Re: EFF Privacy; hopefully Ubuntu will listen to users
On 05/11/12 15:08, Jordon Bedwell wrote: We've just had the Ubuntu Developer Summit during which the next release was planned, and everyone was welcome (both in person and online). I must have missed the session on privacy, or did nobody propose one? I don't think there was one, I think this is a case of the few speaking and protecting the many and the few not having the same power as the many because some people won't do anything until the many step up and embarrass the top brass. What I am saying is, at this point I am to believe that Canonical and Ubuntu do not care one bit about this privacy cock up and they don't care that the few notice and are trying to help the many. They are probably gonna hold off until the many step up and embarrass Canonical. Or is it the case that nobody bothered to file a blueprint? Bear in mind that *anybody* in the community can create blueprints for UDS, not just Canonical. The last time I looked at the list of blueprints that had been filed for UDS-R, I can't remember seeing one on privacy. It would have made an interesting subject for the community track and would have officially documented the issues, what can be done about them and how they will be addressed. Someone much smarter than I am said in a recent blog post something along the lines of the best way to ensure a process doesn't ignore your needs is to engage with it. So the best way to get something done in Ubuntu is to engage with the development process, which means to engage with UDS where the development schedule for the next 6 months is worked out, which in turn means writing a blueprint for consideration at the next UDS. It's a couple of weeks late for UDS-R but what about creating a blueprint for UDS-S? Get the discussion going, gather examples of privacy issues and what could be done to address them. Then at the next UDS, we can work out solutions that satisfy Canonical, privacy conscious users and the rest of us and that can be implemented in time for 14.04 LTS. Disclaimer: I do *not* work for Canonical and only have limited experience of UDS as I only took part for the first time last week. Cheers, Bruno -- 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: Post-natty changes to unity [Was: Re: Ubuntu Gnome 3.4]
On 06/01/12 15:58, Evan Huus wrote: On Fri, Jan 6, 2012 at 10:41 AM, James Haighjames.r.ha...@gmail.com wrote: I was a keen Unity fan when Natty came out. However, since Oneiric, Unity is the main reason why I'm still using Natty. I find that very interesting. My personal opinion (and the general majority opinion, as far as I can tell) is that the Oneiric changes to unity were clear improvements. I second that: I installed Oneiric beta on my new laptop rather than Natty because Unity in Natty was flaky with my hardware. Bruno -- 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: Feature suggestions: optionally placing home folder into separate partition during ubuntu install
On 26 December 2010 10:55, Phillip Susi ps...@cfl.rr.com wrote: This is what manual partitioning is for. Also /home can not be on NTFS since it does not support ownership and permissions. Well, yes and no: manual partitioning is for advanced users who know what they are doing. The fact that it's the only way you can install Ubuntu with a separate /home partition isn't easy for the average user, who doesn't know what the benefits of a separate /home are in the first place. What would help is a simple option such as Install Ubuntu on the whole disk using separate partitions for user and operating system files, which would choose sensible defaults for /, swap and /home and create the partitions accordingly. A look at other distros that already do that could help as highlighted in this review from The Register: http://www.theregister.co.uk/2010/10/11/ubuntu_10_10_review/ Bruno -- 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: Emergent: Oracle's behavior re Java
