[15:57] Schedule Change: This morning we will start with Mark Shuttleworth Q+A [15:57] We apologize for the last minute change [15:58] Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | Ask Mark - Mark Shuttleworth === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | Ask Mark - Mark Shuttleworth [15:58] oops [16:09] Thanks for being patient everyone, Mark is on his way [16:16] hello all [16:16] Welcome sabdfl [16:16] hey jcastro [16:17] what's the format? [16:17] Ok everyone, thanks for being patient, please start asking questions in #ubuntu-classroom-chat [16:17] sabdfl: popey and I will paste questions here, you answer, when you want the next one just say "next" [16:17] jcastro: you doing this or me? [16:17] lo sabdfl [16:18] this is the easiest set of questions *ever* ;-p [16:18] 16:17:44 < toobuntu> QUESTION: For easing corporate deployments, it would be nice to have tasksel options for installing a directory server and a workstation that authenticates to that server with automounted home dirs. Do you see Ubuntu heading in the direction of more installer options on the alternate CD? [16:18] That sounds reasonable, yes. [16:19] the server team is now in pretty good shape, they did amazing work for 8.04 [16:19] both on the hardware enablement front and at the level of specific services [16:19] directory is a focus, and AD integration saw a lot of work, with Likewise [16:20] as for the specifics, it would be best to join the server team and participate there [16:20] who's up next? [16:20] 16:18:34 < qense> QUESTION: When you're a student or someone who has just graduated, what jobs(parttime?) does Ubuntu/Canonical offer for you? [16:20] 16:19:11 < qense> (adding to my question: what type of work, how much, parttime?) [16:20] we have relatively limited scope for part time folks [16:20] we have done some summer internships, working on things like bzr [16:21] guys have contributed good chunks of code on that basis [16:21] but the core of canonical is generally full-time, committed employment [16:21] in addition to that, we are starting to work with folks who become certified professionals around ubuntu, or bzr [16:22] so, a company approaches canonical for expertise, and we project manage but outsource the work to a known-good professional [16:22] the best way to win a job at canonical is to demonstrate real competence in some aspect of ubuntu [16:22] next ? [16:22] 16:18:49 < Lardarse> QUESTION: What are your plans and/or visions for the 3rd "phase" of Ubuntu development? [16:22] Lardarse may want to elaborate? [16:23] Meaning until the next LTS [16:23] right now, we need to become commercially successful while retaining our community engagement [16:23] we think ubuntu is uniquely positioned to be an open, community oriented but nonetheless commercially viable platform [16:24] we know that people love ubuntu as a desktop and as a project [16:24] now we need to see if they will deploy it as a commercially-critical platform [16:24] if we can succeed in that, it will mean that all of our members can start to treat ubuntu both as a passion and as a source of commercial opportunity [16:25] it will increase the value of ubuntu skills [16:25] and karma ;-) [16:25] that's a significant challenge - the project has grown, and the community has scaled (relatively) well [16:25] we have good delegated leadership on many of our community fronts - loco's, irc/forum/list councils etc [16:26] we have motu, and dedicated teams for server, mobile, desktop, bugs, translation etc [16:26] that's all pretty amazing [16:26] now, can we keep those folks passionate and engaged while ubuntu becomes a platform that businesses depend on? [16:26] so, that's what I see as the real next phase of Ubuntu [16:26] as for the technology, we are certainly starting a new meta-cycle [16:27] the next LTS will be 10.04 [16:27] unless red hat and novell agree to sync LTS / RHEL / SLES and want a different "first sync release" date [16:27] so, we are thinking now about the things we want to get done over the two years to 10.04 [16:28] the first major discussions will be at UDS-Prague, hope to see lots of you there [16:28] things I expect to be on the table are: [16:28] - pervasive networking, roaming from 3G to WiFi and WiMax or Bluetooth [16:29] - desktop technology investments - I think we need to look at energising GNOME, and perhaps Ubuntu / Canonical can help [16:29] - virtualisation, based on KVM or Xen, focused on the use of Ubuntu in the cloud [16:29] - mobile, MID's and smartphones [16:29] ... and of course a bunch more [16:30] i'm willing to make an investment in upstream development on the desktop now [16:30] if i can see a vision articulated by developers that i believe can actually deliver it [16:30] which will put Linux ahead of the Mac or Windows in terms of experience [16:30] next? [16:30] related to part of that... [16:30] 16:20:02 < rick_h_> QUESTION: how has the talk about trying to bring release dates for most of the big OSS apps gone? Do you think this is something viable for most OSS projects? [16:31] yes, I think it is very viable for OSS projects [16:31] GNOME has shown that it's very possible, even for complicated projects [16:31] i'm delighted that KDE has said they will try it too! [16:31] with distributed VCS like Bzr, it's much more feasible to land features "when they are ready" [16:32] which means you can release almost at any time! [16:32] and with test-driven development, you can maintain much higher code quality even as big new features get landed [16:32] all of that means that OSS projects should be able to embrace time-based releases [16:32] and syncronization makes sense [16:32] whether or not Red Hat and Novell will agree to syncronise major releases is totally unclear [16:33] but i think it would have huge benefits for collaboration [16:33] imagine if the kernel guys knew exactly which version of the kernel the distros would all ship on their long-term releases, a year in advance! [16:33] at the very least, i think they would make sensible decisions about how they responded to that [16:33] so, we'll see [16:33] next? [16:33] 16:20:15 < nikolaidis> QUESTION: (when) Can we expect to see an option during setup to allow authentication against an Active Directory/LDAP server so an end-user can easily join an existing Microsoft business network? [16:34] i don't know! [16:34] but the infrastructure for that, with Likewise, went in to 8.04 [16:34] that sounds like something that should be discussed with the desktop team at UDS [16:34] next? [16:34] 16:20:30 < toobuntu> QUESTION: When will Canonical have its IPO? [16:34] 16:20:44 < doctormo> QUESTION: Does Canonical need an IPO? [16:35] no need for an IPO [16:35] the plan is to build Canonical and Ubuntu until they are sustainable [16:35] then make the best decisions for the project and the company [16:35] impossible to know far in advance what those might be [16:35] next? [16:35] 16:20:45 < madrazr> QUESTION: My institution uses Windows IIS on it server. Its a HP server, I donno the exact configuration. We are facing lot of problems with respect to maintainance. We students and admins want to migrate to a GNU/Linux distro and we prefer Ubuntu, but convincing the management is a big task, because management feels there is no one knows the maintainance properly. How will Ubuntu/Canonical help us? [16:36] realted.. [16:36] interesting question [16:36] 16:21:03 < aoakley> QUESTION: When I try to convince people to migrate to Ubuntu, usually their main question is whether they can browse the web and open/edit Microsoft Word documents. What will Canonical be doing to increase awareness of OpenOffice? [16:37] Your institution could easily buy a support contract for Ubuntu from Canonical [16:37] and there may be other local companies that could provide support too [16:37] I don't think we can spend a lot on trying to convince your management to adopt Ubuntu unless they would buy a support contract [16:37] that's a question for local advocates [16:38] aoakley: i think OO.o is quite obviously enabled in the desktop [16:38] some of the UI ideas we are looking at for the 10.04 cycle make it more obvious, because that's such a common use case [16:38] when you double-click on a .doc attachment, the right thing happens [16:38] next? [16:39] 16:21:40 < ubunturos> QUESTION: Has there been considerable increase in profit margins from the last LTS release assuming most enterprises have opted for it [16:39] there's been a considerable increase in evaluations, but we are still very young in the enterprise OS game [16:40] canonical isn't profitable yet, but the LTS releases are very important for the support business, yes [16:40] because folks who are interested in support are usually also folks who want to run the OS for long periods of time, on the server [16:40] and on the desktop [16:40] next? [16:40] 16:22:01 < Solarion> QUESTION: What does Canonical offer for computing cluster management? [16:40] I know there are clusters running Ubuntu [16:41] and there are a couple of tools, like specialised filesystems, which are available [16:41] but I don't know more than that [16:41] the server team would have more insight, chat with dendrobates perhaps? [16:41] next? [16:41] 16:22:34 < selinuxium> QUESTION: Is there any thought following on from the JeOS project that Ubuntu will build a lightweight VM host OS? [16:41] well, with KVM in 8.04, you have the beginnings of that [16:42] to be a really goot host you need more than an OS, though [16:42] you need all the infrastructure provided by companies like VMWare or Qumranet, or XenSource [16:42] i think Ubuntu will become a great bare-metal host OS [16:42] but I assume that there will be a need for a lot more management infrastructure [16:43] we will partner with the companies that develop that infrastructure, encourage and support their use of Ubuntu, but leave them to profit from that infrastructure itself while we focus on the OS [16:43] next? [16:43] 16:22:38 < sadiq> QUESTION: What are your thoughts on the progress UbuntuMobile's been making? Have you received much interest from OEM hardware manufacturers? [16:43] sadiq: yes, we have more work than we can handle on the mobile front, but it's still very early days [16:44] the devices we are working on are "round one" devices, they would be used to get other manufacturers interested in the round two's [16:44] i think it's going to take a few years for the Ubuntu Mobile story to shake out [16:44] though, for folks who know exactly what they want to deliver, Ubuntu is already a good platform [16:45] i think Ampro announced an embedded Ubuntu port recently, for example [16:45] Intel is doing very good work on Moblin [16:45] i hope other companies get involved too, so that it can become a broader-based open source initiative [16:45] Intel is a pelasure to work with from an open source perspective [16:46] so I am confident in their understanding of "what makes a good open source project" [16:46] next? [16:46] 16:22:58 < BonesolTeraDyne> QUESTION: Do you believe the news of Adobe opening their SFW and FLV formats will help Ubuntu, or Linux in general, in any fashion? ( http://www.adobe.com/openscreenproject/ ) [16:46] yes, absolutely [16:46] because like the web, it separates application from OS [16:47] the web has made people much more willing to consider non-Windows OS's [16:47] because they perceive the real value to be inside the browser, and Linux / Mac both have good browsers [16:47] hopefully, Adobe's leadership will do the same for other areas [16:47] next? [16:47] 16:23:38 < BluesKaj> Question: why KDE4 ?...it's broken and it's awful [16:47] :S [16:47] (i would still like to see Gnash take off) [16:48] BluesKaj: it's a 1.0 for 4.0 ;-) [16:48] i think it's pretty awesome, personally [16:48] have played with it some, not too extensively though [16:48] I was about to say....awful is a matter of opinion.... [16:48] i think the Kubuntu team made a good choice in doing an unofficial remix of Kubuntu-KDE4 [16:48] It's certainly improving very quickly. [16:49] it's available for those who want it, but it's not pushed to everyone automatically [16:49] hopefully, Kubuntu helps to get very widespread testing, which results in rapid improvement of the codebase [16:49] so we can make KDE4 the default in Kubuntu sooner rather than later [16:49] RIddell will make the right choice, I'm sure [16:49] next? [16:49] 16:23:55 < tech0007> QUESTION: is it true that Dell is not offering ubuntu on its server markets? [16:50] yes, that's true, dell does not off pre-installation of Ubuntu on servers in their default web configurator [16:50] if you are ordering servers in bulk, just ask for it [16:50] i don't see any problem with that - dell will wait till enough customers tell them they bought something else because they didn't offer ubuntu [16:51] we have to prove that there's real commercial demand for ubuntu [16:51] so, if you are buying Dell servers, make sure you email them and ask about Ubuntu, because it helps to make the case [16:51] but your "vote" carries a lot more weight if you are really buying servers [16:51] next? [16:51] 16:25:59 < yann2> QUESTION: Do you plan to stop shipping CDs for free to become economically profitable faster? [16:52] if we needed to, we could do that, yes [16:52] but i love the fact that we make it easy for people to try free software, feel very good about the spend on CD's, and don't really see a need not to [16:52] next? [16:52] 16:26:13 < J-_> QUESTION: From the business aspect, can you recommend a good book to read on investments, and how to get the ball rolling? How does one start up a Venture Capital company to deloy an ongoing business relation of reinvestment? Thanks for what you've done, and how you're doing it. [16:53] J-_: read the writing of a guy called Benjamin Graham [16:53] back in the early 1900's IIRC [16:53] next? [16:53] 16:26:32 < michaelramm> QUESTION: Why was a beta candidate (Firefox) released with the final HH release? Most add-ons do not work currently and, from what I am hearing, it is hard to remove Beta 5 and replace it with FF2 to resume productive web surfing. [16:54] * Hobbsee already answered that [16:54] we'll push FF 3.0 RC into 8.04 in due course [16:54] expect 8.04 RC for 8.04.1 in two months time [16:54] and there's a firefox-2 package. [16:54] July [16:54] next? [16:54] 16:38:39 < Solarion> QUESTION: When can we expect Ubuntu/Linux advertisements on TV? [16:55] Solarion: when it makes economic sense [16:55] you will probably see advertisements for Ubuntu based devices, which is not quite the same thing, first [16:55] next? [16:55] 16:26:47 < chad_W> QUESTION: Have they learned from the servers going down 2 releases in a row now? Plans for improvement? [16:55] chad_W: heh. we were pumping 10gbits per second out of the DC in London [16:55] i think that's pretty amazing [16:55] and we have 200 mirrors [16:56] i have to compliment the team for keeping most services up under that load [16:56] next? [16:56] [01:49] QUESTION: Every good project has continuity plans to avoid reliance on key individuals in case the worst happens. Can we be confident that Ubuntu will continue as a successful legacy in the (hopefully never) situation that you were to fall under a bus? (A young programmer