[07:15] how to edit ubuntu code? [07:15] pls help me [07:15] iwant to have a special ubuntu [07:16] :) [07:26] offerpook, if you just want to change your look, ask in #ubuntu how to do that (appearance is something that will help you, check it out before you ask there ;) ). if you want to change the code of ubuntu itself (programming), try learning some programming languages, grab the code and change it how you like it. thats difficult though === fucrum is now known as fulcrum [11:55] hi [12:49] lo [12:59] woo! welcome everyone [12:59] Bug Guru session in 1 hour! === ChanServ changed the topic of #ubuntu-classroom to: Welcome to the Ubuntu Classroom - https://wiki.ubuntu.com/Classroom || Support in #ubuntu || Upcoming Schedule: http://is.gd/8rtIi || Questions in #ubuntu-classroom-chat || Event: Ubuntu Open Week - Current Session: How to Become a Bug Guru! - Instructors: hggdh [14:00] Woo, we'll get started in a minute [14:00] hggdh: about ready? [14:00] Logs for this session will be available at http://irclogs.ubuntu.com/2011/10/20/%23ubuntu-classroom.html following the conclusion of the session. [14:00] jcastro: as reaqdy as I will ever be [14:01] hello. I am one of the Ubuntu Bug Squad/Bug Control admins [14:02] I will be a bit slower today, my notes for this class perished in a backup accident [14:02] (not a real bug, though, just an user error) === Wilczek_ is now known as Wilczek [14:03] if you have any questions, please post them on the #ubuntu-classroom-chat prefixed by QUESTION: <- note it is upper cased [14:03] the question itself, of course, can be mixed case [14:03] OK === zyga is now known as zyga-afk [14:05] We at Ubuntu receive an amazing amount of bug reports; Launchpad (the program we use for -- among others -- bug reporting) is logging, right now, more than 850,000 bugs [14:05] those are *all* bugs on LP, Ubuntu or not. Ubuntu-specific bugs are less, I would say around 700,000 [14:06] of which ~70,000 are still open... [14:06] (for the record, we do not have 70,000 people analysing these bug reports) [14:07] so. Much more bugs than bug triagers. What to do? [14:07] first off: we can try to have bugs (more) correctly opened [14:08] this is important -- we spend a lot of time trying to figure out which version of Ubuntu is the reporter running, what package/program is this all about, etc [14:09] no counting the description of the problem -- "I did not work as I asked" [14:10] this causes a bit of triage work -- first making sense of the sentence above (not that difficult, but already predisposes the triager against), then having to ask the reporter "what did not work as you expected? What did you expect? What did happen?", etc, etc [14:12] given things like the above, and the sheer amount of new bugs every day (monitor the channel #ubuntu-bugs-announce to have an almost real-time view of bugs as they arrive), we now prefer that bugs are opened from the reporter's system [14:13] via 'ubuntu-bug' -- you run 'ubuntu-bug ', replacing '' by the correct package (see below) or -- if you do not know the package, 'ubuntu' [14:14] this will collect a minimal set of data -- your Ubuntu version, the package version, and -- depending on the package -- a series of data relative to the package [14:15] and -- as a result -- we will not need to waste time asking VERY basic questions, like "what is your Ubuntu version?", etc [14:16] YOu can still directly open a bug on Launchpad; nevertheless, if you are not an expert on the issue, chances are you will forget to add some crucial tiny bit of data, and will make the bug work a bit more delayed [14:18] for finding the package: you can search by the program's name, using either apt-file, or apt-cache, or dpkg: [14:19] apt-file: must be manually installed (not a default install option); it keeps track of *all* packages, be them locally installed or not [14:20] apt-cache: (IDK if it is an auto-install) -- keeps track of all *installed* package contents [14:20] dpkg: can search the locally-installed package contents [14:21] of all the above I usually prefer apt-file (for searching), since it is not necessary that a bug will be about a package I have installed [14:23] for the record, you use 'apt-file search blahblah', or 'apt-cache search blahblah', or 'dpkg -S blahblah'; these programmes have many more options, and most of them are interesting for someone working on bugs [14:23] so, please, *always* Read the Man Page. [14:24] is there any difference between installed, and local installed packages i might have missed? thought