On Fri, 2010-08-13 at 09:11 -0400, Evan wrote: On Fri, Aug 13, 2010 at 9:00 AM, John Moser john.r.mo...@gmail.com wrote: snip Something to keep in mind is that this won't just affect Ubuntu. If this does become a major issue, Debian will have to deal with it, Mono (as a project/community) will have to deal with it etc. Since we effectively inherit from all of these projects who will be much more directly affected, I think our best course of action is simply to keep on eye open, and be prepared to participate in any discussions that happen upstream. To be honest, I don't think Ubuntu has to worry about this. Irrespective of whether Oracle's actions are justified or not, the underlying reason why they are going after Google is that the version of Java that runs on Android is very bastardised, is incompatible with any Java edition (JME, JSE or JEE) and as a result fragments the platform. If Google had developed a JME implementation same as what Nokia, Motorola et al have done for years, Oracle would have said nothing. So I don't think any open source implementation of the JVM is at risk, as they all implement a JDK and JVM that are compatible with Sun's: you can run code built with GCJ or OpenJDK on the Sun JVM and vice versa. This is not true of Dalvik. At the end of the day, love them or hate them, Oracle are no fools: they will go after who they think is harming their business. Proprietary and open source implementations of the JVM that follow the specs and are compatible with the Sun JVM are good for Oracle because they help spread Java. Android's implementation is bad for Oracle because it fragments the Java mobile market. In fact, I was surprised that Sun didn't start that action when Google originally came up with Dalvik, the same way they sued Microsoft all those years ago when they tried to extend Java 1.0. I suppose Sun realised at the time that they didn't have the financial strength for an expensive lawsuit against Google, while Oracle certainly do. My £0.02 Bruno -- 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: Emergent: Oracle's behavior re Java
On Sat, 2010-08-14 at 03:12 +0530, Manish Sinha wrote: snip They are going after Google as they think they can make a huge amount of money since Google has deep pockets. If Google bends, then the patent deals would be a huge monetization for Oracle. As I said, DVM is not JVM and has different instruction set. Oracle needs to prove that their patents are being infringed in DVM. I don't get about fragmentation. The dev end code is same, the VM is different. The dev cares more about the code *if* he knows that it's write-once-run-anywhere. Well no, the code is not the same: DVM doesn't ship the whole of the standard API and adds bits of its own, in particular the UI packages. Also, DVM is the only JVM for which a dev has to compile specifically because the byte code is different. For any other implementation of the JRE, the byte code is the same so you can compile your app and ship a jar file rather than source code and it will run. From a development point of view, if you want to create a Java app for mobile, you do have to target two environments: JME and Android; hence the fragmentation. Anyway, I'm not saying that Oracle is right or wrong in doing this, I'm just trying to understand the motivation behind it. Bruno -- 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 0.8.4-0ubuntu2.1 alleged bugs report
On Wed, 2010-06-02 at 17:50 +0200, Emmanuel Michon wrote: Hello, I'd like to report few bugs I believe are due to the compiz component on a fresh ubuntu amd64 If this mail should not be sent here, please direct me to the proper location. Hi Emmanuel, Thanks for your interest in reporting issues with Compiz. Can you file the bugs in Launchpad under the compiz package please? You can do that easily from the command line by typing: ubuntu-bug compiz Note that you will have to register in Launchpad when you file the bugs. This is so that bug triagers can contact you if they need more information. Bruno -- 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: Proposal: DNS cache by default for ubuntu-desktop
On Wed, 2010-06-02 at 09:50 +0100, Conn O'Griofa wrote: Hi Kurt, [snip] As you can see from the community instructions, you need to edit your /etc/resolv.conf and add the local machine's address (127.0.0.1) as your first nameserver. This is a problem because any changes to this file will be overwritten by NetworkManager each time it initializes a new connection (which is warned in the commented section of the file). The only solution that I am aware of, is to edit your active network connection in NetworkManager's applet, change the setting from Automatic (DHCP) to Automatic (DHCP) addresses only, and then manually input your DNS servers like so: 127.0.0.1, your regular DNS server 1, your regular DNS server 1. And this in turn would break configurations where the DHCP server provides the name of the DNS server to clients when they connect. Such a configuration is extremely common for any network where people are transient (public and semi-public networks like libraries, company networks where users move from office to office, etc). If we are going to integrate dnsmasq into the default installation it may serve us better to investigate the possibility of adding some kind of autodetection of the presence of dnsmasq to NetworkManager itself. We cannot expect users to manually edit their network connections for each new installation (or indeed, each new wireless network profile they create). I second that: integrating whatever solution (like dnsmasq) with NM is essential. Let me also mention that dnsmasq has the capability of providing ICS (internet connection sharing) [3]. This would be an excellent feature to add to NetworkManager's graphical interface, which is yet another feature that competing operating systems provide by default. It would be very nice indeed. Bruno -- 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: Rethinking Ubuntu's Repositories