friend of mine sadly died last year leaving all kinds of mess) [16:57] i hope my demise would not be messy [16:57] in practice, or in law :-) [16:57] of course, i think i'm irreplaceable, but fortunately most of the leaders in ubuntu and canonical don't agree [16:58] there are plans in my will to continue ubuntu, and canonical, on a reasonable basis [16:58] and there are very good people who lead all of the key teams [16:58] next? [16:58] 16:31:14 < doctormo> QUESTION: For home use of ubuntu the biggest barrier is support, what thoughts does Mark have on helping LoCos collectivly offer that support in a way that is accountable? [16:59] no significant plans in that regard [16:59] the loco's do a critical job of raising awareness of Linux as a real alternative [16:59] and their very existence provides encouraging evidence of a local support network [16:59] which is what really matters [16:59] Ok, that about does it for time. [16:59] nobody buys windows to get support from microsoft [16:59] they want support from local specialists and service providers [17:00] and loco teams, together with the growing network of companies offering local services on ubuntu, play that role [17:00] thanks very much everybody! [17:00] Thanks everyone for showing up, and thanks sabdfl for sepeaking, next up is Christer Edwards who will share his tips on running Ubuntu on the Eee PC. [17:00] sorry we only had an hour this time === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Ubuntu on the Eee PC" - Christer Edwards [17:01] ,--8<- [17:01] |== Introduction == [17:01] | [17:01] |Welcome to the "Ubuntu on the EeePC" Presentation. Glad to see everyone here! [17:01] |I think this is a really cool sub-notebook which has really taken off! [17:01] | [17:01] |UPDATE: Linux Journal Magazine has declared the Asus EeePC the "Linux Product of the Year!" [17:01] `-->8- [17:02] sorry, bad paste there. [17:02] I'll be presenting on the EeePC and using Ubuntu on the EeePC. Stick around till the end, I have some cool screenshots :) [17:03] I should mention that I don't represent the EeePC in any official sense, its just my new favorite laptop. [17:03] I have done quite a bit of documentation on getting in going well on ubuntu. [17:03] As a basic outline today I'll be going over the following: [17:03] - Specs on the current and future EeePC [17:04] - Who might want an EeePC? [17:04] - Get Ubuntu on that sucker! [17:04] - Screenshots Baby! [17:04] - Plans for the future [17:04] A little bit about the history and specs for the machine. [17:04] The Asus EeePC was designed by Asus and Intel and released in the US and UK in the latter part of 2007. [17:05] It is known for its lightweight, portability and Linux default OS (Xandros variant) [17:05] The 700 and 701 series are the most widely used at this point, the 900 series coming soon. [17:06] The model that I have, which is the basis of most of my documentation (community wiki + my blog) is the 4G 701 series: [17:06] - 512M RAM (upgraded to 2G for $35. NewEgg FTW!) [17:06] - 4G solid-state flash drive (added 2G via MMC/SD slot, mounted as /home) [17:06] - 7 inch TFT LCD (800x480 resolution) [17:07] - 900 MHz intel celeron, factory underclocked (can be overclocked) [17:07] Intel 915 integrated graphics card (Yes, Compiz works on the EeePC!) [17:07] - 4 cell battery (3-3.5hr battery life) [17:07] - 10/100 wired, 802.11b/g wireless (both atheros drivers) [17:08] - weighs about 2lb [17:08] - $399 US at the time I bought it. [17:08] Its just super tiny, easy to carry around. My boss refers to it as my "laptoplet" [17:09] The new 900 series improves on the 701 series by increasing specs, including: [17:09] - 8.9" display (1024x600) [17:09] - 4G + 16G (SD) for 20G total flash storage. (only 12G if you buy it with XP.) [17:10] - 900 MHz processor (basically just not underclocked) [17:10] - 1G DDR (upgradable to 2G) [17:10] - 4 cell battery (2.5hr battery life) [17:10] - $549 US is the estimated price. [17:11] So, that's a bunch of specs. As you can see its not a powerhouse but its definitely usable! [17:11] I run this EeePC as my production, primary machine most of the time and I'm perfectly productive. [17:12] Who might want an EeePC? [17:12] I tihnk it works great for normal users. Web, email, chat, code, etc. All of those things work just fine. [17:12] its super lightweight. its tiny. Great for traveling. [17:13] I travel two-weeks per month for my job and my EeePC is what I take a long with me. [17:13] works just great. [17:13] People who might *not* want an EeePC? [17:13] probably everyone who left the room before when we started :) [17:13] gamers [17:13] my dad (he can't see his 800x600 LCD) [17:14] ...giant people with abnormally fat-fingers [17:14] I have big hands and I can type on it just fine. Its the only laptop that I can palm like a basketball! :) [17:14] So what are some of the complications of installing Ubuntu on the EeePC? [17:15] 1) no optical drive so installation has to be done another way [17:15] - USB installer (migrating .iso to USB), USB CDROM or PXE based [17:15] - boot.img.gz on USB boot + network install is my preferred method [17:16] 2) It uses solid state flash media, which will wear out faster than traditional drives [17:16] - ext3 is arguably fine with noatime or relatime. Maybe ext2 if you prefer no journaling. [17:16] - no swap is common for gaining additional drive space and avoid swap writes to disk [17:17] - 4G drive doesn't leave a lot of space for additional stuff after installation [17:17] - I have 100M /boot, 3.9G /, 2G /home. All ext3 and no swap. [17:17] - The 2G /home is via an SD card I've added [17:17] 3) 800x480 is not a lot of screen real estate! [17:18] - Gnome can be shrunk quite a bit! You'd be surprised. [17:18] - I also suggest gnome-do --mini for navigation. (screenshot later) [17:18] - My desktop has no menu, auto-shrinking panel + gnome-do. [17:19] 4) Wireless does not work out of the box.. (surprised?) [17:19] - Takes about 5 commands to compile the driver -- not hard at all. [17:19] - ndiswrapper is also available.. if you're into that sort of thing ;) [17:19] 5) Overclocking or "I want my 300Mhz back!" [17:20] - The EeePC is fairly easy to overclock to the 900Mhz but its not "supported" [17:20] - I've run mine overclocked and it works great. Actually "bursts" to 900Mhz when needed but idles at the 630Mhz [17:21] 6) ACPI support is not 100% [17:21] - instructions also on the wiki for this. Fairly simple package installation [17:21] 7) Not all the hotkeys are supported [17:21] - Also, documented how to get this working. not complicated. [17:21] So, now that I've mentioned all the difficulties and you've left the room, what *does* work? [17:22] Remember, it all *does* work it just needs to be configured. [17:22] The wireless for example isn't shipped because part of the driver is non-free. We *can't* ship it. [17:22] The Desktop needs to be customized because, well, not many displays are this small. [17:23] solid state flash media is still fairly new so configuring for that is also something a little "non-standard" [17:23] The EeePC is not your normal desktop/laptop, which is what Ubuntu Desktop is normally aimed for of course. [17:23] with just a little customization we can get everything working and it makes for a nice, easy to use machine. [17:24] Plus, another thing to note is that the default os, Xandros, is also a debian variant so I'm sure spending a little time with them and we can get everything supported the same way they do. [17:24] It'll just take a bit of communication, something that I'm more than happy to work towards. [17:24] I would love to see the EeePC fully supported by Ubuntu 8.10, and I *will* work towards getting that done. [17:25] I love this little thing, and if any distribution can excel at supporting hardware & making a great desktop experience its Ubuntu! [17:25] Now, let me share some screenshots of the EeePC desktop after Ubuntu is installed. [17:25] I did a fresh install just last nite of Ubuntu 8.04. [17:26] After using the UI shrinking tweaks that we have on the wiki it looks a bit like this: [17:26] http://zelut.org/dropbox/mini-desktop.png [17:26] basically we've just shrunk everything down to size. [17:27] I then like to shrink that down even more & add gnome-do for navigation. That looks a bit like this: [17:27] http://zelut.org/dropbox/gnome-do-mini.png [17:27] That uses a auto-shrinking single panel + gnome-do for all navigation. That adds a bit more real estate. [17:27] ..and a few screenshots for common applications: [17:27] http://zelut.org/dropbox/gnome-terminal.png [17:28] http://zelut.org/dropbox/pidgin-twitter.png [17:28] http://zelut.org/dropbox/fullscreen-FF.png [17:28] http://zelut.org/dropbox/Ooo-writer.png [17:28] and lastly, because I know everyone likes the bling, Compiz on the EeePC (out of the box!) [17:28] http://zelut.org/dropbox/compizftw!.png [17:29] Before we go on lets get some of the questions. I've seen them but I assumed someone would paste 'em in here (nudge popey) [17:30] 10:28 < jonphilpott> QUESTION: what's the boot time, and how disk space is ubuntu 8.04 using? [17:30] The boot time, last I clocked it was 45sec. Default install uses about 2.5G, so that leaves about 1.5 for your files. [17:30] Zelut: I can paste them if you wish [17:31] running df -Th just now on my machine (after adding my fav apps), I am using 3G of the 4G total. [17:31] AndrewB: ok. [17:31] The Xandros boot time is faster, yes, but I'm working on getting to that same speed. [17:31] say next when you want one [17:31] next [17:31] ah popey has appeared [17:31] 17:04:36 < alexpgp> QUESTION: I was impressed with eeexubuntu on my eee. Is the upgrade to 8.04 worth it? [17:32] I think 8.04 has a lot of great new features and definitely suggest the upgrade. [17:32] I only briefly used eeeXubuntu. I prefer the gnome desktop so that is what I've primarily worked on. [17:33] eeeXubuntu is definitely a great variant though--a lot of good work has gone into it. [17:33] next [17:33] 17:06:20 < Odd-rationale> QUESTION: Debian has an easy way to install debian on an eeepc. all you need is a 16mb flash drive. does ubuntu have any plans to benefit from this as well? (http://www.aigarius.com/blog/2008/04/07/debian-on-eeepc-2/) [17:33] I am not familiar with that but now that I know about it I'll be sure to put it on the TODO. [17:34] I'm sure it won't be too difficult to port anything over. Thanks for pointing that out! [17:34] next [17:34] 17:14:16 < ZrZ> QUESTION: is there a community hack to fix badly designed website (through greeamonkey or something else) [17:34] ( I think he is referring to the lack of screen real estate ) [17:34] None that I'm aware of. [17:34] I agree sometimes it is difficult to navigate a website on such a small resolution, but that is the price we pay for the size I guess. [17:35] it would be something nice to develop though [17:35] next [17:35] 17:15:54 < artir> QUESTION: Why dont you make a eeeubuntu (like eeexubuntu) version with all the hacks by default ? [17:36] I would love to develop an eeeubuntu. I am a little hesitant to create yet another variant. I think I might prefer to add the hacks by way of a .deb package. [17:36] that way you can install as normal and just install a package. too many variants and things can become complicated. [17:36] but, again, an all-in-one solution for this is definitely something on my list! [17:36] next [17:36] 17:24:08 < alexpgp> QUESTION: What is the best way, in your opinion, to distribute partitions if you plan to have a 4GB SD card in the machine at all times? [17:37] I like to use the new encryption features in Ubuntu so I use something like 100M /boot, / uses the remaining onboard and whatever you put in the SD is /home. [17:37] Again, I don't use swap (particularly with the 2G RAM upgrade) [17:37] I also like to encrypt /home [17:38] side question - does that make a difference to performance? [17:38] With flash drives getting bigger and bigger putting /home on the SD makes it really portable as well. [17:38] I don't think i've seen any issue with performance in that regard, no. [17:38] ta [17:38] I recently saw some encryption benchmarks and the performance hit was minimal. [17:39] plus, with as small as the laptop is I think encryption is critial. its so easy to steal or even lose! lol [17:39] next [17:39] related to that... [17:39] 17:26:10 < rzr> QUESTION: have you tested some filesytem that support "intelligent" on the fly compression ? this will reduce bandwidth but optimise storage [17:40] I have not. The only filesystems I have used are ext2/ext3 and reiserfs. The jffs, which is meant for flash media, is not really designed for something a user will interact with. [17:40] as I understand it it is better used for embedded devices or something. [17:40] next [17:40] 17:29:34 < rzr> QUESTION: are there some ways to make intellegent usage of user inteface (mousewheel etc) to scroll horizontally ? [17:40] There are no ways that I know of, but I'm sure its something that can be implemented. It is open source after all. All it takes is time :) [17:41] next [17:41] (note the 900 has multitouch) [17:41] 17:29:44 < jonphilpott> QUESTION: what's the boot time, and how disk space is ubuntu 8.04 using? [17:41] popey: I think we covered that one. [17:41] oh sorry [17:41] 17:30:47 < Sionide> QUESTION: Why not Xubuntu? Since XFCE is designed with lower specifications in mind? [17:41] ..and that one lol [17:42] 17:31:14 < jonphilpott> QUESTION: Were you able to get the acpi keys (wifi on/off, etc) working out of the box? [17:42] eeeXubuntu is a variant based on XFCE, which works great. I just prefer gnome so that is what I've focused on. [17:42] Those are not yet working in the default install out of the box, but they can be made to work with some minor tweaks. [17:42] again, by 8.10 I would like to have much better support for teh default install. [17:43] everything *can* work now, just not by default. My goal, again, for 8.10 is to improve that as much as I can. [17:43] next [17:43] 17:31:39 < jd2100_> QUESTION: Would the 900 model need shrinking? [17:43] ( I don't know what this means here ) [17:43] if that is referring to the display I'm sure it wouldn't hurt. the 1024x600 is still fairly small. [17:44] the drive space will be plentiful on the 900 model though. 