 Folding is. Is this everybody? Then we'll begin. It's great to be at another debconf. Thanks everybody for turning up. I am not going to bore you too much. First of all, I have got to extend my heartfelt thanks to the entire debconf team. In particular the two local guys who have done a ridiculous amount of work. Cymru'n ei wneud eich bod yn gyfynnu am Anto o Sizar, i'n gweithio ddechrau i'w bwysig ac yn gweithio bydd o'r byd, ond ei ddechrau. Cymru'n gweithio yma, o'ch ddweud yma, o'i amser yn ardal yn Argentina, yn gweithio'n gweithio, rydyn ni wedi bod yn ymddangos meddwl arall o'ch gweithio'n gweithio, o'ch gweithio'n gweithio, ac mae'r gwneud yma'n eu bod yn cyffredinol o'ch ddebion bod yn ei gweithio. Ac roeddwch yn gwybod ar gyfer y dynol, ond yn fawr iawn i'nogaeth o'r cyflwythion. Felly, mae'r ddegwch yn ddegwch yn ymddangos? Degwch yn 1993? Mae'r ddegwch yn ymddangos yn ymddangos ymddangos. Felly, yw'r ddegwch yn angen wedi'r ddegwch yn ymddangos. Rydych chi'n i gael, mae'n ddegwch yn ymddangos i ddegwch ymddangos yma cynghwil â'r ddegwch. Degwch yn ddegwch yn ymddangos yn ymddangos. ddechrau mawr hanfer, ond nidwn yn deillwn i SNS, slyg a slymp, would have easy installation. Now, in 1993 easy installation meant something rather different to what it does now. And, back then, there weren't that many other distros. There were handful, frankly, there was very little competition for users. There were not many distros. In 2009, God forbid nearly 16 years later we have as of yesterday and I checked this with the keyring maintainer we have 1029 developers we have 26,000 I think it was 13 I could be wrong binary packages in I386 said we have dozens of I tried looking for all of them and I gave up after an hour I found several dozen there's probably hundreds just I can't find some of them of distributions that not necessarily compete with us but are actually using our code and building on it we're releasing roughly every two years well we're aiming for better than that we're managing better than that we may not be hitting 18 months that some of the journalists think we're claiming but we're still coming in in less than two years it's okay now possibly even the biggest difference we have shed loads of other distributions a quick search on LWN on distro watch and a few other sites showed thousands of different Linux distributions wow now I spoke to quite a few people I've done quite a lot of research for this and noodles found me looking at them and said you're not going to fill your talk up using quotes from lots of people yeah sure as hell I am save me writing crap I found on distro watch a description of Debian which almost fits on the screen apparently this is what they consider as a com we're conservative due to our support for many processor architectures newest technologies are not always included we have a slower lease cycle and discussions on our mailing lists and blogs can be uncultured at times wow that's being quite gentle now other things I've heard in the last week now somebody did say to me over dinner I'm not going to name names why does Debian always seem to be on the verge of imploding and equally in the last week I've heard from a friend that Debian is the best and most stable Linux distribution he's ever used we've got some really very different stories there depending on who you talk to why are those differences so big well there's some obvious answers we have some differences in what the users are expecting we've got some people who love the fact that we release every two years would like prefer it to be three or even four because in that way their data center just works they'd never have to look at it no problems the people with laptops the people who've got the very latest bleeding edge board com chipsets and whatever oh my god they need a new kernel they need new everything every day otherwise it's not going to work we've had differences in experiences for some people things really do just work everything out the box no issues some people well it almost all works but well I don't get working 3D or why can't I play DVDs back or my sounds a bit crap of course we get different levels of user now for most of the people in this room we're going to be developers we're not going to have any issues well hopefully fingers crossed but even if we do well yeah it's fun to go and fix the issues and tell all the people what we've done for a course for brand new user if it doesn't work you know if partitioning fails during the installer or if it doesn't find their hard disk or whatever that's not a cool thing to play with that's a showstopper problem so are we in fact the thing that we've claimed to be for years the universal operating system hands up who thinks we are okay we have a few cool you're optimists I like that so moving on we've got lots of teams in Debian you may remember we've talked about this lots we've got lots of teams that are working well we've got unfortunately quite a few that aren't I did a team survey to find out exactly which teams were doing well and which weren't just over a year ago when I first was elected to DPL I'm not going to do the same thing to you again anytime soon but the information that came out of that was very very useful what I have found out since then is that the individual teams that are having problems may change but overall the average issue is it's about the same we constantly have a situation where we need people to put effort in we