it was the same, but itb [14:25] on what I am talking about, there is no difference -- installed == locally installed == (locally) installed [14:26] but there are a few thousands packages in the archives, and an Ubuntu install is usually between 700-1300 packages [14:26] or something like it [14:27] If you cannot find the package name for running ubuntu-bug, then open the bug against Ubuntu. This is not the preferred way -- we would rather have a correct package in the bug -- but it is better than directly using Launchpad [14:28] ubuntu-bug will collect some data, and THEN open a browser on launchpad, and ask you to enter details of the bug [14:30] please -- PLEASE -- do enter details on what is going on (or not going on, as the case may be). One common mistake we do is think that everybody else knows what *we* know. Nothing is farther from truth... [14:32] please explain your issue. Use your own words, do not try to make it more "technically correct" -- making it technically correct is *OUR* job. But do explain WHAT, WHY, HOW [14:32] WHAT happened, WHY it was unexpected, HOW you tried it [14:34] I understand that all you know is 'it did not work as I wanted/needed/expected'. But, at this point in time you know much more that we. If *we* do not understand the what, why, and how, there is little chance you will get a resolution [14:36] By the way, another thing we are very much insistent on: please be nice or, at least, neutral. We have a Code of Conduct, and we *do* abide by it. [14:37] Of course, most of the folks working on Ubuntu, be it bugs, or coding, or packaging, are _volunteers_. Antagonising a volunteer is usually not a good way to have *my* problem resolved... [14:38] aggressive behaviour/moods are very frown upon [14:39] another point of contention is the Meaning of Status and Importance [14:39] (which are much more discussed than the meaning of life, the universe, and everything) [14:40] Xubuntu class? [14:40] for the real gory details, please look at https://wiki.ubuntu.com/BugSquad/KnowledgeBase [14:41] and specifically https://wiki.ubuntu.com/Bugs/Importance and https://wiki.ubuntu.com/Bugs/Status [14:41] but... INVALID means 'this is not a bug' [14:42] but it is rather generic, and quite strong [14:43] we close (a bug) invalid when the reporter recants ('very sorry, but I found that I was doing blahbleh instead of the correct yaddayadda'), or is asking for *support*, or is about a non-ubuntu package, or... [14:44] support is a very special case, and most prone to get end-users confused [14:45] a BUG is a *problem* in a package -- program crashing, does not start, does not accept (correct) input, fails to uninstall, etc [14:46] a SUPPORT request is a *question* about how to use a package/program/whatever [14:46] launchpad deals with BUGs, not support [14:46] for support... a marvelous site is http://askubuntu.com/ [14:47] for the record, I usually go there if I have a doubt on how to get something done [14:48] and I have started providing some answers of my own; after all, if I use it to resolve my doubts, it is only fair to also answer about what I know [14:50] There are 10 minutes remaining in the current session. [14:50] Finalising, we -- bug triagers and similar -- usually congregate at #ubuntu-bugs. If you have a question on *triaging* -- how-to, can it be set as triaged, etc --, this is the pace to go [14:50] and this is it. [14:50] Thank you very much, and you all have a nice day [14:52] ok, just about 8 minutes until the next class [14:52] so we'll take a break, smoke if you got em! [14:53] Will that be the Xubuntu class? [14:54] pasti: no, Ubuntu Orchestra [14:55] There are 5 minutes remaining in the current session. [14:58] pasti: xubuntu will be the one after === ChanServ changed the topic of #ubuntu-classroom to: Welcome to the Ubuntu Classroom - https://wiki.ubuntu.com/Classroom || Support in #ubuntu || Upcoming Schedule: http://is.gd/8rtIi || Questions in #ubuntu-classroom-chat || Event: Ubuntu Open Week - Current Session: Ubuntu Orchestra -- Servers, Live and in Concert! - Instructors: kirkland [15:00] Logs for this session will be available at http://irclogs.ubuntu.com/2011/10/20/%23ubuntu-classroom.html following the conclusion of the session. [15:00] ok kirkland, take it away! [15:00] howdy all! [15:00] this session is about Ubuntu Orchestra [15:00] which is (what I think to be) perhaps the most exciting feature in Ubuntu 11.10 on the Server side [15:01] you can think of Orchestra as *the* way of installing Ubuntu servers