Japanese Input with Karmic?
Does anybody here use Japanese input as a secondary language? My computer was installed using the default English language, but I need to converse with Japanese people on a regular basis. So, I went into the System->Administration->Language Support just like I did on all the versions of Ubuntu. But SCIM does not seem to be popped up any more. When I press Ctrl-Space... nothing. Right now while in Firefox, scim says that anthy is active... does this look like Japanese. Not to me either. In order to get any Japanese support, I have to go into GEdit. Pressing Ctrl-space does not work, but I can right click and select Input Methods->SCIM Input Method, and presto, romanji is converted into にほんご (Japanese). I can then paste that into my Firefox, Thunderbird, OpenOffice, etc. So its not a language support issue, its an input issue. I need to write to our Tokyo office and need this to work. Before filing a bug... Can someone else confirm or tell me how they got theirs to work? Thanks Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 -- 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: cancel the 9.10 release... it is not ready
Remember, Lucid is an LTS release. This will have four big side effects: - Generally fewer new features (Though rumor of Gnome 3.0, and pushing for a 10 sec boot is kinda scary) - Fewer Alpha releases - Two Beta Releases - Two weeks of RC release instead of the usual one. Next release already has the controls in place to be more conservative. Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 From: ubuntu-devel-discuss-boun...@lists.ubuntu.com [ubuntu-devel-discuss-boun...@lists.ubuntu.com] On Behalf Of Shentino [shent...@gmail.com] Sent: Monday, November 09, 2009 8:44 AM To: Mohammed Bassit Cc: ubuntu-devel-discuss@lists.ubuntu.com Subject: Re: cancel the 9.10 release... it is not ready Seeing as the horse is already out of the barn, I don't think there's anything more we can do except hope for updates. Meantime, I think focusing on getting an early polishing to Lucid as a preventative measure would be a good idea. On Mon, Nov 9, 2009 at 4:03 AM, Mohammed Bassit mailto:webceo...@gmail.com>> wrote: > El 10/26/2009 12:28 PM, Mohammed Bassit escribió: > > [...] > >> I really like to persuade people to use ubuntu. But as long > >> as it looks unready it will strengthen their opinion, that linux > >> is only for nerds. Pleas learn your lesson from the debian > >> community and release a new version only if its ready. > >> > >> > > > > You are free to use Debian ! > > > After being migrated from Jaunty to Karmic in three machines (one at > work, one mine, and one of a friend), specially seeing that all that > works in Jaunty (wireless, ethernet, sound!) stop working in Karmic Can you give us more details about what went wrong with your network and sound ? was it working in Jaunty out of the box ? did it just break once you upgraded to Karmic ? And those three machines you're talking about, do they have similar network and sound cards ? are all three of them from the same make and model ? > in my end-user friend machine (I replaced her Windows Vista to Kubuntu, > and since Jaunty she was happy and I was proud of that), I seriously I don't see why you decided to upgrade to Karmic if the user was happy with Jaunty ? > considering that option. In this moment I think Debian releases are more > "user friendly" that the "for human beings" latest distribution. > I have to agree that Debian releases tend to be more stable, but I'm not sure about the user friendly thing !! > Sorry, I don't want to make FUD, but I'm really sad about it. And I see > I'm not the only one :-/ > I have seen a few people complaining about their upgrades from Jaunty to Karmic that went bad. I personally upgraded three computers to Karmic without any issues (Actually the jack sense on my work computer's sound card wasn't working in Jaunty and it works now in Karmic). Anyway, it doesn't seem like a common issue, I have only seen a few cases of upgrade fails, and in most of them it was the users "fault" (using apt to upgrade for instance). > > Best regards, > Natanael. > > > Thanks, > > Thanks, -- Mohammed Bassit mailto:webceo...@gmail.com>> -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com<mailto: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: Ubuntu Domain Server
eBox has hosed up a number of my servers network and services configuration. Ubuntu was stupid for choosing this. I always use Webmin. It works and has support for more services. Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 From: ubuntu-devel-discuss-boun...@lists.ubuntu.com [ubuntu-devel-discuss-boun...@lists.ubuntu.com] On Behalf Of Onno Benschop [o...@itmaze.com.au] Sent: Thursday, October 22, 2009 8:45 PM Cc: ubuntu-devel-discuss@lists.ubuntu.com Subject: Re: Ubuntu Domain Server On 23/10/09 10:33, Ryan Dwyer wrote: > So then that brings up the question of what web based tool should be used. Ubuntu Server has chosen eBox as that tool. -- 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 Domain Server
> I don't think there's any use discussing whether we think a GUI or CLI > is better. Shouldn't we focus on what the typical business wants and > what they're prepared to use? > > Although it seems that the first topic to discuss is how the uid/gid system > can be changed or mapped to support local and domain accounts > (forgive me if I'm not using the correct terminology). > > -Ryan I mentioned this the other day, and other than a few people making off line comments indicating that they had never heard of the product, my suggestion of GOsa got completely ignored. While not perfect by any stretch of the imagination, I think it solves the problem that is being acknowledged here, without being mentioned. That problem is that end users (despite lack of proper education), want to feel like they are in control. That is easily obtained if you just twist your thinking a little, and take a look at GOsa. GOsa is a glorified LDAP editor. A shiny, pretty, drop dead easy to use LDAP editor. It is the Debian equivalent to the Apple directory services. And it is already in the Ubuntu repositories. With GOsa, you can allow your end users to add accounts, email, ftp access, SAMBA access, etc, without actually giving them access to the actual system. Admins should always handle the big stuff, such as firewalls, and whatnot. But managers want to state who has access to the VPN... Enter GOsa. Store that stuff in LDAP, and allow the end user to feel like they are driving, while the admin actually deals with the much more difficult job of configuring the software. I will repeat again... Look at GOsa, it is exactly what you guys have been asking for. Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 From: ubuntu-devel-discuss-boun...@lists.ubuntu.com [ubuntu-devel-discuss-boun...@lists.ubuntu.com] On Behalf Of Ryan Dwyer [ryandwy...@gmail.com] Sent: Thursday, October 22, 2009 6:02 PM To: Onno Benschop Cc: ubuntu-devel-discuss@lists.ubuntu.com Subject: Re: Ubuntu Domain Server -- 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
Modem Manager Dev Package
The README file for modem manager makes reference to a command line python program that demonstrates the D-BUS API to manipulate the modems. It also makes mention of a couple of generic classes that can be used to create custom modem classes for your 3G device... AWESOME, exactly what I want... Problem, these files do not exist in the main package. So I looked for a dev package, and could not find one of those either. Does anyone know where the dev resources are so that we can point some programmers at it? Thanks Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 -- 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
WiMax Japan
Anybody have WiMax working with UQ Communications in Japan? I would love to talk with you offline regarding your experiences and how to get it to work. Thx Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 -- 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
Confirmation: US15W & UNR Issue
Hi guys, looking to see if someone can confirm this behaviour, or is it just this prototype board that I am working on (not sure if this is a bug in Ubuntu or hardware issue) I have a COM module and carrier board that sports a Z530 CPU and the US15W chipset. When I try to install Ubuntu UNR, it fails to properly read the crazy IDE bus present on the US15W chipset, and therefore fails to see my SSD. However, when I install Jaunty, it sees it just fine, and installs without a hitch -- after which I am able to install the UNR task, and get to where I should have gotten in the original install Again, just looking for confirmation of this inconsistency, thx Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A division of Japan Communications Inc (303) 708-9228 x326 -- 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
The iPhone, VBox, and devio
