Re: Feature Request: Better partitioning wizard
On Wed, 2008-07-09 at 10:12 -0400, Mackenzie Morgan wrote: > The possibility for error with resizing partitions is why I think LVM > would be a good thing; however, from what I hear, LVM is pretty buggy > on Ubuntu. I strongly disagree on both counts -- LVM still relies on filesystem resizing features, which are still risky. Even with the so-called "reliable" ext3, I've lost several filesystems due to on-the-fly resizing, either because of the process being interrupted by a freeze / power outage or for other inexplicable reasons. I wouldn't consider LVM a good workaround for fixing partitioning mistakes later -- it only makes it a LITTLE less difficult to work around. This also requires a separate /boot partition, which could land you back at this same problem anyway. Also, I've not experienced any kind of bugginess with LVM. I use LVM on most of my systems. John -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking bug for ext4
Please stop filing nonsense bugs without first understanding the situation. ext4 will become the default filesystem once upstream recommends it for adoption (i.e. 2.6.28). GRUB still does not support reading ext4 so we will probably need a separate /boot on ext2/ext3, or wait for one of the SoC projects to magically finish. There is no need to clutter the bug tracker with this. On Tue, Nov 4, 2008 at 3:31 AM, shirish <[EMAIL PROTECTED]> wrote: > Hi all, > I made a tracking bug for having ext4 as the primary filesystem in Ubuntu. > > https://bugs.launchpad.net/ubuntu/+source/grub/+bug/293465 > > Perhaps the description and stuff can be overhauled. > As a simple generic desktop user having an online defragger and better > journaling, data integrity is what I'm looking at. > > From what it appears, it may take jaunty or jaunty+1 cycle for this to > become a reality. > My goal is essentially when I can use ext4 either from the Live or the > alternate CD. > Looking forward to people's comments on the same. > -- > Regards, > Shirish Agarwal > This email is licensed under > http://creativecommons.org/licenses/by-nc/3.0/ > http://flossexperiences.wordpress.com > 065C 6D79 A68C E7EA 52B3 8D70 950D 53FB 729A 8B17 > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: how to exclude --.gvfs from disturbing stuff
That error comes from lsof which cannot access ~/.gvfs because root != shirish. Pass -w into lsof to suppress that warning. On Mon, Nov 3, 2008 at 8:51 AM, shirish <[EMAIL PROTECTED]> wrote: > Hi all, > Please look at this bug > https://bugs.launchpad.net/ubuntu/+source/gvfs/+bug/225361 . > > Now I'm looking for a way so I can use sudo to do my stuff without it > coming in every time. > > For e.g. I'm having something like this :- > > Pass 1 of 10, 1/25 (4%): 502.1 frags/MB /usr/sbin/popularity-contest > lsof: WARNING: can't stat() fuse.gvfs-fuse-daemon file system > /home/shirish/.gvfs > Output information may be incomplete. > Fully defragmented! > > This output is from a tool called pyfrag which was made by John Dong > quite sometime ago. > > http://ubuntuforums.org/showthread.php?t=169551 > > Although using that tool may or may not be good, but getting that sort > of an output certainly isn't helpful. > > Can I do something about it. > > Specifically > > lsof: WARNING: can't stat() fuse.gvfs-fuse-daemon file system > /home/shirish/.gvfs > Output information may be incomplete. > -- > Regards, > Shirish Agarwal > This email is licensed under > http://creativecommons.org/licenses/by-nc/3.0/ > http://flossexperiences.wordpress.com > 065C 6D79 A68C E7EA 52B3 8D70 950D 53FB 729A 8B17 > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking bug for ext4