20G means 5x more space than the 700 series! [17:44] next [17:44] 17:44:13 < jd2100_> popey: screen shrinking [17:44] to clarify [17:44] 17:32:15 < rzr> QUESTION: are there works done to replace default bios by opensource one such as coreboot (linuxbios)? [17:45] not that I am aware of, but there is a lot of scattered tutorials and work on the eeepc. Its still been a project just collecting it all. [17:45] If there is, and if it is possible, it will be something we will document on the ubuntu wiki. [17:45] next [17:45] 17:37:35 < maffru> QUESTION: what are your thoughts on fastinit? http://helllabs.org/blog/20071231/asus-eeepc-fastinit-reimplementation/ [17:46] I have not tried fastinit with the eeepc yet. (thank you guys by the way for these links!) [17:46] looks like it could be a nice potential improvement for 8.10 [17:46] next [17:46] 17:38:47 < rzr> QUESTION: is it possible to merge what xandros did in any distribution ? [17:47] A few of the things Xandros did to its default install are non-free, so that will make things difficult. [17:47] Anything that they do that are free software we'd be more than happy to cooperate on [17:47] What things are non-free out of interest? [17:47] the wireless driver (atheros) has binary blogs. some of the desktop software I believe is non-free. [17:48] *blobs [17:48] I have not done a full audit by any means, but I'd be more than happy to incorporate what I can. [17:48] next [17:48] 17:43:23 < maffru> QUESTION: how hard would it be *g* to put a unionfs layer on, like the Xandros shipped; and other than for security checks, is there any point? [17:48] (Note: This is what Canonical does with the Intel Classmate PC Ubuntu Build) [17:49] that isn't something i've even looked at yet. If there is interest in it I'm sure we can collaborate on it. [17:49] next [17:49] thats your lot [17:49] thanks [17:50] with the last few minutes let me say what I would like to see happen [17:50] I know there are a lot of Ubuntu community members with EeePCs. I think it'd be great to start working together (beyond just the wiki) to make it better. [17:51] I've learned about a few things just from the questions--things I hadn't even started doing yet. I think if we collaborate we could definitely improve the support and functionality. [17:51] if anyone would be interested in working together on a project like this feel free to email me: christer.edwards@ubuntu.com [17:51] perhaps an ubuntu-eeepc team? :) [17:51] yes! [17:52] get a spec together before uds prague! [17:52] I'll champion it there [17:52] I don't think there is any reason that Ubuntu can't be as much of an option an Xandros or EeeDora or anything else. [17:52] popey: perfect! [17:52] Let me also drop some links for resources: [17:52] https://help.ubuntu.com/community/EeePC [17:52] that is the ubuntu community wiki page. please feel free to improve it. [17:53] http://wiki.eeeuser.com/ubuntu [17:53] this is the eeeuser global wiki. note: not creative commons licensed! [17:53] http://wiki.debian.org/DebianEeePC [17:53] the debian wiki, of course [17:53] http://forum.eeeuser.com [17:53] ..and the eeeuser global forum. [17:53] < leftyfb> QUESTION: not really a question, but a uselful link: http://wiki.eeeuser.com/getting_ubuntu_8.04_to_work_perfectly [17:53] there you go. [17:54] also useful, this place are selling EEE 900s now :) [17:54] http://www.laptopsdirect.co.uk/Asus_Eee_PC_900_Linux_Intel_Mob_1024MB_-_White_EeePC900-W005/version.asp [17:54] if you have any additional links *please* add them to the community wiki. [17:54] yeah, I guess the 900 is out now in the UK. /me is jealous of popey [17:54] ordered two already :) [17:54] I believe the 900 will be available in the US in another two weeks. [17:54] likely available at newegg, amazon, etc. [17:54] any other questions? [17:55] not from -chat [17:55] great session Zelut, thanks! [17:55] again, please contribute to the wiki or contact me (christer.edwards@ubuntu.com) if you'd like to help improve ubuntu support on the eeepc! [17:55] I guess that'll do. Thanks for coming guys. Appreciate it. [17:56] \o/ [17:56] * Zelut waves [17:57] https://wiki.ubuntu.com/MeetingLogs/openweekhardy/EeePC log for that last session, and marks https://wiki.ubuntu.com/MeetingLogs/openweekhardy/AskMark === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Preseesing d-i / Ubiquity" - Evan Dandrea [17:58] evand: ok, you're up! [17:59] hooray [17:59] Ok, I'm going to be talking about preseeding as a means of automating the alternate and desktop installers. [17:59] Before I turn this into a Q&A, I am going to briefly speak about the current status of automating the installer. [18:00] I'd also like to remind you that many common questions can be answered by reading through the installation guide, specifically the preseeding section (Appendix B) and the wiki page on preseeding ubiquity (https://wiki.ubuntu.com/UbiquityAutomation). [18:00] You can read the latest version of the installation guide by typing: [18:00] sudo apt-get install installation-guide-i386; w3m /usr/share/doc/installation-guide-i386/en/index.html [18:00] Note that this will be the version of the installation guide for your release. [18:01] So if you're still on 7.10, you'll get the 7.10 guide by doing that. [18:01] There's an older version of the guide on help.ubuntu.com as well. [18:01] Ubuntu uses a modified version of Debian's debian-installer system as the basis for our alternate CD. [18:02] Preseeding in debian-installer is a way of answering questions that you know the installer will ask for the purpose of reducing the amount or type of input the user needs to provide or for completely automating an installation for mass deployment. [18:02] For example, if I wanted to avoid asking what username to use for the default user on the system, I would create a file and put the following in it: [18:02] d-i passwd/username string evan [18:03] I could then place this file in the initrd as preseed.cfg, or I could reference it on the kernel command line. For example: [18:03] url=http://my.server.com/preseed.cfg [18:03] Would tell the installer to fetch the preseed file from my webserver. [18:04] Note that you can also preseed questions directly on the kernel command line. For example, I could've just put this on the kernel command line and avoided using a file altogether: [18:04] passwd/username=evan [18:05] While there isn't a canonical list of possible questions, the installation guide should provide you with all of the questions you need. Beyond that, you'll have to look at the "templates" file in the debian directory of the source package for the d-i component in question. [18:06] For those of you coming from a Red Hat environment, note that kickstart files can be preseeded into an install by passing them via the ks= kernel command line parameter. https://help.ubuntu.com/community/KickstartCompatibility has more details. [18:06] I should note that my experience with that bit of code is minimal. Though I'll try to answer any questions on kickstart to the best of my ability. [18:07] I should also mention that I've noted the kernel command line a couple of times. It can be accessed by pressing F6 at the isolinux menu on the CD. [18:08] Ok, so the desktop CD installer can be preseeded as well. [18:08] Ubiquity, the live CD installer, put simply is actually a layer of GUI code on top of debian-installer. Ubiquity takes the data the user enters on its pages and preseeds it as the answers to the relevant debian-installer questions. [18:09] Each ubiquity page maps to one d-i component, with the exception being the partitioner, which maps two pages to the partman components. [18:10] Again, this is greatly simplifying what is actually happening, but I only mention it to explain the link between the two installers. *Most* of the questions you can preseed in d-i can also be preseeded in ubiquity. [18:10] The exceptions to this are d-i components that ubiquity does not use and questions that ubiquity provides similiar functionality to, but under a different name. [18:11] One such example of the latter is ubiquity/success_command, which will run a command or shell script upon successful completion of the installation. This of course is similar to preseed/late_command. [18:11] Another example would be ubiquity/reboot, which just as the name suggests, skips the final installation completed dialog and triggers a reboot of the system. This is roughly equivalent to preseeding finish-install/reboot_in_progress. [18:12] Also note that if you want to skip the usplash "please remove the CD and press any key to reboot" message, you'll need to add 'noprompt' to the kernel command line. [18:13] Now, by default ubiquity will still ask questions that you've preseeded the answers to. This can be avoided by either passing --automatic to ubiquity in a terminal window, or adding ubiquity-automatic to the kernel command line. [18:14] Note that ubiquity-automatic gives you the same stripped down enviornment that the new "Install Ubuntu" menu option offers. [18:14] Once you do this, any page that has had all of its questions answered correctly will not be shown. An invalid answer, such as selecting cupsys for the username, will result in the page being shown with an error message and the script or command preseeded in ubiquity/automation_failure_command being called. [18:15] FWIW, there is also a generic ubiquity/failure_command preseed key that is called whenever the install crashes. [18:16] Also worth noting is that a failure to preseed one page properly does not mean the rest of the install will be manually done. [18:16] Once the user corrects the error and presses next, it will continue along until it encounters another page with an unanswered question or starts to install. [18:17] If you preseed all of the questions correctly and preseed ubiquity/reboot and add noprompt to the kernel command line, you'll have a completely unattended installation. [18:17] This is what Wubi, the Windows Ubuntu Installer, does. Its code, preseed file included, lives at https://code.launchpad.net/~ubuntu-installer/wubi/hardy . [18:18] One last point -- what to do when things go wrong. If the installer is still showing a page despite all of its questions being preseeded, run the installer in debug mode (ubiquity --debug). [18:19] This will tell the installer to print the communication with debconf to /var/log/installer/debug. You can then search the log file for "question will be asked". [18:19] This can also be handled on the kernel command line via the debug-ubiquity argument. [18:20] Ok, I think that about covers it. I'm going to open this up to questions. If you run into trouble later or we run out of time before you can ask, the ubuntu-installer mailing list is an appropriate forum for preseeding questions. We also have an IRC channel, #ubuntu-installer. [18:21] Do I have someone to help me paste questions? [18:21] Omahn> QUESTION: In the PXEInstallServer wiki page it mentions using PXE with Kickstart. Is the same PXE installation method compatible with preseeding? [18:22] evand: (I can for a bit here) [18:23] Yes, just replace the ks= with file= and the location of the preseed file. [18:23] tonyyarusso: much appreciated. Let me know when you have to leave and I'll take over. [18:23] next. [18:23] Zelut> QUESTION: How much work is being done on the kickseed (kickstart to preseed) compatibility? I'd argue kickstart syntax is more "human readable" than preseed, which fits better into the Ubuntu ethos. [18:24] Zelut: While it's more human readable, preseeding is built into the core of the installer, so kickstart compatibility will always be a matter of translation. [18:24] Colin Watson has done most of the work on the kickstart compatibility and is always quick to fix any bugs that arise. [18:24] Should you find an issue with kickseed I strongly encourage you to file a bug. [18:25] next. [18:25] tonyyarusso> QUESTION: Are there any special concerns when pre-seeding with the OEM install option? [18:25] I'm not sure I understand the question. You have to enable oem mode (oem-config/enable=true), but beyond that it should be pretty straightforward. [18:26] Note that there were a few bugs in 8.04 for this... [18:26] What would those bugs be filed against? [18:26] In particular the default language choice was not being carried over to the oem dialog. I *think* we fixed that, though these kind of bugs often manage to find a way back into the code. :/ [18:27] Either against ubiquity or oem-config. [18:27] 'k [18:27] Omahn> Not sure if this could be answered here or not, but would commercial support from Canonical support the PXE and preseeded install process? [18:27] Omahn: I *believe* so, but you would have to contact Canonical support for a definitive answer. [18:27] next [18:27] tonyyarusso> QUESTION: Can preseeding be used to specify packages to install/remove (like dpgk --set-selections) ? [18:28] tonyyarusso: Yes, pkgsel/include. See section B.4.10 of the install guide for more details. [18:29] Omahn> QUESTION: Is it possible to do a manual install and then 'retrieve' a log of preseeded options to use in future? [18:29] Ooh, that's brilliant. [18:30] Omahn: debconf-get-selections [18:30] So, yes. [18:31] Note that you'll have to do that from the install environment and to my knowledge there's no easy menu item for doing it and pushing it somewhere. [18:31] File a wishlist bug if you believe there should be. [18:31] next. [18:31] Zelut> QUESTION: (building on Omahn's question). The way anaconda auto-builds a .ks at installation would be *awesome* with Ubuntu. [18:31] That's more of a statement Zelut :) [18:31] ^ File a bug and I'll see what I can do :) [18:32] next [18:32] That's all I have at the moment - watching for any last minute ones. [18:33] While we're waiting, I never did formally introduce myself... [18:33] I'm Evan Dandrea, as you may have gathered I work on the installer, ubiquity specifically but d-i to a lesser extent, at Canonical. [18:34] I'm also the author of migration-assistant and have been the Ubuntu end of Wubi's development over the 8.04 cycle. [18:35] If you have any questions, feel free to email the mailing list, or if you don't feel that's appropriate, email me directly at evand - ubuntu - com. [18:35] The mailing list is ubuntu-installer@lists.ubuntu.com [18:35] evand: Is cjwatson still the d-i guy, or has he succeeded in passing that on? [18:36] tonyyarusso: Colin still handles much of d-i, but over the longer term I believe I'll be responsible for most of it. [18:37] If anyone is interested in getting involved in hacking on any part of the installer or the surrounding code (Wubi, Mythbuntu's installer code, etc), please see the wiki: [18:37] http://wiki.ubuntu.com/InstallerDevelopment [18:37] As mentioned there are a few wiki pages on preseeding, I'll try to organize those better going forward. [18:37] Debian's wiki is also a great source for information: [18:37] http://wiki.debian.org/DebianInstaller [18:38] I think that covers it. Thanks to all for your time. Thank you tonyyarusso for fielding questions. [18:42] Ok everyone, next session in ~15 minutes [19:01] All right bryce, it looks like it's time to go. [19:02] Bryce will be talking about the X Windows System, which provides the graphical environment in Ubuntu. [19:02] hi all! [19:02] The title of this talk was to be "Make X Kick Ass" but I guess that didn't pass the censors. ;-) [19:02] X is involved in everything the user sees or touches [19:02] Yet X is probably one of the most ignored parts of the desktop! [19:03] You can make a *huge* impact on Ubuntu's quality by improving X in two ways: Bug squishing, and coding up features. [19:03] In this session, we'll talk about how to contribute to these areas, and to help make Ubuntu's X kick more ass. [19:03] The Ubuntu X team has a mailing list you can subscribe to (low traffic): [19:03] http://lists.ubuntu.com/mailman/listinfo/Ubuntu-x [19:03] Most of our work and discussions occur on IRC: [19:03] FreeNode: #ubuntu-x [19:03] You can join the Ubuntu-X bug team here: [19:03] https://launchpad.net/~ubuntu-x-swat === tonyyarusso changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask question in #ubuntu-classroom-cht, prefaced with "QUESTION:" | See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Making X Rock!" - Bryce Harrrington [19:04] For those who don't know me already, I'm Bryce Harrington and work on Xorg and related components. Timo Aaltonen, Tormod Volden, and unggnu are very active community members who do a huge amount of bug and package work. [19:04] we're always looking to welcome new people to the team too! :-) [19:04] First up is bugs. There's basically four ways to contribute to bug work: [19:04] a) reporting - https://wiki.ubuntu.com/X/Reporting [19:05] b) triaging - https://wiki.ubuntu.com/X/Triaging [19:05] c) researching [19:05] d) fixing [19:05] All of these are important, but they require different levels of experience. [19:05] I'm mostly going to focus on (c) and (d) here since the first two have been thoroughly covered in previous sessions, but first a few brief comments on reporting and triaging for X. === tonyyarusso changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask question in #ubuntu-classroom-cht, prefaced with "QUESTION:" | See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Making X Rock!" - Bryce Harrington [19:05] For reporting X bugs the #1 most important thing to remember is to ALWAYS attach your Xorg.0.log. [19:06] This file is thick with useful info like config settings, module versions, error messages, etc. [19:06] The #2 thing to remember, is to PICK A GOOD TITLE. [19:06] Too often people use a generic title like "X crashes randomly" or "Black screen after startup", and then everyone and their cousin thinks they have the same bug, when really they just happen to have similar symptoms. [19:06] The result is much gnashing of teeth, people not getting their bugs fixed, cats and dogs living together.... [19:07] A lot more info on making good bug reports is at https://wiki.ubuntu.com/X/Reporting [19:07] Regarding bug triaging, I want to thank everyone who has helped in triaging X bugs. [19:07] This is important work, and every hour that you put in saves a developer an hour that they can use to focus on *fixing* the dang bugs. [19:07] Triaging is perhaps the easiest way to get involved in Ubuntu X; there's plenty of bugs, and you can start with little technical know how, and just learn as you go. Pedro gave an excellent talk on this earlier in the week. [19:08] https://wiki.ubuntu.com/MeetingLogs/openweekhardy/TriageBugs1 [19:08] I've outlined some triaging "projects" for X here: https://wiki.ubuntu.com/X/Projects [19:09] Since there's so many bugs, I find it useful to set myself a goal, so I know when I've "finished". [19:09] For Hardy one of my goals was to reduce the total bug count for -intel to <100 by the release. [19:09] that was tougher than I expected, but we did make that goal in time :-) [19:09] For Intrepid, I want to do the same for -ati, which has around 170 currently [19:10] As a more aggressive goal I want to try reducing the total Xorg bug count from 1600 to 1000, but we'll see there - that'll be quite tough. [19:11] I have some plots here to keep track of bug progress - http://people.ubuntu.com/~bryce/Plots/ [19:11] look at the 6-month chart for Intel under X, for open bugs, and you can see how the bug count got reduced during Hardy development [19:11] ok, maybe a few questions at this point? [19:12] would someone mind pasting a couple? [19:12] progfou> QUESTION: what about making X start and stop faster too? does Ubuntu collaborate with Fedora on this project? see http://fedoraproject.org/wiki/Features/OneSecondX [19:12] yep, we keep close eyes on what other distros are doing and have been looking at that one in particular [19:13] a community member dug into that and found a few patches which are low risk, that we might pull in for 8.04.1 [19:13] bryce: (say next when done answering) [19:13] I'll talk a bit more about opportunities for working more closely with upstreams and other distros - this is going to be largely driven by our Ubunt-X community size [19:13] ok, next [19:13] gkatsev> QUESTION: i have problems with full screen video, mostly in flash, does this have anything to do with X, if yes, any ideas how to fix? [19:14] it may indeed be an X issue, but there are several layers in the stack that could be at fault with video issues [19:15] that is actually an excellent question to take us into the next section of the talk, so let me head on into that [19:15] the next step after triage - once you know you have an issue such as "video is failing, maybe flash related?" or whatever, is Research. [19:16] we've got a LOT of bugs like this, where we know a bit of basics about the problem, but don't know how to fix them. [19:16] 1600 Xorg bugs at last count, in fact [19:16] But I've noticed that non-developers can play a huge role in making these bugs solvable, without writing a line of code. [19:16] Here are several techniques I've noticed as effective. [19:16] a) Look for patches in other distros [19:16] b) Look for patches upstream in Xorg [19:16] c) Look in appropriate mailing lists for discussions about the problem [19:16] d) Test the latest upstream version [19:16] e) For regressions, test older versions [19:16] f) Get a backtrace [19:16] g) Talk to an upstream person about the bug [19:17] Providing pointers for anything found (even if you're not certain it's relevant), can help a lot. [19:17] People can try out and verify the fix, and packagers can (usually) easily grab the fix and package it up. [19:17] Testing older or newer versions of X or the driver requires pulling them via git and compiling them [19:17] which sounds technically challenging so probably many don't try it [19:17] but this can be extremely effective at locating a patch to fix the problem. [19:18] (I have some plans to help folks with pre-building git versions of drivers, but need to do more development on that before I can push it. I did a prototype for -intel last month that proved quite effective.) [19:18] (http://people.ubuntu.com/~bryce/bisect/) [19:18] If it works with the newest upstream version, then there may be a patch in the new version we could backport. [19:18] If you find it worked with older versions but does not work in the latest, then this may identify the patch that broke things, which we could then revert or fix. [19:19] Backtraces are awesome. X crashes are some of the most critical kinds of bugs we see. [19:19] Similar to crashes are lockups and freezes. [19:19] In all these cases, getting a backtrace can be extremely illuminating in identifying where X stopped. [19:19] I'll talk more about backtraces in the next segment. [19:19] Directions on obtaining backtraces this are available at this page, so I won't bore you with a lot of details, since this gives a nice paint-by-numbers approach: https://wiki.ubuntu.com/X/Backtracing [19:19] So next time you crash X, don't just reboot! Instead, grab a second computer and ssh into the broken one, attach gdb, and get a full backtrace. :-) [19:20] Okay, let's take more questions, particularly about bug reporting, triage, troubleshooting....? [19:20] tonyyarusso: fire away with the next one [19:20] rodolfo> QUESTION: so far we've seen X and Intel's video cards not working 100%. There are rumors that this will be changed until 8.10 official release. Is this correct? (specifically i9xx series) - perhaps tips on tracking down the bugs and subscribing? [19:21] right, we've been working quite closely with Intel this past release on squishing a lot of bugs [19:22] I mentioned earlier about the reduction of -intel bugs from 180 to 100 - obviously this leaves a lot still to fix. But we plan to continue work on getting the bugs solved. [19:22] Intel has been *great* at responding to bug reports forwarded upstream to bugs.freedesktop.org [19:23] they have very specific expectations about the bugs forwarded there though, so this is why I emphasize the bug triaging procedure so much - collecting all the right info, researching, backtraces, testing current git, etc. all are necessary for upstream to proceed with finding a fix [19:24] we've got good processes for -intel and good momentum, and I'm quite hopeful to see this result in 8.10 being a very very solid -intel release. And it'll be even better with your involvement. [19:24] tonyyarusso: next please? [19:24] qense> QUESTION: Does intel have an own public bug tracker? Or are their employees just reading other bug trackers? [19:24] Intel's employees read the Xorg bug tracker at bugs.freedesktop.org [19:25] they don't have their own outside that. [19:25] next [19:25] gruber> QUESTION: What should people who complain about resolution be told to do if X can't probe for their monitor type or frequency data--how should the new user supply it? [19:25] I like that question :-) [19:25] in fact, things have gotten a LOT better since Feisty, when resolution problems were mentioned in *every* Ubuntu review [19:26] I've got some very detailed directions on solving resolution issues on the X troubleshooting page in the link mentioned above, but a few quick tips: [19:26] first, I always look in /var/log/Xorg.0.log for the EDID section, and then look at what the monitor is reporting, what the card says it's capable of, and what the result is. About 50% of the time this gives a clue. [19:27] many of the common causes for resolution issues in the past have been eliminated. One of the biggest remaining ones is monitor or vid-card breakage, which is treated via Quirks. I'll talk more about quirks later. [19:28] another issue is when the monitor simply won't report edid at all, for whatever reason [19:28] if we can't even identify the monitor, quirks aren't going to help unfortunately. [19:29] there is a tool, displayconfig-gtk, which we no longer use as a stock config tool since it is applicable for only a small number of situations, however it has a good monitor database [19:29] be careful in using it though, and back up your xorg.conf, as it will be (attempting) to modify it [19:29] next [19:29] rzr> QUESTION: does it sound possible to run a repository of upstream snapshot built ? (for testing) (an apt-repo, I assume, PPA or other) [19:30] right, that's essentially what my goals are with the bisect page aboe [19:31] I can't do it with PPAs unfortunately since they only let you keep the newest versions [19:31] next [19:31] rodolfo> QUESTION: For Linux newbies, updating the driver sounds like a painful process. Specially when it's related to X. Maybe because of each file requirements. So, is there a guide for first-timers to update it or something like that available (a self-explained doc or tutorial)? [19:31] indeed there is [19:31] standby [19:32] here we go - https://wiki.ubuntu.com/XorgOnTheEdge [19:32] we have some scripts to help automate building of .debs and such there [19:32] and tormod provides some prebuilts [19:33] I also hope to expand on this during Intrepid to make it easier, since this is such a great way to troubleshoot problems, but also know people like to avoid compiling stuff (yet compiling builds character!) [19:33] ok, I'm going to continue on with the talk [19:34] Okay, now for the hard-core stuff on smashing bugs. Well, "hard-core" is probably a bit of a stretch... What I'm going to describe are more at the "apprentice developer" level - you need to know C, but not a lot of depth into X itself. [19:34] Since we just talked about backtraces, let me explain really briefly what a developer does with them. [19:34] Most crashes are because of bad pointers. Pointers are basically just addresses to places in memory. [19:34] Pointers can go bad in several different ways. [19:34] The most common are NULL pointer dereferences. Basically, pointers are set to 0 (aka NULL) to show they're not valid. If you try to access something using a pointer with value 0, that's a serious error and causes the program to fault. [19:34] You can usually spot NULL pointers right in the backtrace - some variable set to 0x0. [19:34] Another common pointer error is using an uninitialized pointer, that weren't set to NULL and instead point to a random place in memory. These are harder to spot, and usually require studying the source code. [19:34] If you know C and are comfortable with pointers, this can even be kind of fun. :-) [19:34] (well, I find it fun, maybe I'm crazy) [19:35] If you don't know C, don't worry about all the pointer jibberish. The important thing to remember is: If it crashes (or locks up, or freezes, or boots you back to the login screen) get a backtrace. [19:35] [19:35] Another apprentice-level X developer task are Quirks, like I mentioned earlier. [19:35] Quirks are basically hardware-specific fix-ups, and can be specific to a monitor, or to a graphics card. [19:35] When X starts up, it gathers the 'EDID' data from your monitor to get your monitor's id, and then looks up if there are quirks for that id to apply. X does the same with your graphics card, using your pci id. [19:35] Monitor quirks are kept in the X server, in the file hw/xfree86/modes/xf86EdidModes.c. [19:35] Here is a listing of the current quirks: [19:36] quirk_prefer_large_60: Detailed timing is not preferred, use largest mode at 60Hz [19:36] quirk_135_clock_too_high: Recommended 135MHz pixel clock is too high [19:36] quirk_prefer_large_75: Detailed timing is not preferred, use largest mode at 75Hz [19:36] quirk_detailed_h_in_cm: Detailed timings give horizontal size in cm. [19:36] quirk_detailed_v_in_cm: Detailed timings give vertical size in cm. [19:36] quirk_detailed_use_maximum_size: Detailed timings give sizes in cm. [19:36] quirk_first_detailed_preferred: First detailed timing was not marked as preferred. [19:36] quirk_detailed_sync_pp: Use +hsync +vsync for detailed timing. [19:36] At first glance it's not obvious what these mean, but basically they deal with issues where the monitor manufacturer didn't encode EDID properly (like putting values in centimeters rather than millimeters, etc.) [19:36] EDID == Extended Display Identification Data [19:37] basically it's a magical chunk of binary data that is embedded in your monitor, that your computer can query plug-n-play-ishly [19:37] If you're curious about the particulars of any of these quirks, if you look in the source of that file it references bug ID's that explain the problem in better detail. [19:38] [19:38] Graphics cards also have quirks. These are stored in the video driver. [19:38] For example, with Intel graphics, this is in xserver-xorg-video-intel in the file src/i830_quirks. [19:38] Here's the current list of available quirks: [19:38] quirk_mac_mini [19:38] quirk_ignore_tv -- the most common quirk [19:38] quirk_lenovo_tv_dmi [19:38] quirk_ivch_dvob [19:38] quirk_pipea_force -- another common quirk [19:38] Again, look for bug reports for explanations of these. [19:38] Believe it or not, us Ubuntu-ers have been the source for most of upstream's X quirk data . [19:38] This is because Ubuntu kicks ass! [19:39] ... okay, really it's because there are more Ubuntu users than other distros, so we just have a more widespread coverage of desktop hardware. [19:39] But quirks are a good way for us to contribute upstream, and so it's important we recognize bugs that just need quirks made for them. [19:39] tonyyarusso: ok time for a few more questions... next? [19:39] phoenix24> QUESTION: Would you give an overview