On Mon, 2010-05-24 at 09:01 -0400, Evan wrote: On Mon, May 24, 2010 at 5:08 AM, Conrad Knauer ath...@gmail.com wrote: I like the repository system that Ubuntu uses, but I feel that there is a problem with it and I have a suggestion as to how to fix it. ~ The Problem ~ Ubuntu inherited the Debian system of updating software versions with OS upgrades. This makes the most sense when you have many many packages that are slow in updating (e.g. due to code maturity) and/or you are upgrading your OS relatively frequently. An example of where this is a bad idea is Firefox, especially on LTS releases; an excerpt from http://packages.ubuntu.com/search?keywords=firefox showing the releases still supported on the Desktop: Package firefox * hardy (web): meta package for the popular mozilla web browser 3.0.19+nobinonly-0ubuntu0.8.04.1 [security]: all * jaunty (web): meta package for the popular mozilla web browser 3.0.19+nobinonly-0ubuntu0.9.04.1 [security]: all * karmic (web): meta package for the popular mozilla web browser 3.5.9+nobinonly-0ubuntu0.9.10.1 [security]: all * lucid (web): safe and easy web browser from Mozilla 3.6.3+nobinonly-0ubuntu4: amd64 i386 Most Firefox users have already moved to version 3.6 (see the graph on http://gs.statcounter.com/#browser_version-ww-monthly-200904-201005) which is where Mozilla wants you to be also BTW. Getting a new version of Firefox on an old version of Ubuntu can be a pain. Supporting Firefox 3.0.x which is no longer supported by Mozilla (see http://en.wikipedia.org/wiki/Mozilla_Firefox_3) seems silly. PPAs are unofficial. Mozilla doesn't have a DEB repo and even if they do make one, they might not offer packages other than for x86-32. ~ A Solution ~ Now, assuming that there are no technical reasons why Firefox 3.6 can't be built for all the currently supported versions of Ubuntu, we can do the following for future releases; get rid of the main repo as it currently stands and replace it with two repositories: (1) a 'core' which will represent everything up to and including Gnome (for Ubuntu; KDE for Kubuntu, etc.), so to a working GUI including some basic apps (like Totem). This is stuff that most users assume will just work and don't want to fiddle with or upgrade for a while once they're working right. If Ubuntu is a 'software libre supermarket', these are the canned, dried and frozen goods that have a moderate to long shelf life. This repo should retain the 'main' designation. (2) the 'desktop' applications currently in main that people really would like to stay current. Especially Firefox, but also OpenOffice, GIMP, etc. (that is, the 'big' ones (usually recommended by the ubuntu-desktop metapackage, or otherwise in main) that aren't part of Gnome proper...). In the supermarket analogy, these are the big showy fresh fruit displayed at room temperature. Perhaps a line in the sources.list could look like this: deb http://archive.ubuntu.com/ubuntu-desktop maverick main In 'main' cases like Firefox where you can have two versions that are officially supported for a time, have a metapackage (e.g. firefox) pointing at the newest release, but the actual versions in the names of packages that contain data (e.g. firefox-3.5 and firefox-3.6). This will allow users to pick if they would rather transition automatically to the latest version or maintain the old version *while it is still supported* (e.g. for businesses which tend to be slower in adopting new versions... also, for people like my wife who bitterly complain that new releases always break things... e.g. Firefox extensions) since desktop software seems to have unpredictable release cycles very much not in synch with Ubuntu's. In the case of Firefox (let's say starting in 2009 after Firefox 2 reached an end of life), my solution would work like this: - start 2009 firefox metapackage points to firefox-3.0 - June 30, 2009: Firefox 3.5 released firefox metapackage changed shortly to point to firefox-3.5 repository contains