix terminal servers and things like that but they actually have debian on the inside doing that and then as a result of customer demand we actually provide support for debian on our prolonged servers if you're interested in more information about that you can either use hp.com slash go slash debian or there's a newer page go community linux that talks about not just debian but the things that we also do for other community distributions I think that over the years participating in debian has really helped hp to stay connected with the free software community I woke up sometime a few months ago with the strange realization that hp might be the world's largest distributor of gpld software I don't really know how to measure that but when I thought about it the number of systems that we ship that go out with linux on them the number of devices we ship that embed a linux kernel or other elements of free software that include gpld license software the fact that web os which is the palm bone stack which is now part of hp every web os device has gpld software in it most of the pcs we ship have at least one gpld driver buried in them somewhere even if they go out with windows I don't know how to measure that but I think it's entirely possible that's true a couple other things we've done you know we've provided a lot of hardware to debian and as many of you who are developers are aware we've sponsored a linux weekly news subscription for debian developers for a number of years and I'm pleased that we're still able to do that even given some of the changes in the economy in the last year or two so as we look to towards the future you know squeezes our current stable release we're all here talking about thinking about working towards wheezy as the next stable release I don't think anyone should expect radical change from debian because I don't think it's needed and in fact I think the project is alive well healthy and most of the things we care about are going the way we would want them to go I also like to point out that the debian community is very strong the way this project works the way we are structured means that we are very well prepared to and very capable of writing out things like the recent global economic downturn companies come and go projects like debian just keep living one way or another there's always something happening that makes it possible for the distribution to keep going there's a powerful ecosystem of derivative distributions that continue to take the work that we do to an even broader audience and we continue to attract lots of new contributors and finally what is it that you can do to help debian well use it I was very pleased when I asked my questions earlier to realize that most of you in the room do use debian how many of you though report bugs and offer solutions these are things that you can start to think about and do if you don't already do them there are lots of areas that always need more help you know honestly we have a lot of people who are good at packaging software and maintaining packages but there's so many other things the distribution always needs that we would love to have more people to help with and if you're capable committed and really want to maintain packages or do other work like this for debian we would love to have you begin the process of applying to become a debian maintainer or developer in closing I love this quotation from Margaret Mead never doubt that a small group of thoughtful committed citizens can change the world for indeed it's the only thing that ever has that's what debian's about we all understand that we are that small group of people sometimes in a place like this when we all come together doesn't seem so small anymore but in reality we are that small group of people we are changing the world thank you for your time your attention and I hope all of you have a great time the rest of the week now with the timing is there time for questions or no I'm easy I can I'll be here all week so if somebody okay um if you're hungry lunch has started but if you want to stay ask questions and stay here lunch thank you very much I will be here the rest of the week feel free to come find me if you have questions