[15:52] 111111111x111111111 is? [15:53] oh === service is now known as Guest71112 [16:16] hi === guy is now known as Guest14146 === turnip is now known as cheller [17:21] *tap tap* [17:21] this thing on? [17:21] 10 minutes until Q+A with Allison Randall, the Ubuntu Technical Architect! [17:22] ? === ChanServ changed the topic of #ubuntu-classroom to: Welcome to the Ubuntu Classroom - https://wiki.ubuntu.com/Classroom || Support in #ubuntu || Upcoming Schedule: http://is.gd/8rtIi || Questions in #ubuntu-classroom-chat || Current Session: Q and A with Allison Randall, Ubuntu Technical Architect - Instructors: wendar [17:32] welcome to our weekly Q+A sessions with engineering folks in Ubuntu [17:32] this week we have Allison Randall, the Ubuntu Software Architect [17:32] as always, please post your questions in #ubuntu-classroom-chat [17:32] wendar: ok, introduce yourself and take it away! [17:34] greetings programs! [17:35] I see interesting questions coming through === apachelogger_ is now known as apachelogger === Subr00t is now known as subroot [17:37] QUESTION: What does being the Ubuntu Software architect mean? What do you do? [17:38] The technical architect is a new role for Ubuntu. [17:39] I was brought in as a cross-hire from another open source project Parrot. [17:39] I've been the Chief Architect there for 5 years. [17:39] There are several pieces to a software architect role. [17:40] A big one is being the person with the "overall" view. [17:40] Most people focus on one piece of the distribution, the desktop, server, particular applications, etc. [17:41] But, that means that sometimes things slip through the cracks because they're outside the core focus of all the little teams. [17:41] Or, that things get missed because they involve 2 or more teams, who don't have good visibility into what the others are doing. [17:42] Another aspect of an architect role is planning for the future. [17:43] Not just "heads down" to the next release, but where do we need to be 2, 5, 10, 20 years from now. [17:43] next question, jorge [17:43] Question: So what will Wayland mean for the future of Ubuntu. Could it help Wine with OpenGL based games? [17:44] You've probably seen Mark's post on Ubuntu's intention to adopt Wayland in the future. [17:45] In the short-term Wayland isn't quite ready, so we won't be using it in Natty, or even the "O" release. [17:45] In the longer-term, Wayland has great potential. [17:46] Wayland was designed with security as a central focus, so it makes more advanced sandboxing possible. [17:46] It's lighter than X, because it's a more minimal subset of features. [17:47] Performance improvements, especially on netbooks, etc are critical. [17:48] On Wine/OpenGL, a lot depends on what those projects decide to do with Wayland. [17:48] So, I can't predict, but I hope it solves some of the problems there. [17:48] next [17:48] cheller> QUESTION: What is Ubuntu's plan/strategy for HPC in the coming years? [17:49] great question :) [17:49] I'll take a bit of a historical step back there first. [17:50] It's intersting to watch how technology has grown over the decades, our phones today have as much power as a "supercomputer" several decades ago. [17:51] So, in some sense HPC is a false categorization. [17:52] And there's also some question whether monolithic "super computers" are the future of high-performance. [17:52] Or, if we're approaching the limits (in terms of the laws of physics) on monolithic hardware, and so the future of high performance is actually cloud architectures. [17:53] Our focus in Ubuntu is on two fronts: [17:53] - Improving performance of the bare metal install [17:54] That benefits all installs of ubuntu (whether monolithic or cloud) [17:54] - Actively developing Ubuntu as both host and guest for cloud architectures [17:55] With special focus on the ease of use in cloud architectures, which really *should be* as easy to administer as monolithic architectures (but currently are rather a pain to work with) [17:55] next [17:55] kim0> QUESTION: Why do you think mobile OSs like iOS and Android are picking up developers and apps perhaps faster than most desktop OSs. What's Ubuntu's strategy to keep up [17:57] That's something I've been thinking a lot about lately. [17:57] There are several working theories on what attracts developers to a platform. [17:58] One is to provide easy-to-use tools. That developers avoid pain, and so will flock to the easy path. [17:59] That's partly true, and it's something we have immediate and long term plans to work on. [17:59] Take a look at Quickly for a good example. [18:00] We have some more changes coming for that in Natty that will make it much easier to get your Quickly applications distributed in the Software Center. === karthik is now known as Guest7617 [18:00] In the longer term, Ubuntu (and a whole bunch of other Linux-based environments) is looking at the toolchain in a deeper way. [18:01] Not just for a patch here and there, but to really see the developer experience as a whole and remove the pain points. [18:01] That's something I'm particularly interested in, having spent the past 10 years working toward an "easy" developer experience in Parrot. [18:02] But, the "ease of development" is only part of the story. [18:02] I don't know if you've used iOS or Android developer tools, but they're still a long way from "easy" [18:03] An easy, well integrated, social developer experience is more along the lines of scratch.mit.edu [18:03] So, there's another layer to look at, and that is "what motivates developers to target a platform even despite the pain" [18:04] I could talk for an hour on that alone, but I'll summarize [18:05] - Developers want people to use their