I have been monitoring a problem with the iPhone breaking their Windows Virtual Machines on Linux hosts. Basically the problem progressed like this: If you have a Linux host, and tried to run iTunes on it, it would not run correctly via Wine. Most of us tried to create a virtual machine using one of the VM products on the market (this one was Sun's VirtualBox, but VMWare has volumes of details on it also). This allowed iTunes to work better, but as soon as you plugged in your iPhone, the VM would crash. Updates to both VMWare and VBox has stopped the crashing. However, iTunes still does not acknowledge the phone, so you can not backup your contacts, music, etc. So, I like many, have resorted to a dual boot situation to back up the phone. The phone is such an unreliable POS, that not syncing is not really an option. So, we wait, and wait, and wait some more. Finally today, one of the guys over at VBox/Sun came up with this solution: > In case one of you is in for some adventure. The iPhone works if you modify > the linux sources and recompile your kernel: > > Modify MAX_USBFS_BUFFER_SIZE to 128K in drivers/usb/core/devio.c I figure I would ask the developer community before putting anything in Launchpad so I can ask for something that does not make me look as stupid as I am in this area (kernel and drivers programming). * What would be the downside of requesting this change permanently in the Ubuntu kernel modules? * Could it be done safely? * Would any other devices/apps benefit from such a change other than the iPhone? Thanks Kevin Fries -- 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
The iPhone, VBox, and devio
I have been monitoring a problem with the iPhone breaking their Windows Virtual Machines on Linux hosts. Basically the problem progressed like this: If you have a Linux host, and tried to run iTunes on it, it would not run correctly via Wine. Most of us tried to create a virtual machine using one of the VM products on the market (this one was Sun's VirtualBox, but VMWare has volumes of details on it also). This allowed iTunes to work better, but as soon as you plugged in your iPhone, the VM would crash. Updates to both VMWare and VBox has stopped the crashing. However, iTunes still does not acknowledge the phone, so you can not backup your contacts, music, etc. So, I like many, have resorted to a dual boot situation to back up the phone. The phone is such an unreliable POS, that not syncing is not really an option. So, we wait, and wait, and wait some more. Finally today, one of the guys over at VBox/Sun came up with this solution: > In case one of you is in for some adventure. The iPhone works if you modify > the linux sources and recompile your kernel: > > Modify MAX_USBFS_BUFFER_SIZE to 128K in drivers/usb/core/devio.c I figure I would ask the developer community before putting anything in Launchpad so I can ask for something that does not make me look as stupid as I am in this area (kernel and drivers programming). * What would be the downside of requesting this change permanently in the Ubuntu kernel modules? * Could it be done safely? * Would any other devices/apps benefit from such a change other than the iPhone? Thanks Kevin Fries -- 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
IDEA: Commercial Subscription Repositiory
cess, commercial software imposes a new problem... not all upgrades are free and automatic. Therefore, the update process will need to be adjusted slightly. The new process would work like this: - Check to see if there is an update, if not continue as normal - Check to see if the package has a commercial license, if not continue as normal - If there is a license, and there is an update, check to see if the license covers the new updated version, if so, update as normal - If there is a license, there is an update, and an updated license is required, do not upgrade * Once the normal upgrade has been performed, a license checker (similar to the existing update notifier) checks to see if any software has a version upgrade requiring a new license. If so, it notifies the user. To upgrade commercial software, simply update the license and let APT upgrade and keep the software up to date (since the third criteria above is now met). The changes to the existing APT is trivial once the license manager is in place. And the license manager is not all that complicated... its just a database manager with a few key fields, and a blob to hold the binary license the software will require. Not only would this make porting commercial software from Windows to Linux much easier, it should make developing new commercial software for Linux much easier. When boxed software can be sold in Best Buy and Circuit City that is built for, or includes support for Linux, Linux will gain so much in legitimacy in the eyes of the general public. This could be a great first step to achieving this. I only wish I had the time to do it myself. Kevin Fries -- 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
OpenChange
We are in the process of changing our Exchange server here to 2007. This breaks the Evolution in Hardy. There appears to be a fix and it appears to be a plug-in as part of the OpenChange project. It looks like it is supposed to make it into Evolution 2.24, and Fedora is already reporting that it will be part of version 10. Does anyone know if this is going to make it into Intrepid? Thx -- Kevin Fries Linux Engineer, Emerging Technologies Computer and Communications Technology Inc. (303) 708-9928 x326 A JCI Group Company -- 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: BROTHER drivers in Ubuntu
OOPS! Accidentally sent to OP instead of group. Trying again On Wed, 2008-02-06 at 08:48 +0100, Piotr Zaryk wrote: > Hello, > > Would it be difficult to include drivers for Brother printers and > scanners? They are quite common and users have problems with them. If > not by default in Ubuntu install, why not to include them in the > universe repository? > > > user requests on launchpad: > > https://bugs.launchpad.net/ubuntu/+source/cupsys/+bug/25966 > https://bugs.launchpad.net/ubuntu/+source/sane-backends/+bug/16704 > > howto on ubuntuforums: > > http://ubuntuforums.org/showthread.php?t=590793&highlight=brother > +scanner+howto WOW, that howto sucks! I use two Brother Laser printers: HL-2070N (black and white laser); and HL-4040CN (color laser). Both work PERFECT in ubuntu. Combined they took me under 5 minutes to set up. But, I got my drivers direct from Brother. They have fantastic Linux support. Just go to: http://solutions.brother.com/linux/sol/printer/linux/cups_drivers.html#de If anything, just work with Brother to move their drivers into an APT repository so that upgrades can be automatically installed. That would be a far better solution than Ubuntu taking this problem on. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: gThumb
On Tue, 2008-01-15 at 19:34 +0100, Milan wrote: > Kevin Fries wrote: > > This argument has raged before. Has anyone thought of maybe a > > compromise... Something much better than GThumb (not hard) but not based > > on Mono. Maybe something like Blue Marine? > > > > http://bluemarine.tidalwave.it/ > > > > Just a thought. They already package a DEB file and it is about 30M, > > not sure how big the other packages are. > > > > > Is Java really better than Mono? Not speaking of needs to stabilize this > development version... I personally don't have a problem with Mono or Java. Both are excellent technologies in my book. Each has advantages and disadvantages like any technology. But when I last defended Mono and F-Spot, I got jumped like a gang member in the wrong neighborhood. Just trying to show options here. Looking for solutions to defuse the argument that erupted last time this was brought up, before the argument erupts again. Hoping to show that while gThumb is a poor product on its best days (editorial, I know, but face it, its not good), but F-Spot and Mono are not our only solutions. There are other programs out there that are of some serious quality. The BM page says that the product is not quite of release quality, but lets be real, its not of a professional release quality. For the average consumer, its perfectly stable enough. I would not want to rely on this if I was a professional photographer until a few bugs were worked out. But this product has no more bugs than the average FOSS product. BTW, I have just briefly played with Blue Marine, and that is one hell of a program. The type that really takes a run away from the second rate Windows based graphical tools, and right at the far superior Mac based one. It would definitely be a capabilities upgrade from anything in the repos now. If nothing else, putting that in Multiverse would not be the dumbest idea I heard. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: gThumb
resending to group, cuz I accidentally replied to poster On Tue, 2008-01-15 at 15:41 +, Scott James Remnant wrote: > We feel the opposite, that f-spot has a better user experience. This argument has raged before. Has anyone thought of maybe a compromise... Something much better than GThumb (not hard) but not based on Mono. Maybe something like Blue Marine? http://bluemarine.tidalwave.it/ Just a thought. They already package a DEB file and it is about 30M, not sure how big the other packages are. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Kerberos? Does anyone have this running?