of the X architecture or Design ? [19:39] (briefly, I'm sure) [19:39] hehe [19:40] ok, well veerrry briefly, the whole X stack is broken into a client / server model [19:40] the server is, obviously, xserver (although some people run xgl which is basically an alternate server) [19:41] clients are basically anything that runs on X - gui apps, cmdline tools like xrandr or xdpyinfo, window managers, etc. etc. [19:42] there are various protocols that these use to talk to one another. An example many have probably heard of recently is the XRandR protocol [19:42] xlib (recently replaced by xcb) provides the interface between them [19:43] the xserver itself is mainly a big chunk of code, with several libraries broken out, and a series of different drivers [19:43] there are two kinds of drivers - input and output. Input are keyboard, mice, trackpads, etc. Output are basically video cards. [19:44] by and large, most of the stuff we find "interesting" is either in the xserver or in one or more of the drivers [19:44] hopefully that gives a sufficient overview [19:44] tonyyarusso: next please? [19:44] progfou> QUESTION: does the Ubuntu-X team has plan to make X.Org run as a non-root user to make bugs less critical on a security point of view? or do we have to wait for upstream to implement something missing first? [19:46] progfou, upstream is working on making Xorg run as non-root; there are a variety of projects under way upstream like this [19:46] the level of participation is driven largely by the amount of community participation in the Ubuntu X team, and obviously by their own personal itches they wish to scratch [19:47] worst case, we can pull in stuff from upstream as its ready, but it would be preferable to have people involved in assisting with pulling the bits in to test earlier [19:47] next [19:47] mariusss> QUESTION: Hi Bryce! With the latest changes in X and the video drivers (nvidia in my case) everything started to be more automatical (configuration of the monitor, video card, etc). But, are you aware of the fact the some of the latest video cards (expensive ones such as [19:47] GeForce 9800GTX) or some of te worst (most of the on-board ones) are a pain in the a** to make them work in Hardy (twith or without a video driver installed)? [19:47] I joke with Kees that the easiest way to make brand new hardware work is to stick it in your closet for a few months. ;-) [19:48] yes, it has always been true that the newest hardware tends to not be as well supported as older stuff [19:49] we've been doing really well with Intel due to our level of interaction with them, and I hope with ATI's recent increase involvement with the open source community that we'll see big improvements there too (we already have a little) [19:50] as a general rule, the best way for these is to get good bugs reported on them, and forward upstream to Xorg. [19:50] ok, let me continue on with the talk [19:50] We can divide feature coding into three categories: [19:50] a) X config tools [19:50] b) Packager tools and scripts [19:50] c) Core X and driver coding [19:51] X configuration tools are things like displayconfig-gtk or the new Screen Resolution tool, but could also include command line or install-time tools like xresprobe, bulletproof-x, xfix, etc. [19:51] It is important that work on X config tools be done in a manner that upstream will accept, so we can contribute them and benefit from their assistance in maintaining them. [19:51] GNOME and Xorg prefer tools be written in C, for example. [19:51] [19:51] One config tool we will need in the coming year is an GUI XInput Config tool. [19:51] This would allow run-time configuration of mice, keyboards, tablets, and other input devices. [19:51] To start, we need it to be conceptualized - what screens it should include, how they should be laid out, etc. [19:51] Prototypes could be done in Python, but ultimately what we ship will need to be written in C. [19:51] Let me know if this is something you'd like to work on. [19:51] [19:51] There are also a number of improvements that could be made to Screen Resolution, which I've listed along with some other projects here: [19:51] https://wiki.ubuntu.com/X/Projects [19:51] [19:51] Finally, let's talk a bit about core X development work (the Advanced projects at the above url). [19:52] This touches on the fast X boot question. [19:52] Historically Ubuntu has contributed only lightly on core X - mostly bug fixes and fringe work, [19:52] but I know there are some very smart Ubuntu-ers coming on board and looking for respectible challenges, and core X development certainly fits! [19:52] [19:52] You can read elsewhere about what is going on upstream that we could [19:52] participate in. The recent XDC2008 notes are the best place to start: [19:52] http://www.x.org/wiki/Events/XDC2008/Notes [19:52] One of the Ubuntu-X team's top priorities for Intrepid is Input Hotplug. We'd love to see additional testers get involved in this. [19:52] Some other hot topics include kernel modesetting, fast X boot, TTM memory management, redirected direct rendering for GL/Xv, and multi-screen infrastructure. [19:52] These are all under way upstream or at RedHat, and will come to us eventually. [19:52] But one way to accelerate when we see them on Ubuntu would be to try building and testing them on Ubuntu, identify and fix problems, and participate upstream in their development. [19:53] [19:53] Some other areas not under way upstream as far as I know, but that would be cool to see for Ubuntu include: [19:53] Multi-pointer X (MPX) - for using tablet and mouse in conjunction. Useful for drawing apps and maybe games. [19:53] OpenGL with KVM - need to evaluate and/or port VGML to Ubuntu [19:53] OpenGL performance for gaming - would be great to help optimize 3D for OpenGL games, to make it more robust and smoother under Ubuntu. [19:53] Phoronix published a perf test that would be interesting to run and examine on Ubuntu. [19:53] [19:55] I don't want to set false expectations that we have definitive plans for working on any of these, except for Input Hotplug, but to list them as areas of possible involvement [19:56] Here again is where to subscribe to join the Ubuntu-X project: [19:56] http://lists.ubuntu.com/mailman/listinfo/Ubuntu-x [19:56] FreeNode: #ubuntu-x [19:56] https://launchpad.net/~ubuntu-x-swat/ [19:56] tonyyarusso: next questions? [19:56] rzr> QUESTION: what are priority for ubuntu if you have to choose between hw support or reliability or performance ? (or features) [19:57] for an LTS release, reliability is the highest priority, with hw support second and performance third [19:57] for a non-LTS release, I'd probably set the three to roughly equal in importance, but with increased importance on reliability as the release approaches [19:57] next [19:57] that was the last question [19:57] excellent! [19:57] Thanks for filling us in on X Bryce! [19:57] bryce: bumping up against the next one now. [19:58] bryce: There were more questions though - should they join you in #ubuntu-x for those? [19:58] ok, well again, please join us in #ubuntu-x if you'd like to know more, or join in [19:58] yep [19:58] sounds good [19:58] Thanks bryce ! === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "The Future of the Ubuntu Desktop" - Ted Gould [19:59] Next up is Ted Gould, with "The Future of the Ubuntu Desktop" [19:59] This one is going to be awesome. :D [19:59] Uh, oh, now that's pressure :) [20:00] ted1: go ahead and begin! [20:00] Ah, okay. [20:00] Welcome everyone. [20:00] My name is Ted Gould and I'm on the Ubuntu Desktop team. [20:01] Today I'd like to talk a little about the team, and in general about the Ubuntu Desktop. [20:01] One of the things that I think it is interesting to see about the Ubuntu Desktop is that it is really an instance of the GNOME desktop. [20:02] But, yet has it's own identity, and that difference is something that we find very important. [20:02] That doesn't mean that we don't like GNOME, but it means that we don't have to be a straight up copy of GNOME either. [20:02] In general, we've been pretty quiet for this last release as we've viewed the LTS as an opus for the current ideas, not a place to try new ones. [20:03] So for Intrepid we'd like to focus on getting some new things out there, and to start planning for goals for the next LTS. [20:03] Which we plan to be another great work, culminating all of the ideas that have come together over that time. [20:03] I like to say that we're planing for 8.10 and 10.8 at the same time :) [20:04] One of the key places to go for information on the Desktop Team is our wiki page: [20:04] https://wiki.ubuntu.com/DesktopTeam [20:04] (just pasted that in the wrong chat, oops :) ) [20:05] There we want to start working on planning for the next LTS, and looking at this through the lens of designing user exeriences. [20:05] So, we want to focus on how a user accomplishes a task rather than the tools that are used to do that. [20:06] You can find some of the experiences that have been written up so far here: https://wiki.ubuntu.com/DesktopTeam/Experiences [20:06] While those are the beginning, we expect define and work on those at both UDS and beyond. [20:07] I imagine they won't be 100% complete until the next LTS is shipped :) [20:07] Questions on anything so far? [20:07] < Solarion> ACTUAL QUESTION: How will RedHat and SuSE's ignoring of the consumer desktop work with Canonical's support-based business model? [20:08] Solarion: Well, I think that it comes down to Ubuntu and RedHat/SuSE look at the business very differently. [20:08] We're all basically using support based business models, but it comes down to how you handle the ratio of how much money you make per customer, and base your expenses off of that. [20:09] So we're more focused on making $1 off every Ubuntu user (buy a T-shirt ;) ) while they're more focused on getting $1000/year corporate contracts. [20:09] Their model doesn't work for the consumer market. [20:10] So, I think they'll continue to do what they do best, and we'll continue to innovate on the consumer space. [20:10] jcastro: next [20:10] multiple question [20:10] QUESTION: Hi Ted! Do you think this -> http://www.silverwingsonline.com/ubuntu1.jpg will become reality any time soon? It will be nice to have such a futuristic desktop ;) [20:10] and [20:10] QUESTION: Will AWN be default in Ubuntu desktop on the next release, 8.10? [20:12] I don't think that we have any specific goals right now, or a definite look. [20:12] I would say that looking at both of those we have a lot of the same ideas, so I wouldn't be surprised if what we eventually ship has many of those ideas. [20:12] One of the things that we want to focus on is being able to support smaller and larger displays well. [20:13] Which isn't currently done. [20:13] A good example of that is how the panel resizes. [20:13] https://wiki.ubuntu.com/DesktopTeam/PanelResize [20:13] As you can see in many of those screenshots, things look pretty bad as the panels get larger. [20:13] But, most of it isn't the panel's fault. [20:14] What happens is that many of the panel applets or applications with notify icons don't handle the resize events correctly. [20:14] Or, in some cases they don't ship with appropriate icons. [20:14] And that's something we'd definitely like to fix. [20:15] And I think that segways to a lot of how we see the desktop team interacting with upstream projects like GNOME. [20:15] Our goal isn't to be the "Project X" upstream, it's to help all of the upstreams that are coming to the Desktop. [20:15] < Solarion> QUESTION: How far are you willing to drift from upstream? [20:16] So if we could have a patch to all of the apps we ship that implemented resizing the icon, that would be great. [20:16] I think drift is an interesting thing there. [20:16] There are practical limits, in that we're just not that big. We simply can't replace all of the upstream development that is done. [20:17] But I could see us doing things like perhaps replacing the panel. [20:17] I don't see that happening real soon as upstream is interested in redoing that also, but I think it shows the size of what we could accomplish well. [20:17] Not a huge amount of code, but a significant visual impact. [20:18] For instance one of the ideas that has been floating around on the art side is to try and make the panel icons dual-tone, and then use color for notification. [20:19] For example your battery level could be black, but then if it becomes critical it turns red. [20:19] But, this would require a patch, in that the panel is not specifically themeable. [20:19] We would prefer if upstream would take that so we wouldn't drift at all, but if they don't, we could maintain it in our packaging. [20:20] jcastro: next? [20:20] < tonyyarusso> QUESTION: Does Ubuntu have any plans to look into more radical things like BumpTop, and if so, will they be helping to develop [an open source version of] them? [20:20] http://www.bumptop.com/ [20:21] Well, we're trying to be radical in a non-radical way :) [20:21] I don't know if you're familiar with MacSlow's project (I can't think of the name) [20:22] But basically it is a visual way to move files and group them using OpenGL and a very graphical way of moving the mouse. [20:22] We'd love to see that integrated into a nautilus view, but still provide the other nautilus views for people with older machines and who prefer a more familiar experience. [20:23] While we love bling, we need to keep in mind that one of the selling features of Linux has always been it's ability to run on lower end machines. [20:23] We can't loose that advantage, or atleast we need to provide a good experience for those who don't have the latest and greatest hardware. [20:23] jcastro: next [20:23] < Solarion> QUESTION: Are you planning to do some of the cool things that can be done with Empathy/Telepathy? [20:24] Yes! [20:24] I hope that we can do a lot more of that. [20:24] But, I dont' think that we've decided on exactly how, but we really want to get the accounts integrated into the desktop. [20:25] The reality is that people's contacts are more important that religrated to a specific application. [20:25] I think taht social networking sites have shown us that. [20:25] jcastro: next. I think everyone is going through my outline :) [20:26] heh [20:26] < artir> QUESTION: Are you planning to have only 1 panel? [20:26] and related: < eragon> QUESTION: Will Gimpie panel be integrated in the next release, or some other time? [20:27] he means Gimmie [20:27] Lots of panel/dock questions today! [20:28] One panel is an interesting idea, it comes down there are four corners of the display that are very important as they are easy to get to. [20:28] So, while we may not have a "panel" per se, we want to make sure that those are all used for something very useful for every user. [20:29] Today, we have "Applications", "Quit", "Show Desktop" and "Trash" in those corners. [20:29] Now those are all important, but taking up the horizontal space to make those corners useful, isn't necissarrily something we need. [20:30] So I think you'll see us move away from "2 panels" but we'll still use "4 corners" [20:31] Gimmie, that is definitely interesting, and I think that you'll see some of the ideas from it being