constantly need new people joining teams every single team especially the core teams is looking for new people that's never going to change now new maintainers this is something that keeps on coming up hands up in fact if you're one of the people listed here of course you can read this please if you know that you've become a new maintainer last year please stand up let's have a look at you I know there's quite a few of you here yeah go on keep going and I know there's a few more names that I didn't make it on here because we had at least one new developer today I mean for God's sake guys let give me a chance to keep the slides up to date now did I say sit down come on yeah come on stand up so what can you guys do to help Debian what should you do firstly and foremost go and dive in go and do stuff don't be scared yes Anna yes what's funny don't be scared of doing stuff yeah people make mistakes people fix mistakes as well that's what software development is all about in my experience if you want to really help go and see which teams are asking for help the chances are there will be a lot of them you know in some of them you may be able to dive in if not you may be able to dive in in six months time once you've been around a bit longer and you've learned a bit more even more importantly go and find the teams that need help even if they're not admitting to it those are the people who really need it find whatever interests you and I'm hoping that there's got to be something in Debian that interests you otherwise really why the hell did you go through new maintainers and finally think about working on the core stuff too you know I said a couple of slides back we're always looking for more people for core teams that isn't going to change please I don't know Adam come on stand up Adam has dived straight in into dev script he's just about to be officially outed as a new member of the release team I hope well yes please he's great I like him I'll buy him beer right the flip side of that a lot of new maintainers seem to think and this is not entirely their own fault this is there's a lot of documentation out there which is good but it does emphasise possibly overemphasise the idea that people should be working on packaging new things before you go and package something new especially once you're actually in Debian go and have a look do we really need yet another web server do we really need yet another music player even if this one chews do we really need yet another calculator or yet another variation on your favourite shell script that has this brand new gorgeous feature that nobody else has heard of honest even more importantly when you get bored of it in six months and it will happen anybody else maintain it or is it just going to get sat in the archive I mean I've been around probably longer than I care to think I've moved on I started doing a whole lot of audio stuff I don't maintain any audio stuff anymore I've moved on I made sure I orphaned some of those packages at the point when they clearly weren't useful I moved on and now stuck generating CDs every time we do a release God that's really tedious but I haven't got too bored of it yet the thing to be careful of and I'm sure there's a lot of people here would agree we have too much crap in our archive don't add to it if anything the release team and the poor bastards like me who have to try and fit it all on CDs and DVDs at the end of a release cycle will love you even more for pulling out a crap package as opposed to putting in three more innovating are we doing enough cool new stuff who thinks we are oh for God's sake again this is possibly rose tinted spectacles when I joined Debian back in 96 wow it was cool there was so much energy everything new was going on every week and it was impossible to keep up these days I'm not quite so sure I mean we do have go on Martin yes I'm not yes yes you have a very good point and then should we be waiting for other distros to do the cool new stuff instead Jesus Christ now I want to do cool stuff hands up anyone here who doesn't want to do cool stuff when they're developing one of the problems that we have and I'm sure there's a couple of examples that have been on the Debian development last month that people I'm sure will recognise this by God do we do a lot of bike shedding it's we have people who want to do new things who have great ideas rather than saying that's great it's a that's great but shouldn't it be green you know or I think you know shouldn't we do shouldn't we do it this way instead which in the end actually doesn't make the slightest bit of difference to the idea but instead it devolves the whole drags the whole thing into a two week conversation by the time that's finished a lot of people are just going to have lost the effort lost the energy they had in the first place and it may never happen why do we do this any ideas absolutely and that's cool people should feel they're having a contribution but before you start trying to dive in make sure that your contribution really is useful that's not to say don't contribute but think about it please for everyone yes there's a thousand of us that means that if we all really want to have our say bloody hell of mailing lists are going to get very busy very quickly and if somebody else has already said made your point for you leave it you don't have to reinforce you don't have to restate it we also have lots of people saying no to new ideas typically