both firefox-3.0 and firefox-3.5 - January 21, 2010: Firefox 3.6 released firefox metapackage changed shortly to point to firefox-3.6 repository contains firefox-3.0, firefox-3.5 and firefox-3.6 - March 30, 2010: final version of 3.0 (3.0.19) released firefox-3.0 to be removed in a timely manner (a week or two?) - August 2010: final version of 3.5 to be released firefox-3.5 to be removed in a timely manner (a week or two?) - late 2010: Firefox 4.0 (hopefully ;) releases firefox metapackage changed shortly to point to firefox-4.0 repository contains firefox-3.6 and firefox-4.0 etc. ~ Misc. Thoughts ~ Splitting out the desktop apps would mean that old LTS releases (like Dapper, which is expired for the desktop but still supported for the server)
Re: Remove OO Draw from the default install
On Sun, 2010-05-16 at 15:18 +0200, Aurélien Naldi wrote: On Sun, May 16, 2010 at 1:27 PM, Shane Fagan shanepatrickfa...@ubuntu.com wrote: Hey all, I forgot to mention this at the session for default app selection but can we remove Open Office Draw from the default ubuntu install? The reasons are quite obvious it just isnt any good and I dont think any of the regular users actually use it. This probably deserves some discussion. I'm not a huge fan or openoffice in general for various reasons but it seems to be the best free software available for a wide audience (LaTeX, R and other great tools are way too specialised and techie). Back to OOo draw: it seems to me that it is just impress without the effect parts and as such I don't think it uses much space. For the record I do use it (mostly to do simple drawings, export them as pdf and insert them into latex document, so I guess I'm not the main target here...). I agree with this and I would like to add a few extra points. OOo Draw is not the best standalone vector drawing tool around, that's true. On the other hand, it's one of the most accessible for casual users and it is also the best tool when you use it as part of the OOo suite, in collaboration with OOo Writer, Impress or Calc. For instance, I use OOo Draw all the time to produce diagrams that I subsequently include into OOo Writer documents (that generally end up as Word or PDF docs). OOo Draw provides the simplest workflow for that type of usage, which I encounter all the time in business environments. So when it comes to Shane's statement it just isnt any good, I'd say it depends from what point of view. If you are a graphic artist, indeed you need a more elaborate tool and you probably expect to have to download such a specialist tool. If you are a business or home user who just wants to include the occasional drawing into a word processing document, OOo Draw is exactly what you need. Also note that OOo Draw has its quirks but once you understand how things work (such as the colour palette management), it is actually quite good. I first used it under duress because it was the only tool that supported the workflow I needed and I had low expectations but I have been pleasantly surprised along the way. One last that point I want to make it that removing OOo Draw from the default install would mean that you would no longer have a vector drawing tool in the default install (apart from OOo Impress but that would be misusing it to do a job it's not really designed to do, in the same way that thousands of Windows users routinely misuse Powerpoint to include drawings in their documents because they don't have any real vector drawing package at their disposal). In conclusion, I think OOo Draw fits well in the Ubuntu application ecosystem and deserves to be installed by default because it provides casual users with an adequate drawing package that works out of the box, is reasonably intuitive to use and integrates well with the rest of the office suite. I don't mind installing extra software so removing it would be OK for me, but only if it does allow a huge space gain, which I doubt (the size of the .deb isn't a good hint here as impress is tiny and depends on draw). I agree to that too. And I think that the benefits of removing it without crippling Impress are too small compared to the downsides of doing so (namely: working out how to de-couple it from Impress and not break anything as well as the reasons detailed above). Bruno -- 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: Latest Lucid kernel (2.6.32-20) chokes on my hardware