On Fri, 2008-01-11 at 00:49 +0100, Morten Kjeldgaard wrote: > > Tried that and actually configuring DNS. DNS is now fully up and > > running, yet... > > > > Others in the forums report the same failure despite dnsdomainname > > returning correctly. Yet, there were no responses. That is why I > > decided to ask the developers. Thinking it may be in the process of > > being EOL'd or something. > > You also need to have reverse DNS up and running. Yea, struggled with that due to false error messages, but R-DNS is running perfectly, and still refuses to configure correctly. While I have never been able to prove it, is have this sneaky suspicion that my Kerberos problem is due to multiple interfaces. My Setup: +---+ | I | +--+ | n | +-+ Server 1 | | t | (0)+---+ (1)| +--+ +-+ | e ++ LabDC ++--+ Test Client | | r |+---+---+| +--+ +-+ | n || (2)+-+ Server 2 | | e |+---+---+ +--+ | t ++ LabAS | +---++---+ Essentially, LabDC is my problem machine. The Internet is simulated by our Corp Network. LabDC gets its address for eth0 via DHCP. It is also a portal into the simulated corporate network. So, eth1 is a static address into the servers and test client that stand for a typical corporate network. It serves DNS, DHCP, etc accross its eth1 interface. The assumption is that there would be some sort of SSO environment (Kerberos, AD, etc) on that network. So, I was trying to Install Kerberos because I am trying to avoid setting up Windows servers due to resource constraints. The eth2 interface is to a special server which is represented here by a machine referred to as LabAS. This is the heart of our proprietary solution. However, I can not test that, until I get a good Backoffice simulation going. Because LabDC is sitting on the intersection of three networks, and the half-baked way the KBC configuration is designed, I suspect, but can not prove, that this is at the heart of the problems I am seeing. It would be nice if I had more resources to set up additional machines, but, budgets being what they are... Thanks for the help, I am starting to consider giving up on Kerberos, and just settling for a more simplified LDAP based solution. It means that my simulation will not be as representative of a Windows based network, but working and simplicity sometimes just needs to rule the day, lol. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Apt-Cacher again, sorta
On Fri, 2008-01-11 at 01:23 -0500, Fabian Rodriguez wrote: > Kevin, can you provide any URLs to those forum posts ? I can look into > it over the next few days. I originally was interested in writin a MIR > about apt-cacher, just haven' t gotten around writing it yet. At the > time I searched for a good time for reports about any issues with > apt-cacher and couldn' t find any, not bug reports. Remember, Apt-Cacher was just the catalus of a bigger problem. But here are three threads of people complaining about what looks like a crash in the install: http://ubuntuforums.org/showthread.php?t=219861 http://ubuntuforums.org/showthread.php?t=628945 http://ubuntuforums.org/showthread.php?t=580590 -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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
Apt-Cacher again, sorta
At the risk of picking at old wounds... I noticed a behavior this week that probably should be addressed. I fell into it by relying on Apt-Cacher, but a quick search on the forums shows lots of others having issues based upon other types of connectivity problems... I have been building VMs like mad. I have a simulation build that I am in the middle of that include a minimum of seven machines on four virtual networks. I tried this on my laptop... not one of my smarter ideas. I now have a nice server I conned somebody out of. Here is how I fell into the trap: I built a basic Ubuntu Server build. I then added the VMware server from the VMware site (the one in the repositories for some silly reason has no management piece). I then build my first virtual machine using Ubuntu Jeos (yes Neal, its that project, remind me to show you when I see you next). This first server was designed to represent a Linux firewall locked down to within an inch of its life. No pass-through. I installed DNS and DHCP to serve the other machines to come (its also the machine I am fighting with kerberos on). In the wild, this network will be extremely restrictive, and standard desktops may or may not have access to the Internet. To keep the simulation simple, I am assuming no machine has access to the Internet, knowing that I can always relax the restriction in production. Now I built the second machine on the network. The first machine had eth0 set up using bridging, and eth1 and eth2 connected to private networks. The second machine connected to one of these private networks. Updates will be done via Apt-Cacher installed on the first machine. However, when installing on these machines inside the protected network, installation came to a halt, often waiting as much as an hour before tossing an error on the screen and moving on. You see, it got upset over Apt wanting to get to the Internet to configure the Mirrors. Once it finally gave up, it left the sources.list file in a mess, then finished the install. Now I know some of you are going to think that I am going to bring up about supporting Apt-Cacher on install again aren't you? Well consider this my only pitch on that, because there is a bigger more important point here. When I looked up the issue (as I thought the install had crashed after 30 minutes of inactivity), I saw a bunch of people with the same issue. They were complaining about it, and telling newbs to disconnect their network card from the network to make it work. OUCH! Is this what we want to tell newbs right after we convince them that Linux is as easy to use as Windows? While they were not seeing it because of apt-cacher and a network locked down by an insane admin (yes, me), they were seeing it due to either overloaded mirrors at the time, or other networking issues. I know there was conversations about including apt-cacher or some other apt mirroring software into hardy here just a few months ago. But if we are not going to do that, we should at least make the process less painful in the case of simple communications errors. Some of these people out on the forums were just jacked up because the wireless driver was not included on the CD by default. Maybe we need to delay this mirror finding piece until after install? Maybe we need to have this mirror finding piece understand a downed network? Maybe we need to have this mirror finding piece time out quicker? Something, I am a geek to the core. Been using Linux since the RedHat 2 days. So, once I realized what was going on, I went to lunch. Once I got back, Ubuntu had dumped garbage and went on. I could then easily clean up the mess. But Ubuntu is trying to reach out to newbs. And they will not understand. It also leaves no repos installed which can cause a second round of confusion to the less initiated. Just my $0.02 -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Kerberos? Does anyone have this running?
On Thu, 2008-01-10 at 12:30 -0700, Neal McBurnett wrote: > My guess is that the server team mailing list, or #ubuntu-server > or #kerberos would be better places for this discussion. But read on > - my comments are interspersed. > > On Thu, Jan 10, 2008 at 06:17:56PM +0100, Magnus Runesson wrote: > > Do you have the full links to the doc you refer to? > I second the question :-) Makes it much easier for the many readers > of this thread. The original doc I was trying to follow was: https://help.ubuntu.com/community/SingleSignOn it fails at step 3.1.3 > > > Is there something going on with Kerberos as to why there is no help, or > > > bug fixes? Is this package going the way of the Woolly Mammoth? Or is > > > it just not getting enough TLC? Or, is something else going on? > > I don't know, but my guess is that most of the efforts are going into > improving interoperability with Active Directory rather than just > kerberos. The major news on that front is "Likewise": > > likewise: http://www.linux-watch.com/news/NS2350659361.html Unfortunately we are a small shop here. I am already pushing my lab to the limits without adding AD to the mix, lol. > I had a problem like that once, and think that it had to do with not > having a Fully Qualified Domain Name (FQDN) for my system in > /etc/hosts. I seem to recall that the kerberos installation uses the > server's FQDN to make a default realm, and doesn't catch the error if > that doesn't exist. > > Googling for your error string with the word "launchpad" added (which > helps google prioritize authoritative launchpad bugs over chat in the > forums) led me quickly to this: > > https://bugs.edge.launchpad.net/ubuntu/+source/krb5/+bug/159357 Yep, saw that, and followed that thread. But I like many others have dnsdomainname working correctly, and it is still not working. > which gives more details on the root problem: dnsdomainname is not > finding the FQDN. For me, a gutsy install in which I gave it my FQDN > at install time properly put my FQDN in the /etc/hosts file, but if > I just gave it a local name I ran into this kerberos problem. > > This is the format of what worked for me in /etc/hosts: > > 127.0.1.1 example.com example Tried that and actually configuring DNS. DNS is now fully up and running, yet... Others in the forums report the same failure despite dnsdomainname returning correctly. Yet, there were no responses. That is why I decided to ask the developers. Thinking it may be in the process of being EOL'd or something. I may just blow off KRB for this project, since it is not critical, desirable, but not critical. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: LVM on hardy's live installer?
On Wed, 2008-01-09 at 12:36 -0500, Evan Dandrea wrote: > On Wed, 2008-01-09 at 10:26 -0700, Kevin Fries wrote: > > On Wed, 2008-01-09 at 11:50 -0500, Evan Dandrea wrote: > > > To my knowledge there has never been a LVM branch of ubiquity. > > > > I am 85% sure Jeos does, as does the alternate install disk. For some > > reason, I thought the server disk also did. > > Right, but those use debian-installer, not the live CD installer, > ubiquity. d-i has had LVM support for a while, as already mentioned. ahh, I get your point now. There is a bad issue IMHO with the way Ubuntu partitions disks by default anyway. While I always custom install to place /boot and /var or /var/log on separate partitions, I understand other reluctance to do that. But! /home should always be on a separate partition in a workstation build, and placing that as LVM by default is just a great CYA move. If there is any part of the system that is likely to grow out of control, its user file space i.e. /home. This just comes down to long term ease of use for the end user. Adding another disk, could be added to the /home LVM, and provide an environment that is more advantageous to end users than Windows (who would need to keep track of additional disk drives rather than just see a larger home partition). Given that Hardy is a LTS release, I guess the question is to add it now so that the LTS users don't have to wait 3 years to get it? Or consider the feature untested, and wait for Hardy+1 so that there is time to iron out bugs before its added to the next LTS release? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: LVM on hardy's live installer?