At any rate, regardless of whether we use it or not, making a bug in Launchpad is not going to change the course of action. My point was that Shirish should have first checked with the handful of core-devs and kernel devs who have a good understanding of filesystem development in the kernel tree, before filing bugs like this. On Thu, Nov 6, 2008 at 3:56 AM, Chris Cheney <[EMAIL PROTECTED]> wrote: > John, > > 2.6.28 will come out around January and Jaunty will probably ship with > 2.6.29 but it very well might not be a good idea to use it by default > for Jaunty which is what shirish seemed to be talking about. Just > because it is not considered development status doesn't necessarily mean > it is stable enough to use as the default for all Ubuntu installs. > > Chris > > On Tue, 2008-11-04 at 09:30 -0500, John Dong wrote: > > Please stop filing nonsense bugs without first understanding the > > situation. ext4 will become the default filesystem once upstream > > recommends it for adoption (i.e. 2.6.28). GRUB still does not support > > reading ext4 so we will probably need a separate /boot on ext2/ext3, > > or wait for one of the SoC projects to magically finish. > > > > There is no need to clutter the bug tracker with this. > > > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/294454
Now I remember why I didn't subscribe to this list. There's no need to e-mail the list hours after filing the bug. On Wed, Nov 5, 2008 at 7:56 PM, Isaenko Alexander <[EMAIL PROTECTED]>wrote: > Please, consider this bug > > https://bugs.launchpad.net/ubuntu/+source/xorg-server/+bug/294454 > > > > > > -- > > WBR, > > Alexander Isaenko > > ICQ 28055660 > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: kernel scheduler (was Ubuntu 8.10 significantly slower than previous versions)
As listed, the choices are noop, anticipatory, deadline, and cfq. Kernel gurus look away as I try to explain this, lest you risk dying a bit (or a lot) on the inside The default is CFQ which tries to separate IO requests by priority classes, and then provides fair timeslices to each process within each priority class. This is meant to be a good balance between low latency and high overall throughput. anticipatory used to be default up to like 2.6.9-ish and is designed for disks where seeking is costly (i.e. mechanical hard drives). The theory is, this scheduler tries to guess based on past experience after performing a disk operation in an area of a disk, whether to move on to the next request, or to sit there for a bit longer (waiting) hoping for more requests for data in the area. This algorithm is meant to reduce disk seeking, and hence from the benchmarks I've seen, the anticipatory scheduler is meant to deliver the highest net throughput but not optimal response times (latency), since it gives "unfair" preference to data within easy reach so to speak. The deadline scheduler sets a hard deadline on how much time a particular IO operation is given before taking it away and giving disk access to another IO process. It's said that this provides the best response time (latency) to a flurry of random disk reads amongst other heavy disk activity, such as opening up a document while a 20GB torrent is hash-checking in the background. However, it probably delivers the least optimal overall throughput of the schedulers above. And finally, noop simply is a null scheduler -- it makes no attempt to reorder, prioritize, or otherwise optimize IO requests -- I believe its purpose is for underlying devices which have some sort of IO scheduling capability, and/or for debugging purposes. I've heard people recommend using this for solid state media but I haven't seen any reliable sources confirming or rejecting this claim. John On Fri, Nov 7, 2008 at 6:12 AM, (``-_-´´) -- Fernando <[EMAIL PROTECTED]>wrote: > Olá Scott e a todos. > > On Thursday 06 November 2008 22:44:17 Scott James Remnant wrote: > > Also you can just fiddle on a per-disk basis, e.g.: > > echo -n deadline > /sys/block/sda/queue/scheduler > > What other options are there, and in which cases can/should them be used? > > Currenctly this is what I hae on mine: > $ cat /sys/block/sda/queue/scheduler > noop anticipatory deadline [cfq] > > > -- > BUGabundo :o) > (``-_-´´) http://LinuxNoDEI.BUGabundo.net > Linux user #443786GPG key 1024D/A1784EBB > My new micro-blog @ http://BUGabundo.net > ps. My emails tend to sound authority and aggressive. I'm sorry in advance. > I'll try to be more assertive as time goes by... > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Anyone else lost hardware buttons after update from intrepid-proposed?