something on the lines of we've always done it this way well just because somebody and chances are it could be some old get like me or Bidale 15 years ago decided I know let's do it this way that doesn't mean it's white it's just that it's old there's not necessarily any guarantee that it's the best way there's a fair chance because Bidale and I are both good people but hey yes exactly so don't necessarily assume that the new idea that you don't understand is therefore immediately bad give people a chance let them have a go fundamentally try saying yes I know that may sound like a really silly self-help thing go for it go out and say yes to everybody there's a really crap movie been out this year that does exactly that but we are trying to work together here not argue together in the end we are trying to produce an operating system at least I hope we are see how it works out oh that really doesn't work out see how it works out see what people can come up with and now somebody who's very enthusiastic but doesn't have great skills yet will eventually work out for themselves the flaws in their own plan but that's possibly one of the best learning situations for them to be in if you just dissuade them right from the word go then you'll make them think their entire idea is bad not just certain parts of it so what else does Debian fit in we need to play well with others now at various points we need to work well with other people in Debian don't we did nobody try to warm this crowd up before I started if you ain't saying we need to work with our users don't we yes I do why do we need to work with our users pop question if we don't work well with our users what do we got on you developers and where do we press gang those people into signing up for the new maintainer process of course we also need to work with upstreams the vast majority of the software we've got let's be honest isn't actually ours we've polished it we've made it nice we've hacked bits off one place put them in another we've made it play nicely with all the other stuff that we ship in the archive but we didn't write it all so we should be working well with those people who did that try to be nice with them try to encourage them to take our changes but don't necessarily fall out with them obviously there are some special upstreams of those but we've got to try working with them I mean even, God forbid we work with other distros I've been to enough conferences around the world over the last couple of years where I've bumped into people from Fidora from Gentoo from Red Hat from Mandrieva even dare I say it some of Buntu folks and you know what most of them are very similar to us they're trying to do cool stuff and we may be arguing about the really really really nitty gritty details but, God we really do agree on the vast majority of it let's work with them I know a lot of us already do I'm not trying to teach people to suck eggs here but hey, whatever I've got to come up with something and finally, work with everybody share the code whatever changes we've got on a push don't necessarily assume that because you put your stuff into Debian but it's Debian's and nobody else should be able to use it without coming and paying you or coming and buying you lots of beer all point is we're working on this because we love the code we're doing and we want people to use it that was what we started with that was what RMS started with the GNU project it's what Ian Murdoch started Debian for we want to share what we've got and even, I know it's making me sound like a hippie share the love yay, and I have rushed through my slides very quickly what else should we do Debian, I think although certain people might have suggested it's always looking like it's going to implode I don't think it's going to happen well, it's not for a while we do, you know, but yes exactly I mean exactly we're not about to die but please, life support isn't a good place to be let's carry on having fun let's carry on trying to do new stuff because, and this is where I can do the Yoda impression almost more fun leads to more people more people leads to more fun hands up anyone here who doesn't want to have fun doing what they're doing and that's better time thank God for that if you're not really please, what are you doing so, any more suggestions what should we be doing I mean, I've got a few more things to talk about as well but which I don't have slides for because hey, I'm crap, go on, Martin pass on your microphone yes excellent in the beginning of your talk you said where the universal operating system and you highlighted some of the dimensions you said there's users and there's developers and there's data centres and there's laptops and there's desktop people and then there are people who work on the shell and so on that picture at LCA last year I had a buff where I asked this question what does it mean to be universal and we couldn't really come up with one answer to that universal sort of precludes all of the answers we want to be everything all at once so does that mean it's the best possible marketing slogan it is awesome but is it sustainable can we do it is this something that is this the goal that we need to pursue so that we can help and that we need to help ourselves and that we do have problems in the project scaling problems if we continuously try to go into all directions is this really what we want to be doing is this what we can do in the long run that's a very