On Mon, 2010-04-12 at 12:09 +0530, Vishal Rao wrote: Bruno, What are your hardware specs and SSD model? Try passing libata.force=noncq as kernel boot param and see if it helps? Helps with my Intel DP5DP motherboard and Crucial M225 SSD (IndiLinux barefoot controller/firmware)... HTH Vishal Thanks Vishal, I'll try that. The SSD is a Transcend TS64GSSD25-M: I needed a PATA drive as it's an old laptop and that was the best model I found. I notice that your patch[1] talks about ATA errors in dmesg. How do you identify those if there are any? I don't find anything that looks out of place when I run: dmesg | grep -i ata Anyway, I'll try to dig a bit deeper and will post my findings on the forums which, as Luke says, is probably the better place for that sort of discussions. Bruno -- 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
Latest Lucid kernel (2.6.32-20) chokes on my hardware
Hi all, First of all, apologies for cross-posting to different lists, I'm just not sure where is the best place to ask this question. I upgraded my laptop to Lucid beta 2 today, in an attempt to test-drive it. It all worked fine until the system updated the kernel from 2.6.32-19 to 2.6.32-20. With *-20, it won't boot anymore. If anybody could give me some pointers so that I can investigate that problem further and file a bug report that is more specific than computer says no, that would be very much appreciated. I'm happy to dive into the bowels of the OS if need be so don't hesitate to make it technical if required. The laptop in question is an IBM ThinkPad T42. The system seems to fail very early in the boot sequence, as if it didn't manage to recognise the disk, which may be part of the problem as I replaced the original HDD with an SSD when I upgraded to Karmic 6 months ago. I also upgraded the BIOS of the machine to its latest version today. I'm not sure whether that can have an impact. Thanks in advance, Bruno Girin -- 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: Thoughts on quitting and window controls
On Tue, 2010-04-06 at 21:05 -0700, Akkana Peck wrote: Jonathan Blackhall writes: It's very confusing for me when I click the big 'X' in my window controls, only to find that the application I was attempting to close has since been minimized to my system try (or notification area or its respective indicator applet or wherever it goes instead of quitting). I hit that recently with gwibber. Thought maybe it could be a good alternative to the difficult Facebook website ... but then I quit it, or at least told it to quit, and later found random gwibber windows popping up. The result was that it taught me not to use gwibber for anything, because running it even to try it out means a hassle later when I have to run ps to find it and kill it. At least Gwibber gives you the choice: just go to Gwibber - Preferences and you can check / uncheck Minimise to tray on close. Bruno -- 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: Medical Task Force
Hi Sasha, On Thu, 2010-04-01 at 23:32 +0200, Sascha 'saigkill' Manns wrote: Hello Mates, warm Greetings from the openSUSE Community. We all knowing that we have a lack of Software from Medical needs (Medicine Doctors or Clinics). So we (the openSUSE, Fedora and Debian Developers) have founded an Special Task Force for building new Packages with Medical needs. We trying to share our Experience and Patches. So we are pleased to invite the Ubuntu Project, to be a part of this Task Force. If you're an Packager or interested in learning that, you're welcome. Please answer to this List. I have no knowledge in packaging but I would be very interested to learn and to be part of this task force. I have experience working with the NHS in the UK, hence the interest. I'm not sure that's any help though :-) Bruno -- 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: A bug of Linux in Ubuntu 10.04
On Fri, 2010-03-26 at 18:07 +0530, Chandru wrote: You seem to have been bitten by this Virtualbox bug. http://www.virtualbox.org/ticket/6100 The fix is available in the Lucid's Virtualbox . Yes but that's not much use for people who use Karmic and want to test drive Lucid in a VM to make sure it works on their hardware before committing to the upgrade or to contribute to Lucid beta testing without upgrading their production box. Is there any chance the fix could be backported to Karmic? -- Chandra Sekar.S On Fri, Mar 26, 2010 at 6:02 PM, Bill Lee billleecn2...@sina.com wrote: First of all, as a foreign English speaker, I may not be able to express myself complete. If you have difficult understand this mail, I feel sorry. I ran Ubuntu 10.04 on Virtual Box OSE, and find that the ACPI part of Linux Ubuntu 10.04 cannot work with Virtual Box OSE. System will hold up when loading Linux kernel. But if I send parameter 'acpi=off' to the kernel or disable the ACPI support of the virtual machine, the system will work. The version of Linux is '2.6.32-16-generic #25-Ubuntu SMP Tue Mar 9 16:33:52 UTC 2010'. The Virtual Box OSE is running in Ubuntu 9.10 and its version is '3.0.8_OSE r53138'. Ubuntu 9.10 can work with this version of Virtual Box. So I think this problem can be sovled. I wish Lucid LTS will become better. Best wishes, Bill Lee -- 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: Indie Software Dev