On Wed, 2008-01-09 at 11:50 -0500, Evan Dandrea wrote: > To my knowledge there has never been a LVM branch of ubiquity. I am 85% sure Jeos does, as does the alternate install disk. For some reason, I thought the server disk also did. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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
Kerberos? Does anyone have this running?
I am trying to build a single signon environment as an Ubuntu demo. I tried to follow the help.ubuntu.com website to insure I did it the "proper" or "documented" way. There are numerous threads of people asking for help in the forums. In addition, there are 15 bugs filed in Launchpad. Many of these are going back to 7.04. Is there something going on with Kerberos as to why there is no help, or bug fixes? Is this package going the way of the Woolly Mammoth? Or is it just not getting enough TLC? Or, is something else going on? For the record, I tripped upon the problem documented in the forums, and launchpad bug reports, where upon install it does not properly run its full configuration. It never asks me for realm or anything. Once blowing off its config, it then fails to start (yeah I know, go figure). But manually setting up the configuration does not work either. The package keeps coming up with the error: "kadmind: Improper format of Kerberos configuration file while initializing context, aborting". As suggested, my dnsdomainname returns correctly, and I have manually configured the files in accordance with the documentation. No errors are thrown in the syslog. Thx -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Easy "Add/Remove Porgrams" for non-sudoers with local PREFIX?
On Thu, 2008-01-03 at 18:20 +0100, Carsten Agger wrote: > Is there any reason for this - I mean, is the use case of a non-root > user wishing to install software and still using "standard" utilities > like "Add/Remove Programs" considered insignificant? Has there already > been discussions abt this and in that case, do you remember he arguments > for and against? > > (Having utilities like 0install is great, but I was also thinking about > the overall userfriendliness of Ubuntu, and making it easy for non-root > users to install e.g. games from the universe repositories without > bothering the admin and without having to know how to recompile from > scratch seems like an essential use case for me. What you are bucking against is Linux Philosophy more than a technical problem. Lets see if I can explain this without starting a flame war. One of the concepts of Linux (and modern Macs) is the concept of a separation of preferences versus system administration. Microsoft does not have that separation, and this is the cause of much of their security "issues". Linux avoids much of that by defining two areas of configuration, system and user. User configurations are items that only affect the logged in user experience. These items are things such as Colors, icons, window decorations, email preferences, etc. These are set on a user by user basis, and are stored in ~/.program directories. Configuring these settings is done via the System->Preferences menu. System configurations are items that effect users system wide. Available network configurations, Printers, and software available generally fall into this category. The configurations for these items generally exist in /etc folder, and are graphically configured via the System->Administration menu. Some items fall into a great abyss of uncertainty. For example, WEP information. If I put up a wireless network at my office, and have a laptop that several people use, should the key be configured globally? However, being a laptop, authentication of multiple networks is probable, and can not be administered globally. Its a system wide service (wireless networking), but needs to be configured on a user by user basis. These gray areas have never been Linux's strong suit. When it comes to software available though, most of this comes from a historical perspective of centrally administered machines. In the old days (20 computer years = 1 human year, so the old days are not all that long ago, lol) most Unix based machines were shared among several users. All software was centrally installed and configured. If a user wanted, they could always install a custom program into their own ~/ folder. Some Unixes actually placed ~/bin in their execution path. Its this history that you are bucking right now. Having this separation has actually helped the Nixes to avoid the virus problems of the Windows environment. Because end users can only install software to their own directory, they can not infect the core of the OS with a virus. Because of the elevated privileges required, software can not install automatically like it does in say ActiveX. While the practice of having admins do all installs was to keep some sanity in a machine with multiple users, it is this security reason more than any other as to why installs remain the realm of the admin. Modern Nix based systems have a wonderful tool called SUDO that makes getting around this issue extremely easy. If you want someone to be able to admin your box, add them to the admin group on any Ubuntu based system. Then they have sudo access to any root command. If you want to allow non-root users to be able to install software, that is easy also: - Create a group call swinstall - In your /etc/sudoers file add the following line: %swinstall ALL = /usr/bin/update-manager - Add any user you wish to have software install access to the swinstall group. Hope this helps -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Q: Reason for partitioning scheme?
On Thu, 2007-12-13 at 15:58 -0500, Evan wrote: > I definitely agree that the single / partition isn't the best way to > go. My preference is > > swap > /boot > /home > / > > While other partitions can be useful, this covers the most important > areas (user data, settings, and the ability to continue to boot > Windows/OSX regardless of what happens to Ubuntu). In the desktop environment, I agree about /srv. When I do this manually, it is either /srv or /home that ends up in its own partition. I was trying to address both server and desktop installations with one rule. As for the /var/log... I put that one in because I have had runaway processes send info, warning, and error messages to the syslog. That is where they will end up. If the logs fill a drive, I want a simple way to recover. As an advanced user, I would mount a live cd or simply boot to the boot sector. Manually mount that drive. Then fix the problem. reboot. As a end user, they will come to me, and I will do my best Nick Burn Computer Guy... huuuh, move! ... lol Now back to my original question... Is there a reason we don't use one of these more stable layouts now? And should we consider it for Hardy? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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
Q: Reason for partitioning scheme?
rv -> 10GB (secondary use, remainder 18GB) As a non-Linux user interacts with the system, all the area that can cause problems are isolated from areas that can make the system to be unbootable. Especially with noobs, this layout seems to make more sense to me. It has self preservation built in. While I realize some may think my layout is overkill, and that is not the point. I give this as an example of a layout that is easily calculated, protects the end users data, and important settings, and provides hooks to allow real propeller heads like myself to come in and rescue in case of real disaster. If you use partition labels, Ubuntu should even be able to identify a /home and /srv partition from an early install, and attempt to protect data across a complete reinstall! I guess after this long explanation is... is there a reason that Ubuntu, given its target on noobs, did not adopt a more protective partitioning scheme like this? And more importantly, should we? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Fwd: Mono (Re: New Programs for Hardy?)
On Wed, 2007-12-12 at 18:51 -0600, Conrad Knauer wrote: > On Dec 12, 2007 2:36 PM, Kevin Fries <[EMAIL PROTECTED]> wrote: > So going the other way from removing Mono, are there any mono-based > libre software apps in the repos you'd like to see moved onto the > default desktop? Not specifically. What I was really referencing was more of a "lets start building more apps with Mono that are Windows ports" type of argument. But as some have correctly pointed out, that may be a bit tricky given that Linux has supposedly violated every patent that Microsoft has ever owned, and several they have not even written yet, lol. I feel that Mono is that safe haven you give Windows developers. That carrot to encourage them to take that .NET app, and while not breaking .NET capabilities, also allows them to expand their audience. For that reason alone, I would like to see the Linux community put more apps on Mono. The more Windows apps that can easily convert to Linux compatibility, the better for Linux users, the Linux community in general and Ubuntu in specific. Much better idea in the long run than adding a bunch of geek apps like gParted and VIM. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: New Programs for Hardy?