good question Ian I'm sure Ian has an answer for you yes y ffennu sydd yn ymgyrch, y coheraeth cyfwyr sydd yn y cysylltu'r cyfwyr. Ond, mae'n beth sy'n eich bod yn rhaid i'r hyn o'r cyfwyr yng Nghymru, mae'n olygu'n gweithio'n gwybod o'r ddechrau, ac mae'n gweithio'n gweithio'n gwybod y system yn y rhaid i'r ffyrdd. Ond oes y byddai gyfnod hwn yn i weithio, ac os rydyn ni'n amdano, rôl iawn. Na maen nhw'n cael ei gweithio mewn astrach. Rwy'n meddwl â'r quer o'r stylio gym不知道au a allan ag gweithio, Nid yw'r rhiwp yn gwneud y gweithreif e notamment ychydig, oherwydd i chi'n credu ein gweithreif e twists sy'n gweithreif. Maen nhw'n gweithreif e at wel yma, a'r gweithreif e ésg byddai'n ysbrition gyda rheswng gymbrindig a dyfynisiad yw yng nghymru y gallai gynnig, a dweud yn 18-24 maes y cychwyn. A'n mynd i'r bydd ymlaen am y llyfridd ac yn y byddurau a'n ysgrifennu, ychydig i gyfweld yr unig yma. Wel, gallai'n go iawn o'r ysgrifennu. Mae'n ddim yn ddim bod yn ei ddiweddol, mae'n gweithio'n gweithio y bydd ymlaen, a'r bydd unrhyw o'r unrhyw o'r bydd unrhyw o'r bydd unrhyw. Mae'n ddim yn gweithio'r ysgrifennu, if everybody working on the project and everybody working on one of the core teams were to get reasonably specific about what things really are impediments to their ability to make forward progress and to meet the objectives that they and those around them are trying to set, and if there are specific elements of what we do as a project that are causing more trouble than they are adding benefit, those are things that we ought to identify and address, but I get a little worried sometimes when we start attacking the big picture thematic thing without somehow drilling down into specifics fairly quickly. I understand there's always a tension between the bottom of the world and the top-down view of the world, but it's not clear to me that having a long-term vision of being a universal operating system in and of itself gets in the way of our successful in any of these things. It's when some twisted version of what it means that we ought to be doing now in order to... It's allowed to end up sort of controlling our specific behaviors that things go squarely sometimes, so I'm much more interested in figuring out how to sort of... the things that really are, you know, sort of putting impedance in the path to success than I am trying to change the... So... Can I have a microphone, please? Yeah, Andy. Okay. Okay. So, let me please answer a bit of the questions from Madag about... I think I have some opinions about releasing. I think the question is asked wrong. It's not to say whether the release team has to. That's even the wrong part of the sentence. It means do we as Dabian want to release in a certain timeframe if so the release team is delegated by all of us to drive that, but only we as Dabian whole can release. And I really think that a universal operating system is something good and helpful also for the release team. And I don't really think that our diverging goals are a bit different, of course. But in the end if you look back at our social contract it says that the users are our priorities. And I think this is really very, very true. And of course we have very different users with very different wishes. And of course we can't say you make everybody really 100% happy. But we can offer a really great platform where a lot of people can be quite happy with it. I think we are good in Dabian at that. We could be better and we should try to get better. And that's something we need to do not only today, but just every day, every year and then we can even grow farther than we already are. Absolutely. Colin has something to say? Steve, this universal operating system thing really seems kind of easy and I think we need to make it a bit harder. So more seriously it's kind of interesting your definition of universal operating system. I've seen that definition floating around a lot. And I'm sort of puzzled. There seems to be someone missing. All the little Dabian offshoots, you know, the Ubuntu's of the world. The people who are taking Dabian and turning it into a product that's, you know, as Bidel told us, was, you know, running our mobile phone calls. All the people who are taking Dabian and basically using it as a basis for something beyond Dabian. And I at least would like to see that as part of what we're talking about in terms of being a universal operating system as well. Sure. I mean, oh, I should probably clarify. I mean, this isn't necessarily my classification of what the universal operating system is. This is to try and work out why we have different people with such different ideas of what Dabian is. You know, why people, why the quotes that I just provided, at least some of them, were so very different. It's, yeah, it's difficult. Colin. So I've never been a huge fan of the universal operating system slogan. But the thing that attracted me to that sort of presentation of Dabian when I first joined was that it seemed to encourage us to keep an open mind that we wouldn't do certain avenues or certain groups of people or certain organisations just because we hadn't done that. Just because it wasn't the thing that we started out to do. So when we, you know, when Dabian started, we weren't doing desktops, laptops, data centres, mobile phones, telcos. We weren't doing everything under the sun, but we weren't ruling them out. And I think that's a very healthy attitude for a project of this size with