Well of the listed packages, gnome-settings-daemon would be my first suspect. On Tue, Nov 11, 2008 at 6:01 PM, Mario Vukelic <[EMAIL PROTECTED]>wrote: > Hi, > > today a number of updates were pulled, aptitude log excerpt follows > below. They included gnome-settings-daemon and > linux-restricted-modules-2.6.27-8-generic, both from intrepid-proposed. > > I suppose one of them made the hardware buttons on my laptop (hp nc6440) > non-functional, though I only ever used the sound buttons and can't > vouch for the others. > > However, the list in System->Preferences->Keyboard Shortcuts is only a > long list ofin categories Sound and > Desktop. The category Window Management is ok. > > I am not sure for what package to file this in Launchpad, any ideas? > > Aptitude log excerpt: > > [INSTALL, DEPENDENCIES] linux-headers-2.6.27-8 > [INSTALL, DEPENDENCIES] linux-headers-2.6.27-8-generic > [INSTALL] linux-image-2.6.27-8-generic > [INSTALL] linux-restricted-modules-2.6.27-8-generic > [UPGRADE] gnome-cards-data 1:2.24.1-0ubuntu2 -> 1:2.24.1.1-0ubuntu1 > [UPGRADE] gnome-games 1:2.24.1-0ubuntu2 -> 1:2.24.1.1-0ubuntu1 > [UPGRADE] gnome-games-data 1:2.24.1-0ubuntu2 -> 1:2.24.1.1-0ubuntu1 > [UPGRADE] gnome-settings-daemon 2.24.0-0ubuntu3.1 -> 2.24.0-0ubuntu3.2 > [UPGRADE] linux-generic 2.6.27.7.11 -> 2.6.27.8.12 > [UPGRADE] linux-headers-generic 2.6.27.7.11 -> 2.6.27.8.12 > [UPGRADE] linux-image-generic 2.6.27.7.11 -> 2.6.27.8.12 > [UPGRADE] linux-restricted-modules-common 2.6.27-7.12 -> 2.6.27-8.13 > [UPGRADE] linux-restricted-modules-generic 2.6.27.7.11 -> 2.6.27.8.12 > [UPGRADE] update-manager 1:0.93.34 -> 1:0.93.35 > [UPGRADE] update-manager-core 1:0.93.34 -> 1:0.93.35 > > > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Default vfat file permissions - why executable?
Usecase described at https://bugs.edge.launchpad.net/ubuntu/+source/gnome-mount/+bug/78505/comments/10 As Colin said, it seems to be more of a cosmetic issue with Nautilus. On Oct 6, 2009, at 3:21 PM, Jeff Hanson wrote: > Whenever I plug in a USB FLASH drive on a system with Ubuntu 8.04 > (Hard Heron) - 9.10 (Karmic Koala) beta, all the files on it have > execute permissions. Why? Standard security practice on *nix-related > systems is to default to non-executable unless specifically needed > otherwise. I can't think of a use-case where defaulting to executable > files makes sense on vfat. If a user really needs to maintain execute > permissions on a file transfer they can just tar it first. > > I am constantly reminded by this bizarre setting every time I > double-click a text file on a vfat-formatted device from Nautilus and > am asked if I want to run or display it. > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Default vfat file permissions - why executable?