over the network [15:01] it's a collection of the best of open source packages [15:01] for provisioning, installing, logging, monitoring, and managing [15:01] physical systems installed with Ubuntu Server [15:02] in this session, i'll introduce you to the parts of Orchestra, and even demonstrate the bits I can [15:02] for demonstration, please ssh guest@ec2-107-22-49-46.compute-1.amazonaws.com [15:02] the password is 'guest' [15:02] * kirkland waits for anyone else to show up there :-) [15:03] aha, someone else logged on :-) [15:03] okay, good [15:03] so I'll start by installing Orchestra, with 'sudo apt-get install ubuntu-orchestra-server' [15:04] that's going to take a minute [15:04] in the mean time, let's look at the different parts of Orchestra [15:04] first, and foremost, there's a provisioning server [15:05] the provisioning server is largely Cobbler (which is an excellent project from the Fedora team) [15:05] but we've added a few bits to surround cobbler [15:05] that makes it even more useful in Ubuntu environments [15:06] we also use the distro-info command to determine the list of currently supported ubuntu releases [15:06] .oO (you can find more info on cobbler here: https://fedorahosted.org/cobbler/ ) [15:06] and every Sunday morning, Orchestra will go out and fetch the latest Ubuntu mini.iso's, and automatically import them into Cobbler [15:06] so that your Orchestra installation is always ready to install each of the Ubuntu releases [15:07] otherwise, you'd need to do this manually yourself, and frequently to keep up with changes [15:07] the provisioning server also contains a squid proxy [15:07] so that as network installations are performed, the vast majority of the binary data is cached locally [15:08] this ensures that subsequent installations happen very fast [15:08] note that you can also point Orchestra/Cobbler to your local mirror, if you have one [15:08] but even if you don't (mirrors take upwards of 500GB of disk, and lots of bandwidth to keep in sync) [15:08] you'll at least get the convenient benefit of squid cache [15:09] Orchestra also includes dnsmasq, which is configured at your option [15:09] Orchestra/Cobbler can use dnsmasq to manage your dhcp and dns, which allows for some pretty cool features [15:09] but doesn't necessarily fit into everyone's network infrastructure [15:10] so we leave it as optional [15:10] however [15:10] if you want to use Juju with Orchestra (which, I'm not going to get into in this session), you MUST enable Orchestra to manage dhcp+dns with dnsmasq [15:10] okay, so that's the Provisioning Server [15:11] Orchestra also includes a Logging Server [15:11] this is really, really cool! [15:11] the Orchestra server is a rsyslog remote logging server for all of its installed clients [15:11] when performing remote, headless installations, you can watch the installation proceed by tailing the syslog on the Orchestra Logging server [15:12] as part of the installation, Orchestra exchanges SSL keys with the machine being installed, so that this logging happens over a secure SSL encrypted connection (after installation) [15:12] (note that the installation logs are logged in the clear, over UDP, which I think is a limitation of the debian-installation and busybox) [15:13] beyond the Logging Server, Orchestra also includes a Monitoring Server, which, for now, is just Nagios [15:13] the Nagios CGI web front end and aggregate server is installed on the Orchestra Server, and nagios clients are installed on each installed machine [15:14] at UDS in Orlando, we're going to revisit this, and consider some other monitor options in addition, like collectd, munin, and ganglia [15:14] now, the Orchestra-installed clients themselves [15:15] are a little more than your basic Ubuntu server [15:15] we've made a few opinionated decisions with Orchestra to really get these machines working well, locked in tight step, with one another [15:15] all of the machines are running NTP and have their time sync'd [15:15] which is *very* important for the syslogging [15:16] they're all running an SSH server, and they have the Orchestra server's keys installed on them [15:16] all of their /etc configuration directories are managed by etckeeper [15:16] which means that /etc is under bzr revision control, and gets committed any time dpkg or the administrator makes a configuration change [15:17] this *really* helps manage configurations over time (take a look at etckeeper, if you're not already using it!) [15:17] all of the clients run cloud-init (like Ubuntu systems in EC2) [15:17] even though these are physical machines, they can run cloud-init too [15:17] though we don't use metadata [15:17] Juju uses this to pass information to the machine via the preseed [15:18] QUESTION: Is orchestra a 11.10+ only feature, or is there a PPA for my older 10.04 servers? [15:18] jcastro: there is an orchestra ppa, ppa:orchestra/ppa [15:18] jcastro: where the packages are automatically built and uploaded as part of my release-build and release scripts in the bikeshed package [15:18] jcastro: but note that those PPA packages are completely untested [15:19] jcastro: we'll accept and fix bugs in those, as possible [15:19] jcastro: but those PPA packages are on a best-effort basis, no more [15:19] back to the client, it also installs the rsyslog daemon as well [15:19] okay, so that's the basics ... [15:19] our installation is done on our machine in the cloud [15:20] if you're late joining, feel free to join us with ssh guest@ec2-107-22-49-46.compute-1.amazonaws.com [15:20] password=guest [15:20] let's take a look at a couple of things [15:21] first, the configuration in /etc/orchestra [15:21] there's a configuration for the ISOs you want orchestra to automatically import [15:21] we use a decent set of defaults, i think [15:21] but you can override those here [15:21] for instance, if you just want to import Oneiric and Lucid 64-bit iso's, you can do that here [15:21] we also default the locale of each machine to en_US [15:22] for obvious reasons, you might like to change that :-)_ [15:22] we also have a set of settings in /usr/share/orchestra [15:22] these are not in /etc as we can't really support people changing them [15:23] part of what we're doing in Orchestra is getting things working well together [15:23] and that means we have to require a few configurations [15:23] next, let's look at the preseed files [15:23] which are the instructions for the machines that Orchestra installs [15:24] the goal here is to install physical machines [15:24] in a headless, remote manner [15:24] which means we must answer *all* installation questions ahead of time [15:24] and we do that with preseed files [15:24] let's look at /var/lib/orchestra/kickstarts/orchestra.preseed [15:25] these are the base templates for those installations [15:25] Cobbler provides an *awesome* mechanism for customizing these, actually [15:25] and we'll look at the cobbler interface here shortly [15:25] but this is our Orchestra "distro default" preseed [15:25] something that we've tested the heck out of, and *know* works well [15:25] toward the bottom of that file [15:26] you should see some $SNIPPET()'s [15:26] this is another cobbler feature we're leveraging here [15:26] these are basically #include's that bring in installation instructions from elsewhere [15:26] so cobbler ... [15:26] cobbler is *awesome* [15:27] when we first started working on the Orchestra project, we evaluated several different network installation mechanisms [15:27] such as cobbler, fai, uec-provisioning, etc. [15:27] we also considered writing our own [15:27] however, all of us that looked at Cobbler in depth were very impressed with the possibilities [15:27] most importantly, it had 3 things we wanted [15:28] 1) a command line interface (ie, the cobbler command -- see the manpage at http://manpg.es/cobbler) [15:28] 2) a web interface (we're going to look at that in a minute) [15:28] 3) MOST IMPORTANTLY, it has a python API [15:28] (3) was the real kicker for us! [15:28] using the Python API, we can now tie Cobbler (and ultimately Orchestra) to lots of other interesting projects [15:29] like Juju [15:29] and Landscape [15:29] sudo cobbler distro list [15:29] this shows the list of distros imported into cobbler [15:29] remember that "automatic" part? [15:29] here's how you can see that it happened! [15:30] anyway, we could spend hours playing with the cobbler command [15:30] check the manpage, and you'll see that you can list, add, edit, remove lots of configuration entries [15:30] at this point, though, let's take a look at the web interface! [15:31] so i'm going to give you all the URL and the cobbler user's password [15:31] and trust that you don't do destructive things to the database :-) [15:32] one second ... [15:33] i need to ec2-authorize this machine [15:35] hang on, sorry, juju issues here [15:38] okay, gonna need to come back to this one :-) [15:38] i'm starting a second orchestra instance in the background [15:38] turns out cobbler_web and ajaxterm conflict on the web