On Fri, 2010-03-05 at 15:06 -0700, Charlie Kravetz wrote: On Fri, 05 Mar 2010 15:43:40 -0600 Jon Buys jonb...@me.com wrote: Hello, I posted this on the Ubuntu Forums, and didn't get a whole lot of feedback, so I thought I'd try this mailing list. I'm launching a commercial software business. I'm targeting the Mac right now, but I am curious to find out how this would be received by the Linux community. I have an ongoing conversation with a friend of mine about this, and I think that the free and open culture surrounding Linux would mean that there is no market for an indie dev to go after. He thinks that the market is here, am I wrong? Would anyone buy a $20 app on Linux? Thanks, Jon That depends on the application and perceived value. People will buy applications that are better than what is in the free stack. As Charlie says, it completely depends on the application and if it's better than the free equivalents or fills a need that is not covered by free software, why not? One area where this can and has worked is with games, as demonstrated by 2D Boy and their World of Goo game [1]. One major advantage of games is that they don't have to integrate flawlessly into the user's desktop experience as they tend to be full screen and have their own UI. On the other hand, if your software is desktop software, it may be more difficult to make a Linux version profitable because you will spend a lot of time making it play well with different desktop environments. And if integration is less than perfect, users may not want to pay for it. What will also be very important is how you price it: it needs to be low enough that it can be an impulse buy but high enough that you make a profit. $20 is a price that most people would be ready to pay on a whim for an app that actually solves a problem they have but would it be enough for you to break even? How many copies would you need to sell? If the answer is not a lot then is the app good enough to get people to buy it, even at a low price? So to answer your question, people would definitely buy a $20 app on Linux if they felt it's good enough. However, Linux users are a lot more savvy than Windows or Mac users about open source and free alternatives so the competition may be harder. On the other hand, if the app is really good, news will go round and people will buy it. Also, I'm sure Linux users will appreciate an indie software outfit that provides Linux versions of their software. [1] http://2dboy.com/ My $0.02 Bruno -- 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: White-on-black terminal should be default
On Wed, 2010-03-03 at 10:34 +0200, Lucian Adrian Grijincu wrote: On Wed, Mar 3, 2010 at 6:13 AM, Shentino shent...@gmail.com wrote: My personal opinion is that the default should be gray on black like the vc. For some reason the terminal colors just work better that way, particularly with midnight commander in a shell in a term. my two cents. Opinions, opinions. Everyone has a preferred setup. There are people who hate color ls and always disable it. And then there are others that can't tell directories and files apart in ls if they're not colored differently. You won't satisfy everyone. Either setup a poll (on brainstorm, for example) or bring some proof (a study proving one is better in this and that situation). And then, you change it and you get a lot of flak from people who liked the way it was before. The best way to cater for everybody is to make it customisable, which the Ubuntu terminal already supports. To have white text on black background in your terminal, do the following: 1. open a terminal 2. go to the Edit - Profiles... menu 3. select Default and click Edit 4. select the colours tab 5. uncheck Use colours from system theme 6. either select a built-in scheme from the drop down (such as white on black) or customise your colours using the palette. There's a bewildering array of customisation options you can play with, you can even have an image as the background or a semi-transparent colour. You can also create several profiles to use in different situations if you want to. So I reckon, the only thing to do with the terminal is to leave it alone: it's already customisable way beyond what the Windows and Mac terminals offer. Bruno -- 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: White-on-black terminal should be default