On Oct 6, 2009, at 3:27 PM, Chris Coulson wrote: >> > AFAIK you need the executable bit to be able to browse the folders. > Which is separately manageable by the "dmask" mount parameter; The fundamental problem here is on VFAT the exec bit is all-or-nothing at mount time, not per-file configurable. -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Compiz zoom scary, enabled by default in Karmic RC
On Oct 25, 2009, at 4:16 PM, Jordan Mantha wrote: > > > I've accidentally hit this zoom thing many many more times than I ever > accidentally hit the dreaded Ctrl-Alt-Backspace. I've never figured > out a good > way to get out of it other than to reboot my computer so the effect > was about > the same. I'd be all for some sort of "get me out of here" button > when it's > first activated. > > -Jordan Totally agreed, Jordan, Compiz zooming is one of the most powerful features (along with translucency and Expose-like window tiling) that 3D compositing affords, but at the same time, a user unknowingly triggering it could be a usability issue. It wouldn't be a bad idea to quickly pop up a translucency overlay that tells the user how to get out of it, not unlike VMWare Fusion's fullscreen momentarily overlays "Press A-B-C to exit fullscreen mode" -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Compiz zoom scary, enabled by default in Karmic RC
I just helped a Mac user the other day whose cat was stumbling across the keyboard as she scrolled and it ended up doing the zoom gesture. Completely unfamiliar with the feature, she was unable to restore the screen to original zoom level. So, no, while I feel super-scroll is obvious when you're actively hunting for the sequence, if accidentally triggered it can still lead to confusion without some sort of hint at least the first time activated how to restore. On Oct 27, 2009, at 3:15 PM, Joe Zimmerman wrote: I think the Super+Scrollwheel is okay (and more discovarable), but maybe the Super+MiddleClick isn't as useful? I agree -- I think Super+Scrollwheel is definitely very useful, and since the operation is continuous, it's much easier for a user to figure out how it happened and how to use it. (In fact, Control- Scrollwheel works the same way on Mac OS.) -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Java deb package
Doing so probably upsets some Sun legal fairies On Nov 4, 2009, at 5:18 AM, Onkar Shinde wrote: > On Tue, Nov 3, 2009 at 8:45 AM, Evan Hazlett > wrote: >> Greetings... >> >> Is there any way to bypass the license agreement for the sun-java6- >> jre >> package making it possible to silently install the package? >> > > Try this. > sudo debconf-set-selections > > And then inter following lines. > sun-java6-bin shared/accepted-sun-dlj-v1-1boolean true > sun-java6-jre shared/accepted-sun-dlj-v1-1boolean true > sun-java6-jre sun-java6-jre/stopthreadboolean true > sun-java6-jre sun-java6-jre/jcepolicy note > sun-java6-bin shared/present-sun-dlj-v1-1 note > sun-java6-jre shared/present-sun-dlj-v1-1 note > > Press Ctrl + D to finish the process. > > > Onkar > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: The 9.10 boot loader progress bar
The Upstart event-driven bootup doesn't really have the notion of progress, unlike the old SysV Init script bootup. It's hard to provide a linear measure of progress... The same thing happened in OS X -- in 10.4 they introduced a parallelized init daemon and the "progress" bar was just a simple timer (counting to how long it took the LAST time to boot up), and in 10.5+ they just got rid of it altogether. The same can be said about Windows and their parallelized startup control mechanism. On Jan 25, 2010, at 7:09 AM, Amahdy wrote: > I'm wondering why starting from 9.10 the boot loader started to be an > infinite loop progressbar (like windows always does)?? it was better in > previous versions that I'm able to track the percent of loaded and what's > remaining, or at least know that there is a progress (in slow computers) > instead of staying like that without knowing whether the loader hangs and > will never load or there is a slow progress and the user will watch waiting > it ... > > If this was because gathering the information before starting the loader for > a progress-bar tracking takes some time, then I believe +|- 2 seconds is not > the big deal that prevents this feature from existing ... or what's your > thoughts? > > One more thing is that, the loading could be cooler for some ppl (like me) if > it was text-based not graphical (like before that or like servers), to watch > out what's happening (eg. problems, errors, notices ... etc) so if there was > a key (like ESC) to switch, and an option to make it permanent text-based > boot. > > > > -- Amahdy AbdElAziz > IT & Development Manager > 3D Diagnostix Inc. www.3ddx.com > http://www.linkedin.com/in/amahdyabdelaziz > > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: The 9.10 boot loader progress bar
On Jan 25, 2010, at 11:19 AM, John Moser wrote: > On Mon, Jan 25, 2010 at 10:57 AM, John Dong wrote: >> The Upstart event-driven bootup doesn't really have the notion of progress, >> unlike the old SysV Init script bootup. It's hard to provide a linear >> measure of progress... > > This is why I disable 'quiet' ... my boot screen is like > > > Loading kernel modules OK > Mounting drives... OK > Starting networking... OK > > > It's familiar, and when something stalls it's suddenly not familiar. > I don't have to care WHAT it's doing, just as long as it's doing > something, and telling me what it's doing. Apple used to do this in > System 7 and System 8 at least by showing icons during boot, > signifying what part of the boot process it was currently in. There is no "part" of bootup progress anymore. Everything happens together in parallel as long as its dependencies are met, and can be arbitrary order during bootup. IO traffic in an unrelated bootup job can cause a seemingly small other job to "stall". It's not at all surprising that non-linear booted OS'es like OS X 10.4+, Ubuntu with Upstart, Windows 2000+, etc do not attempt to show a linear progress bar. -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: The 9.10 boot loader progress bar