On Thu, 2007-12-13 at 03:51 -0500, Michael R. Head wrote: > On Thu, 2007-12-13 at 02:36 -0600, Conrad Knauer wrote: > > So let's use that as a dividing line; let's keep the 32-bit x86 disks > > as CDs, BUT... let's switch the 64-bit disks to DVDs. > > It's not uncommon for some servers to come with just a CD ROM drive (I > manage a Dell PowerEdge running 64-bit dapper in such a configuration). > > For example, in its base configuration, the Dell SC440 is still being > sold with a 48X CD-ROM Drive. These are just the machines on which > users may wish to install Ubuntu's amd64 architecture. I have been ordering servers more and more without CD Rom drives. I deal with a lot of managed systems, and have opted for moving the install over to a flash stick. One idea that Debian has had for years, that I am surprised that Ubuntu did not follow -- especially with servers -- was the idea of the minimal install CD ( < 50MB to fit on a mini-cd or flash stick) that was little more than a debootstrap install. Then everything was obtainable from the repositories. While I realize that could get ugly for the noobs that Ubuntu goes after if followed exactly... but what about a derivation off of it. Instead of leaving just a command line system, it installed a core system, then rebooted, upon first boot, it asked which U/K/X-buntu version you wanted, then retrieved that from the Apt repositories. Now CD capacity is unimportant. Of course, this only helps in places where broadband internet connectivity is available. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Fwd: Mono (Re: New Programs for Hardy?)
On Wed, 2007-12-12 at 15:46 -0500, Martin Owens wrote: > > I will even help you with one more I would like to see... Scribus. My > > mother uses this along with Inkscape for her scrap-booking (definitely > > not a geeky endeavor), and with a few tweaks to the descriptions, could > > be a very popular addition. > > I'll have insist about the sync support, it's not a geeky endeavour; > and most people avoid it because of the difficulty. It's a feature we > could make better than other platforms, we have all the tools written > already. Sync is definitely gray area (goes to my earlier quality over quantity argument). The problem is that I have yet to see a sync client that truly does it all. They all claim to, but reality is a bit more sketchy. Right now, I will settle for one that does most, reasonably... still waiting > > But there are better places to trim than mono. I personally would like > > to see more mono apps included by default to encourage Wintel developers > > to extend their product to the Linux desktop. That would be a win for > > everybody but Microsoft, but that does not disappoint me so much. > > But as you pointed out, we don't want to include things because it's > good for developers, it's an operating system for human beings, not > wintel geeks. I'm not sure where this fascination has come from that > we need to include mono by default to encourage windows developers. > All the developers I know from the windows world move into Linux by > programming in python, c++ and java. Not through the .net framework. > In fact shouldn't we be installing Eclipse if we're so focused on > developers? You are talking tools, and those that are taking the effort to learn Linux. One of the areas where I think we can all do better is to encourage Wintel geeks to stop being hostile to Linux. Encourage the use of tools such as Wine and mono. Use the carrot and back off on the whip by showing that they can tweak existing programs to gain mono and Wine compatibility, and instantly grow their market without having to actually write for Linux. This then causes more of a crutch on mono, and extends the conversation to Wine. Which come to think about it, would be great to add to the base install. Its about making the transition easier for the new user. Two other programs that by the way also use the mono runtime are Beagle (ok stop laughing, I turn it off too), and gnome-rdp used to access Windows desktops. Less so in homes, but that is extremely useful in businesses, or more importantly... workers at home. Again, services that are built into Windows, so they should be represented in Linux so noobs can see that it can be done easily. The easier they see the transition, the more likely they will be to start the journey. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Fwd: Mono (Re: New Programs for Hardy?)
On Wed, 2007-12-12 at 15:27 -0500, Martin Owens wrote: > > Could you list some example of those things that would be nice there? > > Sure thing: Inkscape, Conduit, opensync, Glipper, firefox-adblock, > rar-free, more languages, obex, gnome parition editor, audacity, > gnomebaker, vim (real vim), PGP keys manager, open office draw, any > kind of irc program, keep backup 2, graphics tablet integrations, any > kind of webcam management software, cheese, devede, atlantik-gtk, > compiz manager, start up manager, Storage Manager, schedual, Dohickey. I'll give you Inkscape and webcam, but I think the rest such as addons, vim, irc, gnome partition editor, etc really do need to stay in the repos. I will even help you with one more I would like to see... Scribus. My mother uses this along with Inkscape for her scrap-booking (definitely not a geeky endeavor), and with a few tweaks to the descriptions, could be a very popular addition. But there are better places to trim than mono. I personally would like to see more mono apps included by default to encourage Wintel developers to extend their product to the Linux desktop. That would be a win for everybody but Microsoft, but that does not disappoint me so much. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Fwd: Mono (Re: New Programs for Hardy?)
On Wed, 2007-12-12 at 12:58 -0500, Martin Owens wrote: > > F-Spot and gThumb are very similar in capabilities on local images. > > Though F-Spot's interface is a little cleaner. But the big difference > > comes in Web2.0 integration. GThumb has none, while F-Spot integrates > > with Flickr, Picasa, etc. Hands down, this is what end users expect, > > and it is gThumbs that needs to be eliminated. > > Users who want any of these applications can get them from the > repositories; just because you like the applications doesn't make them > good for inclusion by default. The argument seems to be that we should > sacrifice 60MB of CD space for a handful of extra features. The space > instead could be used to have many more user friendly features than > just nice sticky notes and Flicker integration. They get these features in Windows, if we want to reasonably expect to get users to switch over, we need to supply them also. This goes to user expectations. Windows and Mac are the front runners, We need our best foot forward to compete with iPhoto. A better program in the repositories means nothing when you pop that CD in to show the end user they can have the same capabilities with Ubuntu than they can in Windows or Mac. As for me, given the programs that are there, and the ones that will make hesitant users switch, the 60MB must be sacrificed. Putting lesser programs on the disk to add in what? Multimedia is so mainstream that hardware manufacturers are making it available from an otherwise turned off machine. The only think that brings more bling would be codecs, and that is not going to happen for licensing reasons. So, I could not agree with you assertions more. Programs that integrate better with more features are far more important that number of features. I would rather see a second disk than to back off on these features. And I feel a second disk is a huge mistake. > Although I didn't see you mentioning that F-Spots flicker integration > will be removed and moved into Conduit (which we don't include by > default) so you argument does loose some merit with the way things are > going. And if we should be or not was not the question. I have no problem with the conduit. But to install a lesser program to get more software is a DSL type decision. That is what they do best, not us. We need to keep everything about providing a highly usable desktop with the features people expect. After iPhoto, they expect more than gThumb. They expect what F-Spot brings. Like I said, the users have already spoken on this exhaustively in both the forums and Launchpad. My opinion seems to be the overwhelmingly popular one. Quality over quantity... That is what makes Ubuntu the best distro. I would also be curious as to which programs you would choose to include to replace that 60MB? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Fwd: Mono (Re: New Programs for Hardy?)