interface :-) [15:38] too much going on with this system [15:38] i'll pause here for questions...before moving on [15:39] nealmcb: yeah I have a cobbled together box that does most of this [15:39] jcastro: right! [15:39] jcastro: that's kind of what we're trying to solve, at a distro level [15:39] I'm betting many, many Ubuntu admins out there have hacked together something like this [15:39] it's kinda fun to setup a pxe + tftp + mirror server, and do some network installs [15:39] maybe the FIRST time :-) [15:39] but once you have to do it more than once [15:40] or, maybe you have to do it for a HUGE network [15:40] that's what we're trying to help you with, in Orchestra [15:40] QUESTION: How can I add other distros to orchestra/cobbler? Like if I wanted to try a Fedora or SUSE [15:40] jcastro: excellent question [15:40] cobbler works with lots of distros [15:40] in fact, it was pretty broken with Debian/Ubuntu, until we started working with the upstream project [15:41] within Orchestra, we don't auto import other distros, but it's pretty easy to do [15:41] when i get the web interface up, i'll show you [15:41] note that we're not testing those other distros at all, right now [15:41] and we're not shipping preseeds/kickstarts for them either [15:41] so it might take a little work on your part to get them working [15:42] but we've also not done anything to specifically disable or break those [15:42] in fact, if you or anyone else wants to contribute kickstarts, etc. to orchestra that makes it work better with other distros, i don't see any reason why we wouldn't take them [15:43] okay, almost done installing Orchestra sever #2 :-) [15:43] QUESTION: How does Orchestra compare with (or work with) tools like Chef? [15:43] mmclark: good question [15:44] mmclark: in the early version of Orchestra (the 1.x series is unsupported, but still in Launchpad), we actually relied quite heavily on puppet [15:44] mmclark: but as we saw the Juju project mature, we eventually switched everything over to Juju [15:45] but, if you have preseeds that set up your puppet or chef environment how you like [15:45] it would be trivial to add those as snippets to your base preseeds in Orchestra [15:45] and Orchestra would basically get you to the point where your system is installed and rebooted on the bare metal [15:45] and if you're an expert Chef/Puppet shop, your Chef/Puppet conf management would take over at that point [15:46] it's actually a pretty seemless transition [15:46] okay! [15:46] http://ec2-184-73-142-29.compute-1.amazonaws.com/cobbler_web [15:46] user and password are both cobbler [15:47] QUESTION: You baked some configuration into /usr/share, including e.g. some apache stuff. Are there places where this makes it hard for folks to install or customize their own services and apps? [15:47] nealmcb: to an extent, yes [15:47] nealmcb: this was a difficult balance we had to walk [15:47] nealmcb: something things (like the squid proxy), we absolutely require to be setup in a particular way, or else the tools just don't work [15:48] nealmcb: apache and squid, etc., are fundamentally "embedded" dependencies of what we're doing with Orchestra [15:48] nealmcb: file bugs, though, as you run into problems [15:48] nealmcb: and we'll do our best to get them worked back into /etc, if they're real issues [15:48] nealmcb: we're already using dotdee to fix a few issues like this [15:48] okay, let's look at this web interface! [15:49] this is the old-school Cobbler web ui, with a slick Ubuntu color scheme and CSS skin [15:49] click down the operations at the left side [15:49] Distros -> [15:49] and you should see the list of distros that it auto-imported [15:49] click on one of those distros [15:49] and you'll see some configuration that we auto populated [15:50] click Profiles -> [15:50] There are 10 minutes remaining in the current session. [15:50] and again you'll see a bunch of auto generated profiles [15:50] (note that you can create your own here) [15:50] click a profile [15:50] and you'll see some more stuff you can edit [15:50] like the kernel parameters, for instance [15:50] click Systems -> [15:51] and you'll see this is empty here :-( [15:51] unfortunately, this is the MOST interesting of all the sections [15:51] as this is where the physical machines actually live [15:51] click Create New System and you should see what a machine looks like to cobbler [15:51] note that Daviey wrote an awesome utility called cobbler-enlist [15:51] which is now on the 11.10 Server ISO [15:52] which