On Wed, 2010-03-03 at 16:48 +, Scott James Remnant wrote: On Fri, 2010-02-19 at 13:22 +0100, Fred . wrote: Everybody knows that a terminal has white text on black background. Everybody knows that the bikeshed is green. Seriously, there's a whole tab of preferences devoted to letting you pick what colour you want your terminals. You can have them lurid pink on green if you like. The default is the way it is because it provides the best contrast on the majority of screens (TFTs, LCDs, etc.) A lot simpler than that: the default follows the theme's colours, which makes complete sense: there's no reason for the terminal to be any different from other applications. Having said that, everybody(*) knows that a real hardcore terminal should be like the VT100 of old: green text on black background ;-) (*) everybody = me and my non-existent dog Bruno -- 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: White-on-black terminal should be default
On Wed, 2010-03-03 at 16:04 -0700, Charlie Kravetz wrote: On Wed, 03 Mar 2010 22:48:13 + Bruno Girin brunogi...@gmail.com wrote: On Wed, 2010-03-03 at 16:48 +, Scott James Remnant wrote: On Fri, 2010-02-19 at 13:22 +0100, Fred . wrote: Everybody knows that a terminal has white text on black background. Everybody knows that the bikeshed is green. Seriously, there's a whole tab of preferences devoted to letting you pick what colour you want your terminals. You can have them lurid pink on green if you like. The default is the way it is because it provides the best contrast on the majority of screens (TFTs, LCDs, etc.) A lot simpler than that: the default follows the theme's colours, which makes complete sense: there's no reason for the terminal to be any different from other applications. Having said that, everybody(*) knows that a real hardcore terminal should be like the VT100 of old: green text on black background ;-) (*) everybody = me and my non-existent dog Bruno That was a black background? ;) Seen in the deficient lighting of the computer room at 4 o'clock in the morning, it looked black. In natural light, it was probably the darkest shade of green that technology could produce at the time but I couldn't tell as I don't remember seeing much daylight when I was using those ;-) Bruno -- 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: Ask for a nickname in users-admin; Was: Should Short really be username when creating a user in users-admin
On Sun, 2010-02-28 at 09:53 -0400, Derek Broughton wrote: Thomas Templin wrote: I would vote for ... and also for 'Full Name' (DE_de: 'Vollständiger Name') or better 'Name Surname' (DE_de: 'Name Nachname') You'd know better than me if that's better in German - but it's certainly wrong (confusing) in English. Translations don't (and shouldn't) have to be literal. Full Name works in English. Other options like Forename Surname, Given Name Surname are clumsy (largely because we can't decide what to call the first name). Name Surname suggests you are asking just for a Surname (as in Name: Surname), and your users will either presume the short or login name is the given name, or be waiting for another prompt. Agreed. Add to this that the concept of name, surname, first name, last name, etc is also language and culture specific and it's a lot better to keep it generic and simple. Here are some examples of full names from different places: US: Harry Downey Jr = given name + family name + suffix Spanish speaking world: Gabriel Garcia Marquez = given name + 2 family names (from the mother and the father) Middle East: Mohammed bin Rachid al Makhtoum = given name + father's name + family (tribe) name Asia: Mao Zedong = family name + given name Europe: John Smith = given name + family name So Full Name is the only solution that really works. As for the other one, Short Name tries to be less technical than Login Name or User Name but becomes very confusing as it's not something that people would recognise: you never get asked for your short name in normal life so why would your computer ask you for it? It makes sense to ask for something technical like login name or user name because it is then obvious that it is related to the computer. Furthermore, the concept of login name or user name is consistent with what you're being asked for on all web sites that require you to register. Nickname is not a good substitute either because a lot of people don't consider themselves as having a nickname or may not feel like they want to use to nickname they were given at school, even they like it. If there is an issue of making people understand what a user name is, it might be better to have a short piece of explanatory text above the field, such as Your computer needs to be able to recognise you as a user, so you need to provide it with a name you will use every time you want to use it. (I don't like that sentence but I can't find anything better) My £0.02 Bruno -- 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: ubuntu : A mantis problem while updating