On Wed, 2007-12-12 at 17:22 +0100, Sebastian Heinlein wrote: > Am Mittwoch, den 12.12.2007, 05:06 -0600 schrieb Conrad Knauer: > > > Mono by default takes 48MB of space on the CD. The ISO download is > > 690+ MB. Therefore, it is taking up valuable space that could be used > > for a whole host of other things. Also, for that 48MB, there are just > > two applications which use Mono. These are F-spot (photo manager) and > > Tomboy (note application). Ubuntu also includes two other programs > > which do a similar job, gThumb (photo manager) and GNOME sticky notes. > > > > In my opinion, these two applications function well enough to warrant > > the removal of Mono dependent programs. This argument has been made and lost in Launchpad, and the consensus seems to eliminate gThumb. I for one agree with that assessment. F-Spot and gThumb are very similar in capabilities on local images. Though F-Spot's interface is a little cleaner. But the big difference comes in Web2.0 integration. GThumb has none, while F-Spot integrates with Flickr, Picasa, etc. Hands down, this is what end users expect, and it is gThumbs that needs to be eliminated. As for Tomboy vs GNOME sticky notes, this one is even more obvious. Sticky notes needs to go away. GNOME no longer considers it part of the base suite of packages, and has instead worked with Tomboy on tighter and tighter integration. Tomboy can fire links to open on your browser or Nautulus, fire alarms as reminders, and integrates with Evolution. Sticky notes does none of that. I realize the original argument was about the size of Mono. And that is a legitimate argument. But lets also realize functionality and integration needs to be maximized in order to make this distro easy for the noobs it is aimed at. We already have a distro out there that makes sacrifices of number of packages over space... its called DSL. So, when I hear solid arguments on both sides like this one, I generally follow the Dad test (my dad will screw things up faster than my mom because his is more willing to mess with things to "make" them work, mom will just ask). Mom takes pictures with her new digital camera she is getting for Christmas... then asks my dad's help in putting them online so she can share them with my brother in LA, me in Denver, and her sister in Pittsburgh. If they were on a Ubuntu system with gThumb or F-Spot, the pictures can be imported, tagged, and sorted quite well. But with F-Spot, my dad can export to a web page, or use file->export to post those to a Flickr account. Huge win to F-Spot. Taking the same thought process with Tomboy vs Sticky Notes... If I want to take an quick note, both will do the job just fine. But I can also take an email and drop it into a sticky with Tomboy. I can also make the notes appear as to-do items in evolution. If a file on my hard drive is referenced in the note (i.e. Finish sales proposal ~/2008 Sales.odt), I can click on the file name to open it. None of that can be done with Sticky Notes. These two products have been replaced with a Mono based equivalent and both have become standards despite the size hit. There are solid reasons for that. While space is always a concern, the level of integration expected by the modern users can not be sacrificed. And therefore, I would disagree with the OPs suggestion that these products be replaced. If you want to save space, I say get rid of gThumb and Sticky Notes like many other distros have. The mono based replacements have too many modern features to follow the OPs suggestion. Just my $0.02 -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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
[Fwd: Re: Appropriateness of posts to this list (Was Re: evince crash)]
This was sent to me personally, and it has comments directed to others in the group... Therefore, I assume it was meant for the group at large. Kevin Fries Forwarded Message From: Richard A. Johnson <[EMAIL PROTECTED]> Reply-To: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: Re: Appropriateness of posts to this list (Was Re: evince crash) Date: Thu, 6 Dec 2007 13:54:45 -0600 On Thursday 06 December 2007, Kevin Fries wrote: [...] | If I don't get my steak the way I ordered it. I buy my steak from | elsewhere. Ubuntu with no users, is not anything but an exercise in | ego. What the customer wants is the only real metric. You need to | understand that as a developer, and I live with that every day as a | Consultant, Designer, and Implementer. | | Which of those priorities you wish to work on, however, is completely | your own decision. But the customer MUST set the priorities of what | needs done in the bigger picture. And, the customer MUST set the list of | features that need to be implemented. I couldn't agree with you more! | Rule #1 of Business: Its not about you. Actually, this wouldn't be Rule #1, but it is pretty much the Golden Rule of Business. Mark Cuban said it best a few years back, "Treat your customers like they own you, because they do." The hard part with this though in our little neck of the woods is that all of us are also customers, so it can get confusing. | If you do not make your customers wishes and desires #1 on your priority | list, your competition will. And they are (ie. PCLinuxOS, Fedora). | Lets not forget, Ubuntu is a business product, distributed by a real | business. Therefore, its not about you... or me. Its about the | customer. Making the customer feel like they have to talk you into | something, is just not good business. This is why I spend so many hours | providing help to ANYONE who asks. Even people I would rather not. Its | not about me, its about Ubuntu, and what is best for the project. It was all fine and dandy until this paragraph. This is the one thing that really could irk a volunteer to such a project. I have been around this community for a couple of years now and talking to some past developers and contributors, the one thing that was common was that "we are working for free while they are making money from our work." I look at it like this..Kubuntu is giving me more than I could ever give it. How? 1) I have a totally free operating system 2) I don't have to worry about all the other things I would have to with that other OS 3) The development community allows me to participate in which I get to learn the ins-and-outs of what really goes on (after a while, this is a nice CV bullet point) 4) The friends I have made in the process are totally worth every minute I have put in. | Even more so in an all volunteer endeavor, egos must be checked at the | door. Developer's egos, designer's egos, and consultant's egos. We as | the people trying to make this a success, need to listen to the customer | so that there will be more of them. Its the one true advantage we have | over Microsoft which is notorious for blowing off their customer to do | what is in their best interest (Can we say Windows Genuine Advantage, or | Digital Rights Management... I knew we could). I am 50/50 on this paragraph. I wholeheartedly believe there should be the "checking the ego at the door," however a little bit of ego never hurt anybody. For instance, look at Microsoft. They have the biggest ego of all, and they have yet to really fail at what they do. Going on with Microsoft, they do indeed listen to their customers, just because we don't see it simply because we are not their customers, doesn't mean they don't. If they didn't listen, would they really be as big as they are? I mean Apple and other operating systems have been around just as long. Imagine if the Linux community would have really listened to the complaints in the 90s, I think we would then be further than we are today. In our eyes, yes we do have a true advantage over Microsoft, but to the billions of Microsoft users out there, they laugh at that advantage. | You allow the customers wishes to be the only real metric because you | place Ubuntu and Linux's needs before your own. Otherwise, are you | really helping? Very true, but one thing I have noticed from doing so is this: 1) Linux isn't gaining the ground with proprietary vendors. Why? because most distros have listened totally to the customer and have provided them with the proprietary solutions. This isn't helping in my opinion. And the one thing that really sucks with these proprietary solutions, we can't help/support the users when problems occur. The only thing we can do is say "oh well, that is what you get when using proprietary stuff, we can't help you, ask . T
Re: Appropriateness of posts to this list (Was Re: evince crash)
On Thu, 2007-12-06 at 12:03 -0500, Scott Kitterman wrote: > I think you misunderstand my point. No I got it. And I think that that thinking is wrong and dangerous to Linux in general, and Ubuntu in specific. > My concern is the idea that "because a user said they want it" is a meaninful > metric in a largely volunteer project. In Free software projects, the > meaningful metric for what gets done is what the people doing the work think > needs doing (and this applies to all types of work, not just development, in > the project). Volunteers can't be ordered. They have to be convinced. If I don't get my steak the way I ordered it. I buy my steak from elsewhere. Ubuntu with no users, is not anything but an exercise in ego. What the customer wants is the only real metric. You need to understand that as a developer, and I live with that every day as a Consultant, Designer, and Implementer. Which of those priorities you wish to work on, however, is completely your own decision. But the customer MUST set the priorities of what needs done in the bigger picture. And, the customer MUST set the list of features that need to be implemented. Rule #1 of Business: Its not about you. If you do not make your customers wishes and desires #1 on your priority list, your competition will. Lets not forget, Ubuntu is a business product, distributed by a real business. Therefore, its not about you... or me. Its about the customer. Making the customer feel like they have to talk you into something, is just not good business. This is why I spend so many hours providing help to ANYONE who asks. Even people I would rather not. Its not about me, its about Ubuntu, and what is best for the project. Even more so in an all volunteer endeavor, egos must be checked at the door. Developer's egos, designer's egos, and consultant's egos. We as the people trying to make this a success, need to listen to the customer so that there will be more of them. Its the one true advantage we have over Microsoft which is notorious for blowing off their customer to do what is in their best interest (Can we say Windows Genuine Advantage, or Digital Rights Management... I knew we could). You allow the customers wishes to be the only real metric because you place Ubuntu and Linux's needs before your own. Otherwise, are you really helping? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Appropriateness of posts to this list (Was Re: evince crash)