On Jan 25, 2010, at 11:48 AM, Joe Zimmerman wrote: > On Mon, Jan 25, 2010 at 11:41 AM, John Dong wrote: > > It's familiar, and when something stalls it's suddenly not familiar. > > I don't have to care WHAT it's doing, just as long as it's doing > > something, and telling me what it's doing. Apple used to do this in > > System 7 and System 8 at least by showing icons during boot, > > signifying what part of the boot process it was currently in. > There is no "part" of bootup progress anymore. Everything happens together in > parallel as long as its dependencies are met, and can be arbitrary order > during bootup. IO traffic in an unrelated bootup job can cause a seemingly > small other job to "stall". > > It's not at all surprising that non-linear booted OS'es like OS X 10.4+, > Ubuntu with Upstart, Windows 2000+, etc do not attempt to show a linear > progress bar. > > Why not show a panel of greyed-out icons at the bottom of the screen, one for > each major component, and light up each icon as the corresponding component > is initialized? This would reflect the correct abstraction (and if the icons > had captions, users would have some idea of what was wrong if the boot > process stalled). > I agree with you, a lightup panel of Upstart jobs with status indications is a really good bootup visual. I'd like to see this feature implemented. Any volunteers? :) -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: regular fsck runs are too disturbing
A partial check doesn't make sense with the current fsck tools AFAIK. We should do a full filesystem check if anything, and if a user decides to abort it, it's his choice. There should be a graphical or otherwise easily accessible way of re-touching the /forcefsck flag so that users can choose which bootup to do a check on. Another "idea" is on LVM-capable systems to take a snapshot of important filesystems while they are unmounted or read-only then fsck the snapshot device as a background task. If any serious errors are detected in the snapshot, then schedule an uncancelable boot scan. I agree with everyone who says that the current fsck experience is a blemish to Ubuntu's general user-friendliness, and also that we should not be entirely removing the regular fsck as it catches hardware irregularities and potential software bugs with ext3. John On Wed, Oct 10, 2007 at 01:25:54PM +0200, Vincenzo Ciancia wrote: > Problem is that users will just skip the test, and get tired of "having" > to skip the test each time. Perhaps an alternative would be to check > only a part of the filesystem (e.g. randomly choosen) each time, but I > don't know enough about filesystem (even though I should :) ) to say > it's impossible or feasible. > > Vincenzo pgp05N5f7HUAd.pgp Description: PGP signature -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: regular fsck runs are too disturbing
The main roadblock in my mind is that few people use LVM as the main installer doesn't support it. Also, I have no idea how sane this idea is in terms of the abilities of ext3. It's an interesting solution but probably too insane to ship in a distro. Something like autofsck is easier/less risky to implement. Since ext3 has a fsck-on-boot-if-dirty flag, we also have to deal with some sort of usplash hook for fscking during bootup too. On Wed, Oct 10, 2007 at 03:49:45PM -0400, Bryan Haskins wrote: > Honestly it is a tad complex but it is REALLY a cool idea. > > We should write up a formal spec and see where it goes, still needs some > development, but it's really promising In my opinion. -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Restricted tab-completion is annoying
On Thu, Oct 11, 2007 at 08:23:38AM -0700, Aaron C. de Bruyn wrote: > If I modify them, doesn't that mean they will get overwritten by the next > update to the bash package? > No; it is a configuration file, which means dpkg will prompt you whether or not to replace the file. You can choose not to. > I know the reasons for program-specific tab completion--I love it with svn, > but it is annoying when you are trying to tab-complete a filename and it > won't do it because the file doesn't end with PDF or pdf for example. > > To me it seems like this is heading down the windows route. Give all your > file names a 3 character extension so you know what to open it with. > Shouldn't it be more 'unixy' and be based on the mime-type of the file? > You're right, the heuristic for evince is a bit flawed. mime-type would be foolproof, but imagine how long tab would take on a directory with 10,000 files if the mime-type of each file had to be checked, rather than just the name. > I maintain that the bash prompt is not supposed to be a user-friendly > environment. It's the bash prompt afterall. It's where admins, power-users, > and all-around geeks can go to do advanced stuff. > > I don't think it's a good argument to say that people need to have > user-friendly hand-holding at the command prompt. If I want to run 'evince > somefile.asp' I should be able to. I don't care if the extension is .asp > .pdf or .mystupidfile. > For the most part I value a degree of user-friendliness in the shell that gets my work done faster. I work 90% of my time in the CLI and believe that my shell (zsh) tab completion boosts my overall productivity. Instead, there should be convenient/intuitive ways of disabling this functionality where it is not desired. pgp5SpuBoY7Nu.pgp Description: PGP signature -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: 4 More days...