such divergent attitudes and divergent people. I don't think we need to turn that into we are doing everything right now, but... But we're certainly not ruling out doing everything. Exactly, exactly. Yes. And to sort of add on to what Colin said, I think one of the... Oh, you want me to stand up? I guess I can do that. Show off the pretty shirt after all. So I think one of the things that's really important about this is it goes to sort of the individual actions that individual developers take on a daily basis. Part of what I like about the universal theme and part of the reason I sort of reinvigorated it back in my second, I guess, DPL campaign was that I had this notion that we were at that time getting into a situation where individual developers working on packages or pieces of infrastructure would sort of exercise more editorial control than they really needed to sometimes. They were often taking fairly myopic views of how Debian was going to end up being used. You know, a very simple example from my own experiences. I thought I understood how people used NTP on computers. And, you know, all of a sudden the fact that people might actually be trying to do a reasonable time synchronization of machines that didn't have persistent net... a whole different model of usage and that I hadn't thought about, popped up and came along. And so this notion that when somebody else comes up and files a bug report or a wish list or an enhancement request or offers up a patch that adds some functionality that you wouldn't have thought was necessary or useful, the notion that you had this as a contextual framework to sort of expand your thinking a little bit and see, you know, can I help make the world a better place by adding support for that behavior? That's where to me this stuff gets really useful. Not that we try to do everything all the time for everybody, but that, you know, we're inclusive in our thinking. Right. While BDL is still up, I don't know how much people have heard about. There was an announcement just a couple of days ago about a new group being set up in the US. Open Source for America. In fact, do you want to give BDL a microphone back? Sorry, Martin. I want to ask BDL to say a few words about that. This is something that has been only just literally in the last few days come into public knowledge. And it's a really great movement to try and push more and more of, let's be honest, the biggest economy in the world to use more and more free software. I'm sure we'd all like to see that. BDL, please. Yeah, this is a new organization that has just coalesced within the last couple of weeks. We first heard about it Sunday, last Sunday, so it's been less than a week ago, and very quickly got in touch with Steve and we had a little discussion about the documents that were floating around very privately at that time, and we agreed that it seemed 100% like something that Debian would want to be involved in, so we very quickly pushed the right buttons to get Debian listed as one of the original founding members. We didn't immediately go talk about it on public mailing lists and stuff and a complete blackout on public information and media release of knowledge of the existence of the organization until it was announced at, I guess it was Tim O'Reilly's keynote at OSCON on Wednesday morning in the US. As you might be able to imagine, certain other participants in the IT business world might have had strong motivation to try and stir up the media a little bit in confusing ways if they'd gotten prior knowledge of what was happening. What this organization is about is to advocate for more use of free software in the United States federal government, and it's really about that simple. It's not specifically a lobbying organization. There's a big distinction in the US tax code and behavioral rules for organizations between a lobbying group and a advocacy group. This will be an advocacy group. The participation in this does not in any way shape or form mean that we are endorsing the specific behaviors or positions taken by other members of the alliance, but simply that all of the members of the alliance have agreed to collaborate on trying to encourage the United States federal government and all of the things that it puts money into taking more advantage of and including more looking at free software alternatives in their procurement processes and so forth for the future. I think this is mostly of value and import to folks inside the US. On the other hand, as we've learned in the last few years, things that happen in the US often tend to percolate out to other places. Hopefully adding this to the growing list of national advocacy organizations around the world that are pushing for more use of free software, open source, whatever we choose to call it at the moment in various places will help make the world safer for the bits. Martin, did you want this one? Is it okay if I go back to the topic of universal operating system? By all means. I wouldn't want to put you off. Early on in your talk, you asked the question or you asked the audience to put up those people who think that we are actually the universal operating system. Could I see those hands again, the ones that were up earlier? Is that a maybe? We're trying. That's exactly the point that I'm trying to make. It was not the majority of people who had the hands up right now. I'd like to know what's stopping us, what those people who didn't put their hands