facilitates the autopopulation and injection into this row in the cobbler mysql datbase [15:52] this is where you would import the 1 or 10 or 10000 machines you want Orchestra to install [15:52] you can poke around here for a bit [15:52] i'll leave this instance running for a full hour [15:52] back to jcastro's question .... [15:53] jcastro: look at the Import DVD link on the left [15:53] jcastro: this is how you would use the web interface to import a new distro [15:53] note that *everything* you can do through the web ui you can *also* do from the command line and api [15:53] jcastro: so you'd first import the iso [15:54] jcastro: then you'd need to click Profile [15:54] jcastro: and create a Profile [15:54] jcastro: sorry, [15:54] jcastro: 1) import iso, 2) create distro, 3) create profile [15:54] jcastro: it's kind of a multi part, complicated process [15:54] jcastro: which is why we auto-import all of the ubuntu isos [15:55] There are 5 minutes remaining in the current session. [15:55] QUESTION: I was under the impression that Ubuntu Orchestra 2.x still uses puppet extensively, as per deps in ubuntu-orchestra-common... just, juju automates throwing that out to your cloud env [15:55] lunitik: Orchestra 2.x doesn't use puppet at all implicitly; though it's certainly possible for you to add a couple of puppet snippets and extend Orchestra to use puppet [15:55] lunitik: Juju was first written to talk to virtual machines and EC2/Cloud instances [15:56] lunitik: however, that team did some AWESOME work to get Juju talking to the Orchestra/Cobbler python API [15:56] lunitik: and now Juju can work against bare metal too [15:56] lunitik: so 'juju bootstrap' would ask Orchestra/Cobbler for a physical machine to bootstrap [15:56] lunitik: and 'juju deploy ...' asks Orchestra/Cobbler for a physical machine to deploy [15:57] lunitik: and in those cases, Orchestra triggers a new installation of a machine [15:57] lunitik: and ~5 minutes later, there's a brand new orchestra client that's waiting for instructions from Juju [15:57] ah, i forgot about powernap and powermanagement in Orchestra! [15:57] all Orchestra clients are running powernap in powersave mode by default (though you can tune that up or down) [15:58] and if you add powermanagement information into Cobbler (ie, WoL, impi, etc) [15:58] then Cobbler can automatically power on/off systems, as you (or Juju) asks for them to be deployed [15:58] QUESTION: Ok so one could just add enlisting in a kickstart rule or whatever and this would autopopulate? [15:58] jcastro: i don't understand the question [15:58] but I'm about out of time [15:59] I'll follow up later [15:59] feel free to talk to me or others in the #ubuntu-server channel [15:59] btw ... [15:59] roaksoax, spamaps, mathiaz, daviey, zul, smoser, and a bunch of others [15:59] did a lot of work on Orchestra [15:59] the whole Ubuntu-Server team kicks ass :-) [15:59] thanks all! === ChanServ changed the topic of #ubuntu-classroom to: Welcome to the Ubuntu Classroom - https://wiki.ubuntu.com/Classroom || Support in #ubuntu || Upcoming Schedule: http://is.gd/8rtIi || Questions in #ubuntu-classroom-chat || Event: Ubuntu Open Week - Current Session: Xubuntu: What is that? - Instructors: charlie-tca [16:00] Logs for this session will be available at http://irclogs.ubuntu.com/2011/10/20/%23ubuntu-classroom.html following the conclusion of the session. [16:01] I'm Charlie Kravetz, known as charlie-tca on irc and the mailing lists. I am Xubuntu Quality Assurance Lead and Xubuntu Project Lead. [16:01] Welcome to Open Week and Xubuntu [16:02] We are going to talk a bit about Xubuntu. We will take your questions in #ubuntu-classroom-chat, and [16:02] will have a few minutes at the end to answer questions you hold on to. Feel free to ask questions at any time, [16:02] but please start them with QUESTION: so they easy to spot. For example, [16:03] QUESTION: What is Xubuntu? [16:03] The answer would then be [16:03] Xubuntu is Ubuntu with the Xfce desktop instead of Gnome and Unity. Xfce emphasizes conservation of system resources, which makes Xubuntu an excellent choice for any system, new or old. [16:03] Xubuntu is pronounced "zoo-boon-too" in english. [16:04] As a ubuntu derivative, Xubuntu maintains the same high standards and quality that Ubuntu has. Xubuntu is built and maintained by volunteer contributors and developers combined. [16:05] We still have bugs to track down and resolve, and we do help with the bugs in Ubuntu also. Testing is a combined effort, since the