software, so they'll go for a sympathetic audience [18:05] Ubuntu has a large number of users, so it's growing to the point that it's appealing for new app development [18:07] - Developers want to feel connected with their users, so the new Ratings and Reviews feature in Natty is an important step forward [18:07] (that feedback in iOS and Android is a big part of their success) [18:07] - Developers want instant gratification [18:08] It's a huge damper on motivation if you write a quick app in a weekend, and then have to wait 2 weeks (or worse, 6 months) before anyone sees it [18:09] so, there's a good bit of work going on now to improve the developer experience with submissions, so they get apps up in days [18:09] and with some good sandboxing for safety, we can get those days of review down to minutes [18:09] I better say 'next' [18:09] om26er> QUESTION: which team are you part of? [18:10] I'm on the Ubuntu Engineering team [18:11] I'm a peer of Jason Warner (Desktop Manager), Robbie Williamson (Server Manager), Kate Stewart (Release Manager) [18:11] And report to Rick Spencer (Ubuntu Engineering Director) [18:11] next [18:12] mhall119> QUESTION: Since you brought it up, where do we need to be 2, 5, 10 and 20 years from now? [18:13] Linux has traditionally played a little of a "catch up" role. [18:13] Like, we're trying to implement features that other operating systems have already implemented. [18:14] That seems a little odd, when you think about it. [18:14] Free software is actually a very agile space. [18:15] It has a very high pace of evolution. [18:15] That is, having more than one way to do things is actually an advantage in pushing forward the state technology. [18:15] We can rapidly try out 5 different ways to solve a problem, and then pick the best one and move on to the next generation. [18:16] But, we aren't using that unique evolutionary advantage as well as we could. [18:16] Linux shouldn't be following other OS's, we should be leading. [18:17] That means deciding where we want technology to be 20 years from now, and actively making that reality happen. [18:17] the 2, 5, and 10 years are all steps toward the future of technology [18:18] * wendar debating whether to go into specifics [18:18] better move on for time [18:18] QUESTION: why the ubuntu team dont make something "unique", not importing things from Windows and Mac OSx?? [18:19] (feel free to chat with me on freenode at any time, btw) [18:19] Yes, exactly! [18:19] So, I see Unity as a good example of this phenomenon. [18:20] It's a bit of a change of pace in the desktop environment. [18:20] There are 10 minutes remaining in the current session. [18:20] Not just "what should Linux be like", but "what should the computing experience be like". [18:21] It's something unique to Ubuntu, and an evolutionary step forward. [18:21] Not the highest evolved state (we've got decades ahead), but a great next step. [18:21] next [18:21] om26er> QUESTION: for what I have seen there have been nothing done to boot performance work after 10.04 although the release notes of 10.10 claimed boot performance but I followed many blueprints closely and there was no work done on that side, are there any future plans on the boots speed. [18:22] There's a standard statistical curve around performance improvements. [18:22] The first set of improvements yield big gains for small effort. [18:23] As you go on, you get smaller and smaller gains for greater effort. [18:23] We assessed boot speed at the begining of the Natty cycle, and determined that although more gains were possible, they were relatively low impact. [18:24] That is, the greatest slow downs now aren't the boot, but starting up services after the boot. [18:24] I.e. what we need is more general performance fixes across the board. [18:25] next [18:25] QUESTION: What's your view on where cloud computing is taking the world. Now that IaaS is almost nailed, is Platform as a Service what everyone actually wants [18:25] (one more after this) [18:25] There are 5 minutes remaining in the current session. [18:25] briefly [18:26] I'm pretty well convinced that future technology is going to be centered on distributed processing rather than monolithic processing. [18:27] But, we haven't hit the even horizon yet. [18:27] event horizon [18:27] And cloud technology needs to be much easier to use before it can become truly universal. [18:27] next [18:27] mhall119> QUESTION: How about getting developers of things like "Angry Birds" making their apps available on Ubuntu? [18:27] last one *chirp* [18:28] I would love to have Angry Birds ported to Ubuntu. [18:28] I'm one of the board members on the new Application Review Board. [18:29] where we work on the "fast" process for lightweight apps like that to get into the Software Center [18:29] If you know an app developer who's interested in making a lightweight app, put them in touch with us. [18:29] end [18:29] (Like making Places for Unity!) [18:30] ok thanks everyone for participating but we are out of time [18:30] thanks allison! *clap* *clap* [18:30] thanks all for the great questions [18:30] Logs for this session will be available at http://irclogs.ubuntu.com/2011/02/18/%23ubuntu-classroom.html === ChanServ changed the topic of #ubuntu-classroom to: Welcome to the Ubuntu Classroom - https://wiki.ubuntu.com/Classroom || Support in #ubuntu || Upcoming Schedule: http://is.gd/8rtIi || Questions in #ubuntu-classroom-chat || [18:30] hey alright, smoke if you got em === xindz`off is now known as xindz === Subr00t is now known as subroot === kenvandine is now known as kenvandine_ === kenvandine_ is now known as kenvandine === zaphod is now known as Guest37599 === guy is now known as Guest47772