used. [20:31] But I'm not sure that you'll seem Gimmie directly. [20:31] One of the beautiful things about Gimmie is that it puts what you're using right now directly available. [20:31] And we'd definitely like to see optimization on the "current task" throughout the desktop. [20:32] jcastro: More questions? Or go through my outline some more? [20:32] outline for a bit [20:32] Sounds good. [20:32] then I will put in more questions [20:33] One of the things that Mark talked about in his Intrepid announcement is that we want to see more of a focus on networking for Intrepid. [20:33] One major part of that initiative is to start looking at things like Telepathy and Empathy, and trying to get them into the desktop more. [20:34] It is looking like they might become part of GNOME directly, which makes it easier for us, but I think that we'll start looking at them seriously for Intrepid irregardless. [20:34] Another thing that effects networking on the desktop is the GNOME project moving over to GIO. [20:35] GIO replaces the old GNOME VFS, and provides a cleaner interface for most applications to access things like network shares. [20:35] Or even odder devices like OBEX transfer to your phone. [20:35] But, that move is not complete. [20:36] So the reason I'd like to talk about it here, is that if there are people who want to contribute to the Ubuntu Desktop from a development perspective, this would be a great way to help us achieve our goals of seamless networking. [20:36] You can get information on the GIO Port: http://live.gnome.org/GioPort [20:36] And instructions on how to do it: http://library.gnome.org/devel/gio/2.15/ch15.html [20:37] What is important to realize here is that even though those links are very focused on the GNOME project, we ship many more applications than that [20:37] And for users, they don't understand what is GNOME and what isn't. [20:37] So for people who want to contribute, you don't need to do it directly with GNOME, it's helpful with any GTK+ project that is interested in moving over to GIO. [20:38] Oh, and going back a little... [20:38] One thing I also forgot to talk about regarding experiences. [20:38] One of the things that we'd like to do with experiences is both decide what we want the experience to be, but also record what it is. [20:38] What happens when someone plugs an iPod into a Hardy machine? [20:39] We're trying to start cataloging that information here: https://wiki.ubuntu.com/DesktopTeam/Experiences/Current [20:39] This is another good way for people who want a small project can help with the Desktop team. [20:39] It takes a little longer than you'd expect, because you really need to create a new account with defaults, but what it does is help us determine where we are and where we're going. [20:40] So when we evaluate things like which CD burner we should be shipping with, we can see what we gain and what we loose in a fairly objective manner. [20:41] jcastro: I probably should shut up for a minute and let people ask questions :) [20:41] ok [20:41] < qrdhol> QUESTION: How will Ubuntu balance adding cool (possible complex) new features, yet still appeal to novices, i.e. the human beings. [20:42] We like human beings :) [20:42] That's always a trick, and I think they key here is providing good defaults. [20:42] One of the things that we like to do is provide features like wobbly windows, but spare people from configuring Compiz. [20:43] So if you look at the desktop, there are settings that can be done, but not all of them. [20:43] And so I think we'll continue with that approach, providing reasonable defaults and visible config options. [20:43] < johanbr> QUESTION: I think desktop bluetooth integration (in particular audio) is something that could still be improved. Are there any plans for that? [20:44] But still providing ways to install more configuration options. [20:44] ted1: we're running out of time and still have more questions so please type faster. :D [20:44] johanbr: We would really like that, I don't know of any hard core plans for that. [20:45] Things are getting better with bluetooth, but no, it's still not as good as an OS like OS X. [20:45] jcastro: next [20:45] < Az32> QUESTION: Will be there a specific implementation of such thing like Ubuntu SDK? In this way the developers would have the possibility to develop some kind of native-Ubuntu applications. [20:46] Az32: That's an interesting question, I don't think we've thought about that. [20:47] For the most part we're still using all upstream libraries, so there's not anything Ubuntu specific there. [20:47] Now, as far as documentation goes, I think that's a good idea, but not something we've done. [20:48] It'd be nice for developers to know what libraries would best integrate with Ubuntu in general. [20:48] jcastro: next. [20:48] < seele> QUESTION: What are some ways you plan on improving these experiences and how can contributors help? [20:49] Well, the easiest answer is come to UDS :) [20:49] But, of course that isn't practical for a lot of people. [20:49] We will be continually discussing and revising them on the ubuntu-desktop mailing list. [20:49] And I'm sure they will be discussed in the Ubuntu Desktop team meetings. [20:49] Those are weekly on IRC in the #ubuntu-meeting channel. [20:50] Thursdays, about 6 hours earlier than this meeting started. [20:50] (adjust for local time :) ) [20:50] I believe that they've been canceled between no and UDS, but they should pick up again right after. [20:50] s/no/now/ [20:51] Basically, show up and make your voice heard! :) [20:51] jcastro: next [20:51] < bryce> QUESTION: What are the top 3 things that Ubuntu Desktop would like to see from Xorg? [20:51] Whew, that's a loaded question ;) [20:51] I think the first one is that we'd really like to see OpenGL work on multiple X instances so that fast user switching really works. [20:52] This helps in the most obvious case of multiple users, but it would also let us do things like keeping GDM on it's own display. [20:52] I'm going to side step the question a little bit and say that number two would be Linux kernel mode setting. [20:52] While not directly an X issue, it is a big user issue. [20:53] The fact that your screen changes as you go from BIOS, to upslash, to GDM -- it sucks. [20:53] Lastly, I think we'd really just like to see good drivers. [20:53] The reality is that most of the things that we want to do can be supported by older video hardware, the drivers just aren't that good for it. [20:54] While a lot of the "eye candy" you're seeing today is flashy, it doesn't require that much GPU power. [20:54] So, it really comes down to good drivers. [20:54] I don't know if that's an X thing or a hardware manufacturer thing though. [20:54] jcastro: next [20:54] < eragon> QUESTION: Why don't you ship advanced desktop effects manager by default? People don't have to use it, since they can also use appearance settings effects tab, but at least it will be available to people in a default installation, [20:54] this will be the last question [20:55] Basically it comes down to having too much choice, you'd be surprised how confused people get. [20:55] There is a fun book on it called "The Paradox of Choice" -- mostly looking at consumer products, but it does apply to UI design also. [20:56] So, while we have no problem with users having those tools, if the majority of users don't use or need them, it doesn't make sense to put them in the default install. [20:56] It just makes unneeded confusing. [20:56] confusion. [20:57] And that goes directly to "support costs" of having things like lots more bug reports. [20:57] So basically it comes down to "don't have to use" doesn't really work. They do use. [20:58] I want to thank everyone for coming. [20:58] And invite everyone to #ubuntu-desktop [20:58] thanks ted! === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Community Support with Ubuntu Forums" - Mike Basinger [20:59] Next up is Mike Basinger ... Community support with ubuntu forums! [20:59] take it away Technoviking! [21:00] Greeting all [21:00] I'm Mike Basinger an admin on the Ubuntu Forums and a Forums Council member [21:01] The Ubuntu Forums starting back in 2004 by Ryan Troy (ubuntu-geek) at a community driven support forum for Ubuntu [21:02] The forum is large community with over 71000 active members in the last month alone [21:03] btw, ask questions at any time [21:03] QUESTION: Are there any plans on providing full-featured technical support from the Ubuntu staff at Ubuntu Forums (a private zone or something)? [21:04] Not as of now, but that is an interesting idea [21:04] < Venko> QUESTION: Are there any plans to switch to a free (as in freedom) forum software? Using proprietary software for Ubuntu discussion seems somewhat hypocritical. [21:04] We do have people who on Ubuntu and Canonical post to the forum regularlly, plus vendors who use Ubuntu (such as Dell) [21:05] Venko: a famous question, we have consider other forum software, but we are afraid they could not handle the load [21:06] the week Hardy came out, we `had over 24000 viewing the forums at one time, quite a load [21:07] Technoviking: venko has further points [21:08] Venko: I feel it is no more hypocritical than Launchpad, to be honest [21:08] vBulletin has been compared to other forum software before and has been shown to be quite inefficient in terms of load. The software is aimed at features and control, not effiecincy [21:09] I do not see the forums using any specific features that available from vBulletin so it seems odd to me, that's all. I used to be involved in forum discussions for another community. [21:09] ok, moving on [21:09] Venko: the features that vBB adds are very popular with our community [21:09] < eragon> QUESTION: how does one become a moderator on the forums? [21:10] eragon: by showing to be a helpful forum member [21:11] when we look for new staff, we ask our current staff who are good people in the forums community [21:12] I would like to talk about asking for and providing support on the Ubuntu Forums [21:13] There are two ways to use the forums, as a guest (view only) or as a member [21:14] to join the Ubuntu forums you will need to lookm at the Forums code of conduit [21:14] http://ubuntuforums.org/index.php?page=policy [21:14] it is basically the same ideas as the Ubuntu Code of Conduct more designed for the forums [21:15] same idea applies, "Don't be a jerk" [21:15] after joining you may question question in the forums [21:16] but before to do, I suggest searching the forum [21:16] The forums has a fantastic search engine, which is very quick [21:17] you can do a basic search by entering a term in the search box in the uppoer right hand corner of the forums [21:18] if you click on the serach link on commond ba near the top of the forum main page who can do an adavanced search [21:18] you can search by title only, or limit it to a single forum [21:19] also, the forum has been spidered by Google, so search via Google work well also [21:19] in Google just type [21:19] serach term site:www.ubuntuforums.org [21:20] to search forum via Google [21:20] any questions so far (search or otherwise) [21:20] ok Let talk about posting [21:21] Posting is very easy in the forums, even to a new user. [21:21] The post editor is a nice WYSIWYG interface [21:22] When you post first think about where to post [21:22] The [21:23] Absolute Beginner Talk forum is a great place to post with any such of beginner question [21:23] or you can get more specific in our other forum areas [21:24] There are areas for specfic hardware (amd64, Apple, Dell) and areas (wireless, multimedia,etc..) [21:25] When you are asking a questions on the forum, provide as much detail as possible [21:26] Include machine spec, OS version, 32 or 64 bit [21:27] Also tell people if you have any third party software installed ( Ubuntu/debain packages, not provided in the Ubuntu repos) [21:28] We also provide prefixes to post titles (such as Ubuntu, Kubuntu, Xubuntu, etc...) to help people uderstand what your problem maybe [21:29] Always follow the Code of Conduct when posting, ask for help, don't demand help [21:29] Remember people are volenteering to help you so be nice:) [21:30] and be patient, sometimes you may have to wait for someone to respond to your question [21:31] BE sure to post question to the whole forums and don't ask for help via a private message. That way more people see you problem [21:31] any questions? [21:32] ok, lets move on answering questions [21:32] < Heartsbane> QUESTION: Would you care to comment on the nature of distruptive commentary in the forums (i.e. people who post the solution being 'rm -rf /') has it subsided? are there any plans in the future to curb those types of comments? [21:32] Heartsbane: I was just getting to that [21:33] last year we had some forum member who thought it would be funny yo trick users into typing sudo rm -rf [21:34] sudo rm -rf /, which would delete the root directory of your Ubuntu install. major bad juju. [21:35] The forum staff willing ban any person to post malcious commands on the forums, trying to trick people in to using them. [21:36] There was a grassroot movement, and you will notice many older forum users have in there signature " Don't sudo rm -rf /" [21:36] when answeing post would you ask that people do a few things [21:37] Be understanding: The person asking the question maybe new to Ubuntu/Linux and maybe afraid of the command line and other linux terminology [21:38] Explain the solution in detail, once again, the person asking the question may need extra details in a answer [21:39] i.e., they may not know where the fstab file is located [21:40] Also, someone if the answer requires sudo, realize that you are asking someone to give you root permission to there box [21:40] be a good human being and respect that trust [21:41] to find new questions that have been posted to the forums goto [21:41] http://ubuntuforums.org/search.php?do=getnew [21:41] any questions? [21:41] < wolfger> QUESTION: why encourage use of the forums instead of use of "Answers" in Launchpad? [21:42] wolfger: both great ways to receive help [21:42] it is all about personal choice, which way works best for you to receieve help [21:43] maybe in the fture the two could be linked some how. [21:43] the last thing I would like to talk about is forum governace [21:45] The forum staff or moderators are all volenteers that help answer questions, remove spam (we get a lot), and keep the Ubuntu Forums a civil environment. [21:47] the Forums Council is a Ubuntu team that decides on direction of the forums, and acts to help mediator for users problem. [21:48] The Forum Council has monthly meetings, and you have view the meeting agenda here [21:48] https://wiki.ubuntu.com/ForumCouncilAgenda [21:48] I got about 10 more minutes, so more questions [21:49] < Az32> QUESTIONS: Is the Ubuntu Forums team developing some additional features for the forum, or you just use the bundled or external ones? [21:49] Az32: there are some social networking feature with the new version of vBB that we are looking into. [21:50] In the last year we added the Thanks feature, which has been a big hit [21:50] You can Thank someone for providing good help. Sort of like karma in launchpad [21:51] < Yaro> QUESTION: You commented on being helpful members on the forum. I like that. Perhaps you could eleborate on many ways one can be helpful on the forums? [21:52] Yaro: Answering questions, being polite to other forum users, etc... [21:53] Just being a good comunity member :) [21:53] Yaro: right now, with Ubuntu 8.04 out, we have a ton of new Ubuntu users needing help [21:55] < toobuntu> QUESTION: Is there a wiki page or something that has stock answers for pasting into the forums for things like opening a terminal, editing text files, and similar? [21:55] last question [21:56] toobuntu: not that I can think of, but the wiki is a fantastic place for getting help with Ubuntu. [21:56] I refer people to it all the time when answering forum questions [21:57] also, if you write a guide or how-to on the Ubuntu Forums, be sure to add it to wiki,ubuntu.com [21:57] it rocks [21:57] wiki.ubuntu.com :) [21:57] ok, that about does it for time [21:57] thanks Mike! [21:58] thanks all for coming === jcastro changed the topic of #ubuntu-classroom to: Ubuntu Open Week | Information and Logs: https://wiki.ubuntu.com/UbuntuOpenWeek | How to ask questions: https://wiki.ubuntu.com/UbuntuOpenWeek/Rules | Ask questions in #ubuntu-classroom-chat, prefaced with "QUESTION:" |See https://wiki.ubuntu.com/UbuntuOpenWeek/JoiningIn to filter out channel noise | "Python Packaging" - Emilio Pozuelo Monfort [21:58] next up is Python packaging [21:59] thanks jcastro [21:59] This is a new subject this time around, due to popular demand, so keep the ideas coming! [21:59] So who is here for the session? :) [21:59] jcastro: could you please -m the channel? [22:00] pochu: me :) === asac_ is now known as asac [22:00] +1 [22:00] _o/ [22:00] * wolfger is here to learn python packaging [22:00] +1 [22:00] +1 [22:00] +1 [22:00] This is my first session, so please forgive me if I go too fast or whatever, and just tell me [22:00] _o/ [22:00] In this session we will learn how to package a Python application [22:00] so let's get started [22:00] +1 [22:01] first let's do the preliminary work [22:01] we will need some packages, let's install it [22:01] $ sudo aptitude install wget dh-make devscripts [22:01] +1 [22:01] +1 [22:01] and let's prepare our working area [22:01] $ mkdir ~/python-packaging && cd ~/python-packaging [22:01] +1 [22:01] please tell me once you are done [22:02] n [22:02] (oops) [22:02] hmm, there seems to be some people joining, let's wait a couple of minutes [22:02] done [22:02] done [22:02] done :) [22:02] done [22:03] ok [22:03] nice :) [22:03] we're going to package terminator, which is a terminal emulator that can split several terminals in the same window [22:03] https://launchpad.net/terminator [22:03] done [22:03] first, let's download the source package [22:03] $ wget http://launchpad.net/terminator/trunk/0.8.1/+download/terminator_0.8.1.tar.gz [22:03] and let's rename it properly [22:03] $ mv terminator_0.8.1.tar.gz terminator_0.8.1.orig.tar.gz [22:03] this is because Debian packages have this structure _.orig.tar.gz [22:04] ls [22:04] let's unpack the tarball to work on it [22:04] $ tar -xf terminator_0.8.1.orig.tar.gz [22:04] ok [22:04] would it be possible to wait for a second. My install is still ongoing ... damn slow internet [22:04] ;) [22:04] pochu: just for you private use. check out "unp" ;) [22:04] grantgm: sure :) [22:05] while we wait, if you have any questions, feel free to ask at any moment [22:05] and feel free to interrupt if you get stuck [22:05] Are there any plans on bundling a full-featured "visual" Python IDE with Ubuntu? [22:06] can this channel be modded +m and have people chat in -chat? [22:06] pochu, will what we learn here be generally aplicable to all python packages distributed using distutils? [22:06] nosrednaekim: I prefer this way [22:06] it makes it nicer for the logs... [22:06] Az32: there's spe in the repositories, try it ;) [22:06] Thanks! [22:06] What about bundling by default? [22:06] Az32: and if you have questions, stani is the upstream author [22:07] sorry to keep everyone waiting. Its done now. [22:07] I don't think it makes sense to put an IDE in the default installation... very few people will use it [22:07] I second nosrednaekim... please moderate this channel [22:07] avu: I hope so. we will only package one, but I will try to explain everything so you can package anything later [22:07] pochu, ok, thanks [22:08] pochu: At least I would use it... ;) [22:08] jcastro: hmm, could you moderate this again, so that it's easier to make logs later? [22:09] ok, please ask your questions in #ubuntu-classroom-chat [22:09] so let's continue [22:09] $ cd terminator-0.8.1/ [22:09] OK, we have the upstream sources ready, let's start the packaging :) [22:10] we are going to use dh_make, which will create a template from which we will work on [22:10] $ dh_make -c gpl -s -b [22:10] which means: '-c gpl' tells it that the package is licensed under the GPL license [22:10] -s tells it that we just want one binary package. one source package can be splitted in many binary package. This is very useful when a package has huge documentation (so you build a -doc package too), or many images (so you build a -common package), or it's a library. We just need one, which will contain everything [22:10] -b tells it to use CDBS, 'Common Debian Build System', which will make our packaging simple, so we can concentrate on the Python specific things. [22:12] let's get into the packaging folder, debian/ [22:12] $ cd debian/ [22:12] $ ls [22:12] Ok, there we have the template files. There's many files we don't need (all the *.ex ones), so let's get rid of them [22:12] $ rm *ex *EX [22:12] everyone keeping up? +1 in #ubuntu-classroom-chat please :) [22:14] alright, let's continue [22:14] Ok, now let's start filling the template. first we will edit the control file, so open it in your preferred editor. [22:16] oly> QUESTION: what are .ex files i have not come accoss them before are they todo with packaging ? [22:16] oly: dh_make will put example files for a lot of things. we may need some. They are for init scripts, manpages, Debian menu files, watch files... [22:17] we could have used menu.ex and watch.ex, but we want to keep this as Python-oriented as possible [22:17] Ok, now let's start filling the template. first we will edit the control file, so open it in your preferred editor. [22:18] As you can see, the file is separated in two sections. The first one is always the source package info, and the others (in our case it's just one) are the binary packages. [22:18] For the source package, we need to change the Section (misc is a good one, the list is here: http://www.debian.org/doc/debian-policy/ch-archive.html#s-subsections), the Priority should be optional (it's extra for -dev, -dbg packages, and maybe a few others), and the Standards-Version should be 3.7.3 which is the newest one. [22:19] If you package a Python module, section should be 'python', but not for applications [22:20] Regarding the binary package, we need to put a short and long description [22:20] Use your imagination to make one ;) [22:21] And this is important for Python packages, the Architecture should be 'all', as our code is architecture-independent, so we don't need to compile it for every architecture (i386, amd64, powerpc, sparc...). This is true always for Python code, but it's different when we are packaging Python bindings written in C for example. [22:21] FrankNiessink: QUESTION: so the first line is the short description and the next lines are the long description? [22:21] FrankNiessink: right [22:21] progfou: QUESTION: shouldn't we keep the Standards-Version: field as generated by dh_make since everything else also generated is compliant to this specific version? [22:22] FrankNiessink: for the long description, every line should start with an space, and if you want to put a white line, it should be an space and a dot, " ." [22:23] progfou: that's a good point. However, there weren't big changes between 3.7.2 and 3.7.3 that would make our package incompatibly with 3.7.3, so we are fine [22:23] nosrednaekim: QUESTIONare the right and left arrows kept when writing the description? [22:23] but when we update Standards-Version, we should *always* check that we are still compliant to the new Policy [22:24] nosrednaekim: no, they aren't [22:24] for example, it could look like this: [22:24] Description: Simple and nice music player for the GNOME desktop Decibel is a GTK+ audio player which aims at being very straightforward to use by mean of a very clean and user friendly interface. [22:25] * stani wants pochu to say next when ready answering [22:25] stani: next :) [22:25] rulus: QUESTION: When creating multiple binary packages from one source package (f. ex. Python module + GTK program), what should the section of the source package be? [22:26] rulus: hmm, I'm not really sure TBH, but I think it should be that one which describes better the package [22:27] e.g. if it's mostly an application which ships a public module which *might* be useful to someone else, then put one for the application [22:27] but if it's a module with a small interface or something, put 'python' [22:27] probably looking at the upstream package name will help you [22:28] next [22:28] sebner: QUESTION: Python applications are all but if we package something made with C or something similar the only way is to *test*? how? [22:28] sebner: do you mean e.g. Python bindings written in C? [22:29] pochu: yes, or just a pure C programm. Sry if it doesn't fit here correctly [22:30] ok, you should test it as you would test any other package [22:30] if it's a module or bindings, use them (probably the upstream source contains some examples or regression tests) [22:30] if it's an application, use it as much as possible [22:31] don't simply open it and look at the menus, but look at the functionality and check that everything works as expected [22:31] next [22:31] morten: QUESTION: What are the primary differences between pycentral and python-support, and which should we use, and when? [22:32] morten: both do the same work, and both work very similar as far as I know, so use the one you like more ;) [22:32] or try both and pick up one [22:32] I'll let you guys decide which one to use in the future, although for this session we will use python-central [22:32] next [22:32] QUESTION: sometimes, does it worth to build native (arch dependant) python program ? can you tell when and why ? [22:33] native packages should be only used when the package is specific to Debian/Ubuntu [22:33] e.g. lintian is a good example, or dpkg [22:34] so unless you are developing (and not packaging in this case) something like that, use a non-native package [22:34] next [22:34] Kopfgeldjaeger: QUESTION: Do you need a setup.py file to package python applications or is there another way to do it? [22:34] Kopfgeldjaeger: not really, although it's easy to create a setup.py script, but you could package Python applications which use autotools or anything else [22:35] I maintain a couple of Python apps which use autotools one and just a Makefile the other [22:35] next [22:35] I propose to continue with your tutorial [22:35] alright :) [22:36] Now let's get to the Python specific bits. This is the important part of this session, as everything covered before would be general packaging tasks. [22:36] So let's look at the Build-Depends and Depends. The Build-Depends are the packages we need to build our package, and the Depends are the packages we need to run it. [22:36] Let's look first at the Depends. We need to add here all the modules our application uses: [22:37] $ grep -B 3 -A 3 import ../terminator [22:37] There we have our list of imported modules. Some of them are standard to Python, but others aren't. let's list them all. Also, most of them are required, but if you look at 'import gnome', if it's not installed it will just import the webbrowser module, which is a standard Python module. [22:38] with 'standard to Python' I mean that they are shipped with Python itself, so depending on Python will be enough to cover them [22:38] but there are modules which aren't shipped with Python, so we will need to depend on them so that our package can actually work [22:39] So we have gobject, which is in python-gobject, gtk and pango which are in python-gtk2, gconf which is in python-gconf, and vte which is in python-vte. [22:40] to find this you may need to do some research, although 'apt-cache search module | grep python' may help [22:40] For gnome, we will just recommend it (and not depend on it) as it's not required, because we have webbrowser as fallback. [22:40] So let's add 'python-gobject, python-gtk2, python-gconf, python-vte' to Depends. We can also remove ${shlibs:Depends} which isn't useful for pure Python packages. [22:41] And let's add 'Suggests: python-gnome2' after the Depends field. I've chosen Suggests and not Recommends because now Recommends are installed by default, and GNOME users probably have it installed, and we don't want to install it for KDE users (the less, the better for them). But feel free to disagree and use Recommends if you want :) [22:42] please, tell me if you get stuck [22:42] Now we will have a look at the Python Policy, http://wiki.debian.org/DebianPython/NewPolicy, which explains what we need to do to our package. [22:42] please everybody open that link :) [22:44] pochu: eveybody is following ;-) [22:44] cool [22:45] If you look at the 'Updating your packages' section, you will see it explains what to do for each case (python application, python module, package with extensions...). We are packaging an application, so let's look what it says about them. [22:45] The first point, although it doesn't say it, is for modules. Some time ago modules had a binary package for each Python version (e.g. python2.4-module, python2.3-module...). But fortunately that's not the case anymore, and we are packaging a Python application anyway so we don't care ;) [22:46] The second point affects us. We need to add a ${python:Depends} to the Depends field. That will add the correct Python dependency to our binary package. [22:47] The third point also affects us. We have to add a "XB-Python-Version: ${python:Versions}" field in the binary stanza (where Depends and Architecture are). It doesn't matter where you put it. You can put it between Architecture and Depends, for example. Note that if we had more than one binary package, we would need to put it for each one of them. [22:48] That's used by pycentral/pysupport, and it will get it's value from 'XS-Python-Version', but we will see that later [22:48] morten> QUESTION: Shouldnt we use debian/pyversions now? [22:49] morten: that's used by python-support only, not by python-central, and I think it's deprecated (or at least it's not required if you have XS-Python-Version) [22:49] morten: so we won't use it [22:49] pochu: stani seems to be away. you gave me voice. Should I copy the questions? [22:50] I am still here [22:50] ah ok ^^ [22:50] morten: in the wiki page, in the python-support section, it says: "Note: if there's no debian/pyversions, dh_pysupport will try to use the XS-Python-Version field to find out the list of supported versions. " [22:50] The fourth point doesn't affect us, as we don't ship