On Wed, 2007-12-05 at 22:13 -0500, Scott Kitterman wrote: > For those of us who are volunteers (most of us), the compromise is > someone has to convince me it's worth my time to bother. So I'd say > the other way around. The users who want volunteers to actually do > free work for them need to be convincing why I should be bothered > (hint: threatening to switch back to Windows doesn't motivate me at > all). If you need motivation from external sources, then maybe you are misdirecting your efforts. I am not trying to be mean here, but I use and advocate Linux for many reasons. Nobody has to motivate me to do so. I do so because I believe in the platform, and I want it to succeed. I want it to succeed for selfish as well as altruistic reasons. Many unpaid hours are spent helping someone get started, not because I need to be convinced to do so. I do so because I want Ubuntu to succeed, I want Linux to succeed. And I am not alone. Many of the local Colorado Local Group, are looking for ways to provide help desks to noobs, to get more CDs out, or even get cards out pointing the uninitiated to online resources. None of them are paid either. Nor do any of them need to be convinced to do so. They do so for the same reason I do... Because it is what is needed to be done. There seems to be this growing trend in the Ubuntu community lately, and I am pretty sure that it is an all bad thing. The developers, not all but a growing number, seem to think Ubuntu is their baby. The sweat of their brow, and therefore, only successful because of what they do. While I will be the first to say that these voices are still the great minority, they are getting louder. And diminish the fantastic work done by so many. There are many ways to contribute to a project such as Ubuntu. I have offered programming skills, and was treated quite rudely buy certain members of the programming community. They were extremely territorial and condescending towards my efforts. I have since decided to focus my efforts elsewhere. The comments above are exactly the attitude we need to guard against. In my current day job, I design state of the art hand-held computerized devices. Because of my advocacy here, fewer and fewer of those machines are being considered for Windows Mobile. Because of my efforts here, Windows Mobile is no longer mentioned when new projects come up. That takes my faith in the project, and the developers ability to deliver that project. Without the developers, my faith is misplaced. Without my advocacy, the developers efforts are purely academic. I am no more, or no less important than the developer, and I expect to be treated with the same respect. Not looked down upon so some developer can find motivation. In addition to my day job, I am getting a business off the ground. This business is designed to bring real, solid, Linux based networking solutions to the small and medium-small based business. A market segment that Linux has not had much success with in the past, and Microsoft is pandering to. Bringing Linux to a brand new market segment is not easy. However, I think by doing so, I can build a successful business, allow small businesses to better compete, and advance the Linux and Ubuntu cause. But I do not degrade the efforts of others to justify why I do this. Eau contraire mon frer, I praise them. For if it was not for all that they do, I could not concentrate on resolving the businesses issues. I know that not everybody on this board is American, but one thing that every American child learns in history class is in regards to the American Civil War. The north finally prevailed with a tactic of divide and conquer. We learn the slogan "United we stand, divided we fall" quoted by Thomas Jefferson (attributed as far back as Aesop). And Abraham Lincoln's paraphrase of that statement "A house divided against itself cannot stand" (one of the great speeches of all time). Ubuntu is strong when we all listen to each other, give each other respect, and stop trying to claim that my problems are more important than your problems. EVERYONE, end user, advocate, consultant, developer, or even the businessperson that uses Linux for his/her operating system of choice, is in this together. Or else, we can tear each other apart. It is time to come together, shut our mouths, and listen to the other sides. It is time to do this so that all of our efforts can be maximized, and focused on the common good. It time to stop using the word me (also meant in the metamorphic sense such as end user, or developer) and start using the word us. Thats just my $0.02 -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: apt-cacher in main + apt-zeroconf
There is another problem with apt-zeroconf... it relies on Avahi. Avahi has lots of environments that it does not work in. In my office, the machines are not seeing each other. When we had a meeting at the Google (Sketchup) offices in Boulder, Avahi did not work correctly their either. I don't think its a bug in the software, but instead it has to do with the way the routers are set up. Before turning on anything like that, you would need to insure that it would be reliable. That is why I proposed the "scan" technique. If the scan fails, either due to protocol issues or the fact that there is no server, it fails over to the old way. @Fabian: I agree with you to a point. Auto-detection can be problematic. But I suggest auto-detection like compiz auto-detects. If its not there, don't force it. Or even better yet... set up the always works version, but if I can reliably detect a better way, reconfigure to that better way by default. Decreasing bandwidth without having to remembering to configure every new client is one of those features that make good buzwords with IT managers. Canonical has made statements indicating that they want to go after the corporate desktop... this is exactly the type of feature that plays well with that demographic. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: apt-cacher in main
On Wed, 2007-11-14 at 18:27 -0500, Scott Abbey wrote: > I think the point of moving it is so that it receives official support from > Canonical. That way those on paid support contracts can still expect > assistance from Canonical when using the package. Canonical only provides > paid support for packages in main and restricted. Universe and multiverse > receive community support only. Ahhh (he says with bells going off like a Las Vegas slot machine) In that case, I would like to second that motion. Especially with Jeos coming out. Not all VM need to have a public facing. I often set up servers in a VM based environment where one or more machines are not given access to the outside network (only internal networking between the virtual machines). Without apt-cacher you either need to allow that machine access to the Internet, or do without updates. Example: Several VMs use a central set of accounts. You store those accounts in LDAP. If the LDAP server only has networking between the VMs, you do not have to worry about setting up TLS. But the second that machine has external visibility, TLS it mandatory. Any time you can relax the security, things run faster and more reliably. So, by putting that LDAP server in an inaccessible place, allows you to run without all those layers of security (good design trumps good security every time). Given this example, the machine with no external network support, can use whichever machine has apt-cacher as a proxy to get updates, without compromising its security. Given the announcement of Ubunut Jeos, moving apt-cacher to a place where it will get support. But I still think it would be cool if the install process scanned the local net looking for any machine with port 3142 open, and reconfigured apt to use the cache by default. -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: apt-cacher in main
On Wed, 2007-11-14 at 11:38 -0500, Fabian Rodriguez wrote: > I also think this would be a good candidate to have on the LiveCD > installer and/or the Server CD installer images, as in many scenarios > CDs are used for a first install and then other PCs in the same LAN > could use that first install apt-cacher to save tremendously on > bandwidth - a common situation outside high-bandwidth areas. I am not sure it needs to be moved. But, what would be totally cool is if the installer scanned the local network on install and configured apt-cacher in sources.list instead of the normal repos by default when if finds a server. That would be a terrific usability upgrade. But, since only one server needs it, is there an advantage to moving it from Universe? It not like its in Multiverse which is turned off by default. If I remember correctly, isn't Universe turned on by default on initial install? -- Kevin Fries Senior Linux Engineer Computer and Communications Technology, Inc A Division of Japan Communications Inc. -- 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: Better Remote Upgrade Capabilities - Ideas?
Chris, You are right, but the problem is actually at the core of the DEB/APT system. It will not get fixed any time soon. And for the record, RPM is not any better, it suffers from the same problem. The core of the issue stems from the fact that software exists in a n-dimension matrix. This matrix is being modeled by a flat file structure. This can not be done fully. Now add the untold number of changes done by the developers to make our users desktops more useful. That model keeps changing, and the flat files are trying to keep up. This is not easy, and one little mistake can show up in a multitude of different ways. Toss in the problems caused by third party repositories (and any solution that is incompatible with third party repositories is 100% broke in my opinion) and the problem becomes unmanageable. The real answer would be a object oriented model, but that would break the DEB/APT system, and cause ciaos on an untold number of fronts. The easiest way around the problem is a fresh install. The problem is Debian (and thus Ubuntu) does not make this easy. I feel that stealing the reinstall procedures from Fedora could resolve this problem without resorting to overhauling the APT system. That is why I am seeking to contribute in this way. Kevin Fries On Fri, 2007-09-14 at 18:09 +0100, Chris Warburton wrote: > I find that Debian is the most consistent through upgrades, if no > strange customisations have been made then "apt-get dist-upgrade" can go > from version to version no problem. > > Ubuntu introduces some issues with this that makes the update manager > and metapackages like ubuntu-desktop needed. I have personally gone > through quite a few reinstalls due to badly upgraded Ubuntus, but I > stick with it due to the awful hardware in my laptop and the community > (I feel like I can contribute to Ubuntu, whereas Debian is rather harder > to be appreciated, especially for non-programmers) > > So whilst I'm not overly familiar with all of the issues involved in > upgrading between releases, it is sensible to use the best tool for the > job, and if Ubuntu's upgrade system doesn't fit your scenario nicely > then Debian probably would, and they are not too dissimilar in terms of > underlying technologies either. > > Hope that helps, > Chris Warburton > > PS: This is not having a go at Ubuntu, or advertising Debian, I'm just > saying that in this situation Debian might be a better choice than > Ubuntu. Yay competition! > > -- 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: Better Remote Upgrade Capabilities - Ideas?