releases are on the same schedule. [16:05] Xubuntu is an ideal candidate for older hardware or low-end machines, thin-client networks, or those who would like to get more performance out of their hardware. [16:05] We produce both 32-bit and 64-bit versions of Xubuntu. The images may be burned to CD-r or copied to a USB drive for installation. There is no DVD version of Xubuntu. [16:06] And, since Xubuntu is a derivative of Ubuntu, we use the same repositories, and many of the same applications. [16:06] Those of us on the team really believe the stated minimums should be real! If we can't run what the minimums say, we question it. [16:07] Granted, we don't expect perfect performance from it, but the system should be usable. [16:07] Using Xubuntu, we expect the applications provided to be able to run, one at a time with the minimum hardware. [16:07] As the hardware increases, we would expect the performance to increase. [16:08] New users are often surprised to find that Xubuntu includes a number of gnome applications. [16:08] These are included simply because if an application works well, and is considered lightweight, it fits. [16:08] Any application can be included, and it does not matter if it starts with gnome, xfce, k or anything other letters. [16:09] The target audience for Xubuntu is users who are interested in having a modestly light weight, slim, fast desktop experience. [16:09] Those users should be able to retain the usability and functionality that is required to provide an easy to use desktop environment. [16:09] Xubuntu does not specifically focus on new users or users migrating from Windows; alternative distributions such as Ubuntu may be more appropriate for first time Linux users [16:10] tomswartz07 asked: What influenced the choice to use XFCE over Opebbox or other lightweight DE's? [16:10] AXubuntu does not specifically focus on new users or users migrating from Windows; alternative distributions such as Ubuntu may be more appropriate for first time Linux users [16:10] An excellent question, tomswartz07 [16:11] Xubuntu choose Xfce in 2006 because it was ready. It allowed Xubuntu to put together a quality desktop, which would give Ubuntu the top three desktop environments of the time (Gnome, KDE, Xfce) [16:13] Xubuntu does not specifically focus on new users or users migrating from Windows; alternative distributions such as Ubuntu may be more appropriate for first time Linux users [16:13] (especially first time Linux users who may be particularly at risk of experiencing difficulties due to lack of general experience). [16:13] Xubuntu does not exclusively target users with low, modest, or high powered machines but instead targets the entire spectrum with a strong focus on enabling lower end machines. [16:13] oliverhr asked: Plans to migrate or support for Gtk3? [16:14] Thanks for asking, oliverhr [16:14] As Ubuntu has switched to GTK3 as much as possible, we also are integrating it. [16:15] However, Xfce does not plan to switch to Gtk3 for a while yet. [16:15] Current plans for Xfce 4.10 do not include migration to Gtk3 yet. They are instead looking at stability of Gtk2 for the next release. [16:17] Xubuntu will support Gtk3 applications in 11.10, and we will continue to use both Gtk2 and Gtk3. As a matter of fact, Greybird theme used by Xubuntu 11.10 is the only Xfce theme designed for both gtk2 and gtk3. [16:17] Xubuntu's extra responsiveness and speed, among other positive traits, can be appreciated by all users regardless of their hardware. === zyga-afk is now known as zyga [16:18] Are there other applications that could provide the same functionality? Most definitely. [16:18] We are using applications that are light in resources, and relatively easy to configure for most users. [16:19] You are welcome to use other applications if you desire. [16:19] As a matter of fact, we do routinely check our applictions as well as others to see if they still belong in Xubuntu. [16:19] A few of the more well known alternatives which could make the desktop a bit more lightweight are: [16:20] chromium - an open source browser - http://www.chromium.org/Home [16:20] claws-mail - a very nice mail client with many options - http://www.claws-mail.org/ [16:20] gpicviewer - A Simple and Fast Image Viewer for X - http://lxde.sourceforge.net/gpicview/ [16:21] geeqie - a lightweight Gtk+ based image viewer for Unix like operating systems. - http://geeqie.sourceforge.net/ [16:21] midori - a lightweight browser in development by Xfce. For more information see http://software.twotoasts.de/index.php?