This is not very constructive. All of us here put our heart and effort into the distro and comments like this don't help. Exactly what things are bug-ridden that need attention? It's one thing to raise awareness of last-minute important bugs, but this seems to be nothing more than flamebait John On Sun, Oct 14, 2007 at 04:31:10PM -0700, Scott (angrykeyboarder) wrote: > till the release of the most bug-ridden Ubuntu release yet (unless the > devs go into overdrive in the next few days)! signature.asc Description: Digital signature -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Untrusted software and security click-through warnings
I don't think it'd hurt if we had a warning in gdebi when installing a .deb not from or signed by the Ubuntu Archive key, to the likeness of "Installing packages not from Ubuntu repositories can introduce software bugs, upgrade conflicts, or security vulnerabilities. Make sure you trust the origin of this package" Of course, I think most people will click through that anyway, but at least then we can't say we didn't try. On Mon, Oct 15, 2007 at 05:31:23PM +0100, Ian Jackson wrote: > What I'm suggesting is that if they want to do that they should be > required to do something a little more complicated which is more > likely to trigger an actual decisionmaking process. Like, for > example, typing random commands they found on a webpage. > > Ian. signature.asc Description: Digital signature -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
What is blocking the lrm nvidia-glx-new postrm bug?
https://bugs.edge.launchpad.net/ubuntu/+source/linux-restricted-modules-2.6.20/+bug/106217 This bug has been reported for around 6 months, the cause (typo in a postrm file) has been diagnosed, but it's been sitting there with no developer activity for a good deal of time. I don't mean to be sitting here and waving the magic wand so to speak, (*waves at Hobbsee*) but this one seems trivial to fix. I know it's too late to target Gutsy's release, but it might be candidate for a SRU, and it should definitely be fixed before Hardy. Thanks, John signature.asc Description: Digital signature -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: UI for backports usage
No, IMO the UI, underneath, should be adding the entire backports repository, just all packages pinned back. A UI should be able to unpin specific versions of each backport to let it through, but not let through other backports unless needed to satisfy dependences... An apturl approach would require the archive to create a separate section for each package and a meta "all" section like Debian Backports, which I guess can be done but is more complex. John On Tue, Oct 23, 2007 at 04:06:32PM +0300, Teemu Heinämäki wrote: > What about getdeb with apturl? It can remove the external repositorys > from sources.list after you install from there if you wish. Maybe it can > be modified to do so automatically? > https://wiki.ubuntu.com/AptFirefoxFileHandler > > ti, 2007-10-23 kello 07:22 -0400, Scott Kitterman kirjoitti: > > On Tue, 23 Oct 2007 12:10:06 +0200 Daniel Holbach <[EMAIL PROTECTED]> wrote: > > >Hello everybody, > > > > > >I had some discussions with John Dong and João Pinto and both > > >acknowledged the fact that there's a need for installing just a select > > >few backport packages and not all of them. > > > > > >Let's imagine that somebody has an interest in just the latest version > > >of blender on his stable system. That user does not want to upgrade to > > >all packages available in the backports repository, but just the > > >backported version of blender. > > > > > >How could something like this be realised in a user-friendly way? > > > > > I think that between the getdeb web U/I and gdebi there is a very > > reasonably user friendly approach. I think from a U/I perspective it's > > only major lack is around automated dependency management. > > > > Scott K > > > > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: UI for backports usage