On Thu, 2010-02-18 at 10:23 +0100, Cédric BRIFFON - AIROD Technologies wrote: Hello, i'm working on an ubuntu server , and sometime using mantis. Since few months, i have a problem with the mantis package. The updates are not working properly. You can found the log when i try tu update. Sorry, it's a french log :/ I have search someone who have the same problem, but google didn't help me. So i try to write you this email. what can i do to correct this error ? Préconfiguration des paquets... alias: /mantis not found alias: /usr/share/mantis not found Échec de la préconfiguration de mantis, avec le code d'erreur 1 (Lecture de la base de données... 313416 fichiers et répertoires déjà installés.) Préparation du remplacement de mantis 1.0.8-4 (en utilisant .../mantis_1.0.8-4ubuntu0.1_all.deb) ... alias: /mantis not found alias: /usr/share/mantis not found dpkg : avertissement - ancien script pre-removal a retourné un code d'erreur de sortie 1 dpkg - tentative d'exécution du script du nouveau paquet à la place ... alias: /mantis not found alias: /usr/share/mantis not found dpkg : erreur de traitement de /var/cache/apt/archives/mantis_1.0.8-4ubuntu0.1_all.deb (--unpack) : le sous-processus nouveau script pre-removal a retourné une erreur de sortie d'état 1 alias: /mantis not found alias: /usr/share/mantis not found dpkg : erreur lors du nettoyage : le sous-processus post-installation script a retourné une erreur de sortie d'état 1 Des erreurs ont été rencontrées pendant l'exécution : /var/cache/apt/archives/mantis_1.0.8-4ubuntu0.1_all.deb E: Sub-process /usr/bin/dpkg returned an error code (1) Un paquet n'a pas pu être installé. Tentative de récupération : I am cross-posting this to the bugsquad mailing list as it's probably better dealt with there. It looks like this not an isolated problem: out of the 5 new bugs in launchpad for mantis, 3 of them are about failing to upgrade various versions of the package [1], but none for your exact version. Therefore, I would suggest you open a new bug in launchpad for the mantis package. Include the logs above and as much information as possible. Bugsquad, how should such package update bugs be triaged? [1] https://bugs.launchpad.net/ubuntu/+source/mantis/+bugs?field.status:list=NEW Bruno -- 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: Focus of the introduction for programmers
On Tue, 2010-02-02 at 19:56 -0300, Brian Vidal Castillo wrote: It could sound like a fanboy, but using a powerful IDE is the best choice for programers new to Ubuntu. And I think Eclipse is the right for this tasks. I'd agree to that. Eclipse is very powerful. I've used it extensively and it's hard to beat. Add to this that it is also cross-platform and is one of the most common IDE on Windows and Mac and a large number of developers used to work in those environments feel at home with Eclipse. There are plugins for bzr (whic I use). But a little more love from the community would help a lot. Some of the things that would be very useful are standard packages for the pydev plugin, a bzr plugin and a .deb packaging plugin. [snip] I agree with what you're saying, although imho if you're serious about becoming an Ubuntu contributor on any type of technical level you'll probably want to learn packaging. You might not want to do sponsorship work or merges or anything MOTU-specific, but knowing how to provide patches and being able to upload your own code and bug fixes seems very useful. I think they should be pointed to packaging in addition to what you propose above, and also other tools that are used across Ubuntu such as bzr. -Jonathan I agree and I think that's an area that needs work. As an experienced developer on other platforms, I find packaging to be the biggest challenge to being able to contribute code because it is Ubuntu (or rather Debian) specific and I can't relate it easily to what I know. Learning Python when you know other OO languages is not too difficult, nor is learning bzr when you know svn; but the wheels come off when it comes to packaging because producing a .deb feels a lot more complicated than producing a .jar for a Java application to the uninitiated like me. Bruno -- 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