up. I could make them all stand up now, but I won't. What do you think is coming up from the universal operating system? I wonder whether, when I'm thinking about whether this could be something that we could start a process to discuss this afterwards because some people are going to have something to say about this right now, but I think there is a lot of content that could be developed if you actually sit and think about this for a second. I'd like to encourage everyone to do that. I guess I'll send an email around when I think about, or somebody else thinks about a way in which we could collect these ideas. I'm not sure how many people I speak for, but to me it seems more like a slogan and it's not really a meaningful question to say, are we the universal operating system? What does it mean? If it means what Bdale was talking about, then I think that's what we've already got. Debian is the one place where you can, with a reasonably good chance of success, go to some random maintain different pieces of software and suggest to them that it would be nice if it did something completely weird and actually get a fair hearing. That's really valuable, but to say, does that mean I want to stick up my hand and say we are the universal operating system? Does that mean that we should be the one operating system that everybody in the world should run as is? That's a very... That's a very different question. I think concentrating too much on this marketing slogan, we're a bit in danger of following our own marketing here and we should carry on with that in mind, try to be all things to all people and maybe just go on with it. We may well never get there, but it's a nice thing to try for. Matag ask how many people believe that Debian is the universal operating system. I would like to ask to raise your hand if, in your everyday Debian task, your goal is to make Debian the operating system. If you think of that every time you do things for Debian, it's something that, as Ian said, is a slogan that is hung in the web page, shy at first, but quite a good turn it. OK, so, dead whistle. Anybody else? Good girl. Ah, Aniesca. OK, it's on? Yeah. OK, is this working? It's a bit temperamental. It's gone. Try the other one. OK, this isn't working too. OK, so most of you know I'm not a coder myself, but maybe I can say something from a different perspective, like maybe a user or something. Well, according to the last sentence you were showing on your slides where people were giving feedback or what they think about Debian, the last one was the most stable operating system I ever used. So that's basically the reputation that Debian ran over other ways, others based, create something. I think you already have, or what should I say, work on, sure, because there already is a huge amount of realised and I think there's no reasonable way that's what I think. Right, next thing is, again, this is something I never got land to slides for. We have some summer code students here, I hope. Yes, no? Anybody going to admit to it? Oh, God, they're all hiding. Yes. Well, I can see Arthur white there. So, come on, Arthur. Come on, stand up. Are any of the others around at all? No. Oh, well, white. Arthur. I'm going to pick on people. Now, Arthur, no, was one of our successful summer code students last year. He was so enthusiastic that straightaway afterwards, he came round on to IOC and beat me up for being a crap admin last year. And that was fair. But of course, the second half of that was then, of course, he then had to volunteer to be an admin himself for this year and he's done an astounding job of getting our students organised, much better than I ever did. And he's now in the new maintainer queue. Please, God, let's get him fast-track because we need people like this. I think that is basically it for what I have to say now. Obviously, I'm going to be around. Please, there's quite a few of you. I'd like to have face-to-face discussions with and by beer, whatever. Yes, Martin, I did promise I owed you beer. I can't remember what for, but I'm sure I do. One thing I will point out that the people who in the SPI box this morning will be aware that Debian has rather large amount of money in the bank, probably an unhealthy amount of money. So we're actually looking for suggestions on useful for Debian ways of spending that money. Now, I appreciate holidays in the Bahamas or whatever might sound useful to some, but it might be difficult to justify to the tax authorities. I have added a boff to the schedule for the 29th, which is next Wednesday. In case you haven't seen it already, you need to make sure you update your view of the schedule periodically. Talks can and will move around and new ones appear. If you have any ideas, if you can't make that talk, please, by all means, mail me, come and cost me round a conference, whatever. I don't bite much. Or please, by all means, come along and help in a discussion about what we can do. We have somewhere north of $100,000 in the bank. That's actually beginning to get difficult for us to deal with. If we have too much money, we've become a target for people who want that money. It might sound ridiculous, but really, that's the state of the world. So please, if you have useful suggestions on what we can do with that that would make Debian a better place, that would make the world a better place through Debian, please let me know, okay? Unless there's any more questions, I'll let you all go earlier.