Backports so far have been marketed/branded as Backports, and on Launchpad and the Wiki are both referred to as Backports. If we want to change the name of the project to facilitate user familiarity, it has to be done across the board. The project is fueled primarily by user input (i.e. what packages for us to triage) and hiding the terminology right now would lead to confusion as to how to get more of these. Perhaps we can use the term "Community Updates" or something else that doesn't necessarily label it as dangerous (often times -- if not most -- they are not dangerous to use). Perhaps another important part of a Backports UI is the ability to *request* backports too. On Tue, Oct 23, 2007 at 05:20:47PM -0500, Anthony Yarusso wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Kristian Erik Hermansen wrote: > > On 10/23/07, Daniel Holbach <[EMAIL PROTECTED]> wrote: > >> I had some discussions with John Dong and Jo?o Pinto and both > >> acknowledged the fact that there's a need for installing just a > >> select few backport packages and not all of them. > >> > >> Let's imagine that somebody has an interest in just the latest > >> version of blender on his stable system. That user does not want > >> to upgrade to all packages available in the backports repository, > >> but just the backported version of blender. > >> > >> How could something like this be realised in a user-friendly way? > >> > > > > I think that most people will be searching via Synaptic. I think > > there should be an option in the default installation of Ubuntu, > > such that when you search for 'blender' or any other package that > > is available via a backport repository, Synaptic offers the ability > > to right-click the package name and say 'Install the bleeding-edge > > version' or something similar. I think only tech-savvy guys will > > utilize this feature, and the word 'bleeding-edge' will imply that > > it could be dangerous. You should not have the mention the word > > 'backports' to the end user as this is a term unknown to those > > unfamiliar with software development... > We may want to note that backports aren't necessarily the same as > "bleeding-edge", since they may be completely stable, but are simply > newer than the release. I'd hardly call something released the day > after feature freeze for Dapper to be bleeding edge in most cases now. > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.6 (GNU/Linux) > > iD8DBQFHHnO+KlAIzV4ebxoRAiLvAKCMtlhmBPFfBp1pgzV1LJ3AyWJ1EACgyMaC > dhjeWy9ddiPaA9ZtPRY1X8A= > =VW4p > -END PGP SIGNATURE- > > > -- > Ubuntu-devel-discuss mailing list > Ubuntu-devel-discuss@lists.ubuntu.com > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: UI for backports usage
Wow, never knew about NotAutomatic -- that sounds great! On Wed, Oct 24, 2007 at 08:39:12AM +0200, Reinhard Tartler wrote: > John Dong <[EMAIL PROTECTED]> writes: > > > No, IMO the UI, underneath, should be adding the entire backports > > repository, > > just all packages pinned back. > > This does not necessarily need pinning via /etc/apt/preferences. Apt has > a feature which does something similar. Please compare > > http://archive.ubuntu.com/ubuntu/dists/gutsy-backports/Release > > with > > http://backports.org/backports.org/dists/etch-backports/Release > > wrt to the NotAutomatic: Yes field in bpo. > > I always wondered by ubuntu backports don't have that field. > > See http://backports.org/dokuwiki/doku.php?id=instructions for > instructions how to use them. If backports were NotAutomatic: yes, then > the instructions would reduce to step 3, i.e.: > > apt-get -t gutsy-backports install "package" > > > Synaptic and aptitude already offer a nice interface for selecting > specific version (i.e. backports for instance). > > John, does this satisfy your concerns? > > -- > Gruesse/greetings, > Reinhard Tartler, KeyID 945348A4 > -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