extensions. [22:51] The fifth point wouldn't affect us if we weren't using CDBS, but as we want to make the packaging simple, we will need to pick either python-central or python-support (doesn't matter which one, both do their job). This is good because if we were shipping a module, either public or private (and many applications are modularized), we would need to use one of these tools, so now we will learn about them so when you need to use them you know how t [22:51] morten: /usr/share/cdbs/1/class/python-distutils.mk:60: WARNING: Use of XS-Python-Version and XB-Python-Version fields in debian/control is deprecated with pysupport method; use debian/pyversions if you need to specify specific versions. [22:51] morten: (but you are using pycentral, so i guess its ok) [22:51] Let's go with python-central. If you want to use python-support in your package, you will find that the steps are almost the same, and they are well documented in the policy. [22:52] morten: that's just a warning, not an error ;) but anyway we are going to use pycentral in this session, and not pysupport [22:53] progfou: QUESTION: is there specific things to do for a package only working with one version of Python (eg 2.5 and not before)? and the same question for a package not Python version dependant [22:54] BTW, the session will last some more time, FYI [22:54] there's nothing scheduled after, so take all the time you need! [22:54] progfou: yes, we will see that when I explain how to use XS-Python-Version [22:54] jcastro: thanks :) [22:55] So from steps first and second from 'Using python-central', let's Build-Depend on "debhelper (>= 5.0.38), python-central (>= 0.5.6)", substituting "debhelper (>= 5)" which is minor than 5.0.38. [22:55] The third step doesn't apply to us as we will use CDBS and that's for debhelper. But we will see that later. [22:56] As per the fourth step, we need to add XS-Python-Version to the source stanza of debian/control. The 'current' keyword is deprecated (the Policy needs to be updated), so we can use either 'all' if our package supports all Python versions, or >= X.Y. the README or ChangeLog doesn't mention which Python versions are supported, but it works at least with Python 2.4 so let's put >= 2.4 on it, "XS-Python-Version: >= 2.4". Remeber to put that in the [22:56] progfou: ^-- does that answer your question? [22:57] ebner: Question. Shouldn't we use pycentral >= 0.6 since there are fixes in it IIRC. (just did some bugfixing, transition).. [22:58] progfou: pochu: yes it answers my Python version question, but... I'll wait for next step to see... :) [22:59] sebner: yes, >= 0.6.0 makes more sense as it changed the path were the modules are stored. However our package doesn't ship any modules so we are safe [23:00] But if you package public modules, you should build depend on python-central >= 0.6.0 (the policy needs to be updated) [23:00] but it works at least with Python 2.4 so let's put >= 2.4 on it, "XS-Python-Version: >= 2.4". Remeber to put that in the source stanza, for example just below Standards-Version. [23:01] sorry, that got truncated [23:01] And the fifth point obviously doesn't affect us, as we don't ship any modules. [23:01] 9,10,11,12,13,14,15,16,17,18,19,20,22 < [23:01] simmons.freenode.net/#ubuntu-classro: [23:01] ooops [23:02] So we have this in place now, let's go to the CDBS magic, in the "CDBS + distutils" section. For this to work, our package needs to use distutils (the setup.py script usually). It's a system to distribute Python applications and modules: http://docs.python.org/dist/dist.html, but we don't need to worry about that now as our package uses it. [23:02] pochu: but it also doesn't matter if I *only* use >= 0.6 for all things? [23:02] no, that's fine. although that will make backporting harder [23:03] So the first point tells us to Build-Depend on cdbs (>= 0.4.49) in the control file, let's do it. [23:03] please, let me know if you are keeping up fine [23:05] pochu: please continue :) [23:05] The second point tells us to set the DEB_PYTHON_SYSTEM environment variable in the rules makefile. As we have chosen python-central, that would look like "DEB_PYTHON_SYSTEM=pycentral" before the includes. [23:06] And finally let's add the distutils magic, removing the makefile one (as the package doesn't have a Makefile but a distutils script). So let's change "include /usr/share/cdbs/1/class/makefile.mk" with "include /usr/share/cdbs/1/class/python-distutils.mk" [23:06] We should be done now :-) [23:07] Let's finish with the other files. As we don't need to document anything in README.Debian, let's remove it. We don't need the dirs file, as distutils and python-central will take care of making the necessary directories. [23:07] $ rm README.Debian dirs [23:07] And finally, let's edit debian/changelog. The 'dch' script from devscripts is really useful for this. [23:08] but for this we will need to have a couple of environment variables in place [23:08] so please, add something like this to your ~/.bashrc [23:08] export DEBFULLNAME="Emilio Pozuelo Monfort" [23:08] export DEBEMAIL="pochu@ubuntu.com" [23:08] rulus: QUESTION: What about "DEB_COMPRESS_EXCLUDE := .py" in the rules file? Is it (still) necessary? [23:10] rulus: I don't think it is, at least I don't use that for any of my packages. But I'm not really sure [23:10] rulus: for this package it isn't needed though, as there's no .py file [23:11] ok, now run this so that the terminal knows about those variables we have set [23:11] $ source ~/.bashrc [23:11] and let's edit the changelog: [23:11] $ dch -e [23:11] And let's put a bugnumber there in Closes (we don't have one, so let's invent it). [23:11] Also, we should change unstable with intrepid and '0.8.1-1' with '0.8.1-0ubuntu1' if we are packaging this for Ubuntu (although if you package it for Debian, then it will hit both Debian and Ubuntu, and probably other Debian derivatives) [23:12] also if it's for Ubuntu, the bug should be a Launchpad bug and we should change 'Closes' with 'LP' [23:12] We have almost finished this! We only need to put the right info in debian/copyright, but let's skip that here. That's very important though. Without a perfect debian/copyright file listing all the necessary info, the package will be rejected. [23:14] to write good debian/copyright files, this two mails from a couple of Debian Developers are very useful: http://lists.debian.org/debian-devel-announce/2003/12/msg00007.html, http://lists.debian.org/debian-devel-announce/2006/03/msg00023.html [23:14] So let's now build the package and install it! [23:14] please, make sure you have the dpkg-dev package installed, I forgot about it in the beginning [23:14] $ sudo aptitude install dpkg-dev [23:15] $ cd .. [23:16] $ dpkg-buildpackage -us -uc [23:16] That will build the source and binary package. The '-us -uc' options are not to sign the source and changes files (we would need to sign them to upload them to Ubuntu/Debian/your PPA, but if you have rights to do that, you probably know how to sign them ;) [23:16] If you are interested in PPAs, there was a session about it this Tuesday, logs are available at https://wiki.ubuntu.com/MeetingLogs/openweekhardy/LaunchpadPPAs [23:17] If everything has gone fine, that command will end with this: [23:17] dpkg-genchanges: including full source code in upload [23:17] dpkg-buildpackage: full upload (original source is included) [23:17] Is that the case? :) [23:17] FrankNiessink: QUESTION: I get: dpkg-checkbuilddeps: Unmet build dependencies: cdbs (>= 0.4.49) [23:18] * stani wants again a next blessing after an answer ;-) [23:18] FrankNiessink: ah, right [23:19] we need to install the Build-Dependencies! so let's run: [23:19] $ sudo aptitude install cdbs debhelper python-central [23:20] after that, build the package again: [23:20] $ dpkg-buildpackage -us -uc [23:20] (if it failed in the first run) [23:21] sebner: python 2.4 not found [23:22] gruber: QUESTION: I get: /bin/sh: python2.4: not found [23:24] pochu: ping [23:24] sebner, gruber: ok, please change XS-Python-Version, '>= 2.4' to 'all' [23:24] and try again [23:25] pochu: invalid setting [23:25] hhlp: QUESTIONS : can you put in pastebin an example of controlfile of this example to compare with mine (in the meantime) [23:25] pochu: ah no. same error [23:25] hhlp: sure, http://pastebin.com/f60fedee1 [23:26] (progfou: yep, to make it easier I've changed XS-Python-Version to 2.5 and it works 'til the end) [23:26] gruber, sebner: ok, so then we need to use 'current' [23:26] pochu: yeah. I saw your posted control file ^^ [23:26] That was my fault, I thought 'current' was already deprecated in python-central but it's not yet... [23:27] So to everybody: if we use '>= X.Y', or 'all' it will build the package for every python version. However, this is an application and that's not useful (that good for modules though) so for an application, 'current' will build it only with the default python version [23:28] I guess 'current' will be deprecated in the future, but it's not right now so we can use it [23:28] ok, continue [23:28] lup> could you paste the rules file too? [23:28] lup: http://pastebin.com/f4e6dc243 [23:29] ok [23:29] Ok, let's install the package now. the 'debi' command (from the devscripts package too) will install it if we have just built it from the package directory, so we are fine: [23:29] $ sudo debi [23:29] The last line should be this one: [23:29] Setting up terminator (0.8.1-1) ... [23:30] So we got it!! [23:30] Now you can start terminator with the following command: [23:30] $ terminator [23:30] And there is a terminator menu entry in the Accesories menu (that's in GNOME, I don't know where will it show in other desktops). [23:30] So we are done :-) [23:30] Questions? [23:30] lup: Got a different error: /usr/share/cdbs/1/class/python-distutils.mk:69: *** unsupported Python system: pycentral (select either pysupport or pycentral). Stop. [23:32] lup: what Ubuntu version are you running? (this could be due to old packages, I didn't tested the session in Dapper) [23:32] lup> It was white space after the pycentral, the quotes didn't help [23:32] alright :) [23:33] sebner: Question. Is it in generel better to use cdbs for python things? It may contain black magic [23:34] * stani starts with other questions now [23:34] sebner: not really. Any system is fine. I chose CDBS as that way it would be easier for starters to follow the session and concentrate on the Python bits, but using debhelper is equally fine [23:34] next [23:35] rulus: QUESTION: How do I combine 'multiple binary' with 'cdbs' in dh_make? It seems the one or the other. [23:36] rulus: I have to admit I have no idea, I have never used that, and indeed it doesn't seem to work [23:37] sebner: Question: Why have we a menu entry? We didn't created a menu file or is this black magic by cdbs? [23:37] rulus: file a bug report about it ;) [23:38] sebner: good question. Indeed we removed the menu file, but the upstream package contains a .desktop file, and the setup.py script installs it properly in /usr/share/applications/ [23:38] that's nice, isn't it? [23:38] next [23:38] progfou: QUESTION: how should I deal with real Python version specific files like new encodings having to be put in /usr/lib/python2.5/encodings/ ? should I put them at the right place by hand (using "install -c ...") or is there some more automatic way not requiring to know the exact location? [23:39] progfou: pycentral/pysupport will take care of that [23:40] progfou: e.g. with pycentral it will install the files into /usr/share/pyshared/, and it will byte-compile them into every python version [23:40] progfou: so you don't need to do that by hand, luckily [23:40] next [23:40] oly: QUESTION: how does packaging from scratch differ, i mean if i have no tar.gz file do i just tar up my existing project and work from there does the layout of the source files matter in anyway ?? [23:41] oly: ok, we assume here you are upstream for that project. So yes, you need to make a tarball (a public release for example, mypackage-0.1), so you can then package it [23:42] the layout matters a bit, so if you will use distutils, the distutils documentation explains where to put everything [23:42] next [23:43] lup: QUESTION: what do I need to remove to rebuild the package again? [23:44] lup: good question. We need to remove build/, so you can put that in debian/rules in the clean target [23:44] e.g. [23:44] terminator/clean:: [23:44] rm -rf build/ [23:44] that would fix it [23:45] it's important that a package can build twice in a row now, but i didn't want to enter into that so we could concentrate in the python specifics [23:45] next [23:45] grantgm: QUESTION: why does 'dpkg-buildpackage -us -uc' need to run as root? [23:45] grantgm: so I still haven't been able to build successfully: make: *** [testroot] Error 1 [23:46] grantgm: ah, I guess you aren't running Hardy ;-) [23:46] grantgm: or you may be, but you don't have installed fakeroot [23:46] grantgm: you don't need to, you can use fakeroot to do it [23:47] grantgm: in fact, you should never build a package as root in your production system [23:47] grantgm: so install the fakeroot package, and build it again (use dpkg-buildpackage -us -uc -rfakeroot if you don't have Hardy or later) [23:47] grantgm: 7.10 on this machine [23:47] I see, that should work for you then [23:47] next [23:48] tzeentch__: QUESTION: How does using different build tool change things, e.g. if a package used scons, setuptools or whatever, would cdbs and all other magic still work? [23:48] grantgm: stani: same problem (with fakeroot on 7.10) [23:48] grantgm: did you try '-rfakeroot' ? [23:49] tzeentch__: not exactly. The distutils magic won't work anymore [23:49] tzeentch__: but the pycentral/pysupport magic should work as long as the build system does the Right Thing [23:50] I have never packaged anything using scons or setuptools, but I maintain a Python package which uses autotools. I use debhelper for it though, but I guess I could switch it to CDBS without problems [23:50] next [23:51] I guess no more questions [23:51] but maybe you can give a short word about PAPT [23:52] yes [23:53] so PAPT, or Python Applications Packaging Team, is a Debian team which has many Python applications. There's also many Ubuntu developers and contributors packaging their applications there, as then they will hit both Debian and Ubuntu [23:53] I maintain all my Python applications there, and I have never had any problem in getting them into Ubuntu [23:53] pochu: because they get autosynced ^^ [23:54] yes :) [23:55] http://wiki.debian.org/Teams/PythonAppsPackagingTeam and https://wiki.ubuntu.com/ContributingToDebian/PythonModulesTeam have more info [23:56] Alright, thank you all for coming to the session! I hope you have learnt something with it, and I'll be more than happy to help you if you get in troubles with Python packaging or need sponsorship :-) [23:56] Thanks stani for pasting the questions too [23:57] pochu: yeah. 2 hours ^^ [23:58] also, there's #ubuntu-motu here on Freenode and #debian-python on OFTC where you can ask for help with packaging (the latter only for Python packaging obviously :) [23:58] nealmcb: QUESTION: What is a really good example of a simple, beautifully packaged and written and documented python package?