/pages/midori_summary.html [16:21] Pino - a simple and fast X11 client for Twitter and Identi.ca - http://pino-app.appspot.com/ [16:22] Any or all of these can be installed by the user. Please check the repositories before downloading or compiling applications. [16:22] Please note that the above applications are not presented as approved or recommended by myself or Xubuntu. There are given here as examples. [16:23] Xubuntu does not install these applications by default at this time. Why not? [16:23] Since we have limited developer resources available, we use applications maintained by Ubuntu that fit our needs. The application must also have a good user GUI, if possible. [16:23] The more complicated it is to configure the application for use, the less likely it will fit the requirement. [16:24] Some of the above are still in development, and are not yet released as a stable version. That, too, must be considered before including the application in a stable operating system. [16:24] Are there any other questions? [16:25] We will work very closely to the Ubuntu bugsquad when triaging, and follow the guidelines set in https://wiki.ubuntu.com/Bugs/HowToTriage. [16:26] A very important idea in triaging is that just because only one person had a failure, that does not make it invalid. [16:26] We work very close with upstream xfce, abiword and gnumeric bug teams to help resolve the bugs found. Once you report a bug, it goes through a process called bug triage. [16:26] Bug triage is simply the process of determining that there is a bug, how important is the bug to the overall scheme, and is there enough information for a developer to resolve the bug. [16:26] You will find that I often triage the Xfce, Abiword, and Gnumeric bugs for this reason. I am also the bugsquad contact for these bugs. [16:27] At this time, I would like to hit on some of the key points in Xubuntu 11.10, otherwise known as Oneiric Ocelot. [16:28] The contributors and developers of Xubuntu consider Xubuntu 11.10 to be one the best releases yet. Through their efforts, this distribution continues to improve steadily. [16:28] gThumb is now in the default Xubuntu 11.10 installation, to help users with image transformations and viewing. The Xubuntu developers wanted to insure an enjoyable, usable operating system. By including gThumb, along with Rhistretto, users have the ability to not only view all images and files, but also to perform minor editing of photos. [16:29] leafpad is the new default text editor(was mousepad), and now includes the ability to print. [16:29] For those users who prefer it, gedit also works quite nicely in Xubuntu 11.10. [16:29] pastebinit is now included in Xubuntu 11.10 intallations by default. If you need to use http://paste.ubuntu.com/, you can use pastebinit in terminals to paste directly without copying and pasting the data. [16:30] TonyP asked: How does the Xubuntu footprint compare with Ubuntu. I have an old EEEPC with a 4GB SSD and it is a struggle to update it due to insufficient space. Would Xubuntu be a better choice? [16:30] Thank you for asking, TonyP [16:31] Xubuntu uses about 30% fewer resources than Ubuntu. Where Ubuntu will require about 4 GB for the installation, using a Xubuntu alternate cd to install, it will use as little as 2GB. [16:32] We also can use less ram, but it will affect performance. [16:33] Of course, all numbers do depend on what you add to the basic installation,. [16:34] TonyP: I could suggest adding a USB pendrive to help with the space. Plugging in a 4 or 8 gb drive and using it for /home or even /tmp could alleviate that issue. [16:34] LightDM is the new application that manages logins in Xubuntu 11.10. [16:35] Onboard (the onscreen keyboard) is now included in the default Xubuntu menus, under Accessories. For those who require an onscreen keyboard, this will be much easier to access using only a mouse or touchpad. [16:35] Xubuntu will continue to use Synaptic Package Manager, yet provides the Ubuntu Software Center for those who prefer it. [16:35] Thunderbird is a lightweight mail/news/RSS client. It fits well with the fewer resources desired for Xubuntu, yet remains an easy to configure application for the new user. [16:36] The Ubuntu Font Family is available for your use and enjoyment. It is not, however, the default fonts in Xubuntu. [16:36] TonyP asked: Why the alternate CD? [16:37] TonyP: because of the GUI installer used by the desktop cd, it requires more memory and drive space to complete it' [16:37]