On Fri, 2007-09-14 at 19:20 +0200, Oliver Grawert wrote: > hi, > Am Freitag, den 14.09.2007, 10:43 -0600 schrieb Kevin Fries: > > The version upgrade process from one version of Ubuntu to another has > > been troublesome for the most part. I am beginning to trust it less and > > less. > did you use the update-manager way [1] as we recommend for that or plain > apt-get (which is known to produce issues, thats the reason we > officially discourage this way of upgaring since serveral releases > already) > also did you file bugs for breakages you found during the upgrade > process so the responsible developers are aware ? # update-manager The problems are really basic... There is too much going on to be done reliably. While I held out hope, and still do... its not quite there. I do not ever expect this method to get truly reliable until some fundamental changes in the package management system are implemented. But those types of changes would take Ubuntu into some very harrowing territory. It would mean a break from APT which is horrible unless compared to all the rest. I do not expect Ubuntu to do anything that radical, so I am seeking a way to work around the problem. Thx Kevin Fries -- 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
Better Disaster Recovery
I just sent a remote install process inquiry, but I have a second semi-related idea I would like to begin working on and contribute. If I do this right (a rarity for me), these two ideas could be combined into one... One of the holly grails of enterprise network management is data and system recovery. Backup processes backup data, and sometimes settings (easier in Linux than Windows for sure). However, I have yet to see a good enterprise tool in any OS that allows quick, easy, automated system and/or network recovery. Here was what I would like to start on, and am fishing for ideas on how this might work for Ubuntu: Each machine backs up each night. The backup consists of /home, /etc, and a list of all packages installed (I believe apt stores this info somewhere), and disk layouts. All of this information should be able to be fed into a restore process to automatically rebuild a machine in case of disaster. The server that handles all the disaster recovery, would then replicate that data for all machines, including itself, to an offsite location. Each machine should perform its backup based upon a local /etc setting to allow it to optimize (desktops could update immediately, but a mail server would get overloaded if it replicated every file as it was written, so would take more of a snapshot methodology) for its own needs. But the offsite would be replicated based upon FAM. The eventual goal is to say: "Our enterprise tools in Ubuntu are so good, your entire building could burn down at 11am, and we can rebuild your office infrastructure elsewhere inside 3 hours from the time the replacement computers showed up". Let Microsoft match that one, lol. I think Ubuntu has the tools, I want to figure out how to tie them together in a way to make this happen. Any ideas, or starting points? Thx Kevin Fries -- 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
Better Remote Upgrade Capabilities - Ideas?
For a long time, I used to use Fedora exclusively. The cleaner more user friendly Ubuntu desktop has been much better for me as a consultant in every way sans one. Remote update between versions. Here is how I would do this in Fedora: First I would go to a server, and dump the contents of the current CD into a directory. I would then make that CD available via NFS. Since nothing on that CD was proprietary, I could open that read-only to the world... step one, quite easy. Second, I would create a kickstart file that would essentially give the install just enough intelligence to start up, configure the network, and basically nothing else... step two, tricky, but once done, easily cloned Third, I would ssh into the box, and download the ISOLINUX folder from the CD into the target's /boot directory... step three, trivial Fourth, I would edit the grub menu.lst file to have a menu item that would boot from the /boot/isolinux folder... step four trivial Then I would wait until it was time to do the actual upgrade. Once that time arrived: Step Six, I would edit the grub menu.lst to make the upgrade option the default boot option. Step seven, I would start VNC in listen mode Step eight, I would reboot the target machine Step nine, I would wait for the target machine's install process to start appearing on my desktop, and would then proceed with the install. I could have automated the entire process, but never quite felt comfortable enough with kickstart to give it full control. The version upgrade process from one version of Ubuntu to another has been troublesome for the most part. I am beginning to trust it less and less. If I wanted to produce a similar methodology for version upgrades, where would someone suggest I start? The parts I like most of the above process are the "Start the install process, and remote the screen to a VNC client listening elsewhere to actually perform the upgrade." Kevin Fries -- 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: Support for multimedia/internet keyboards
On Wed, 2007-08-15 at 11:55 +0100, Chris Warburton wrote: > At LUGRadio Live there was some talk from some RedHat/Fedora people in > the power management BoF about a generic system for submitting keycodes > and things to a central store, so power-user types can try out all of > the buttons on their systems (the use-case discussed was laptops), then > this collected data could be put into an XML file in its own separate > package, able to be updated regularly without impacting anything else > (XML can easily be kept forward and backward compatible). Since I do not > know the applications and things involved in this (and therefore don't > know what to Google for) I am not sure of the implementation, or whether > Ubuntu uses it, but I am sure it is relevant. I believe you are speaking of the: System->Preferences->Keyboard Shortcuts I run a Dell D620, and it has just the volume keys, not all those other multi-media keys, but I am sure its the same thing. Until recently, we changed Distros here in R&D quite often until we settled on Ubuntu. (our corp standard for servers and desktops has traditionally been RedHat, though I have worked to get Ubuntu deployed and is quickly taking over. however, all the embedded devices I have built have always been debian based.) When we had SLED 10.1 and Fedora 6, we needed to use this configurator to make the volume keys work. I am sure its no different for the "Mail" or "Web" keys. When I installed Ubuntu, it was the first distro to actually make the keys work out of the box, without needing to run this program. HTH -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Announcement: One Click Installer
On Wed, 2007-08-08 at 10:09 +0100, Matt Zimmerman wrote: > The two are not mutually exclusive, and an ideal solution would incorporate > both. I can't believe this conversation has gone on this long. Its a really ill conceived idea that is either not explained very well, or has evolved during this thread. First of all, the OP wants a one click install. But we already have that in GDebi, and the upcoming apt:// protocol. If you publish software and use the software out of the Ubuntu repositories, both protocols will use the underlying APT system to pull dependencies, and install in a safe and sane manner. If you are not building based upon the Ubuntu core, you are more likely to brick your system than to get any great functionality... so why would we encourage that behavior because Microsoft does? Can we find a better one? Microsoft does not have to worry about different distros and the OP is all upset that Linux can not reach it full potential until some high school kid from Tallahassee, Rio, or Queensland can simply compile there software, post it on the web, and allow it to be installed on all the distros. The problem is that this is not possible. The impossibility does not come from a technical problem, but instead a political one. Technical problems can be overcome with hard work and technology. Political problems will tie you up in knots for decades without any resolution. The real problem is that not all system use LSB nor do all system distribute their software as binaries. Those distros that don't follow LSB will surely break if you install software that does. Due to the nature of Linux, you can not enforce LSB. Heck, LSB even leaves vague where several key items should be placed (lets start with /opt vs /usr/local or /usr/games vs /usr/shared/games) Therefore, any one size fits all installer will surely only serve a small portion of the install base. As an example, I saw talk of re-inventing alien. But a better Alien is only solving the RPM->DEB or DEB->RPM issue. Lets not forget Gentoo's portage system and all its descendants like T2, Rock, Puppy, etc. If one size truly fit all, ever woman in America should be walking around in a Muumuu. Ladies? Guys want to suggest this to your lady? The reason is that women are not all walking around in muumuus is the same reason this idea will fail... One size does not fit all, and different systems will require different solutions. Viva la difference! While the dream is nobble, and probably worth while, this is not the solution. A better solution would be from the compilation and tools side. A better solution would be to provide a single tool that takes the code, and packages it for deb, rpm, ipkg, tar.gz, and an ebuild all in one command. Then package it up with a solid testing and approval process that makes it easy to get it into the approved repositories for each distro. Maybe a clearing house system for packages. Once an independent developer builds their new nifty widget generator, the nwg project could be posted easily to all the major (and even minor) projects all at once. Without running software though the various testing processes to insure it is safe, we will have the same problem that has Microsoft in the situation they are in right now. Microsoft has such a commanding lead, and there market share is slowly dwindling. The battle is being lost in Redmond, and stability, viruses, bloat, and cost are all playing their part. Linux has MS on all these parts. Linux is more popular than ever. Why would we ever want to begin copying Microsoft's bad habits. One step installer sounds great, but it can not be done safely. As for the OPs problem with Synaptic... That is 500% off base. I know this because I have sat down with end users and showed them synaptic, and the gnome installer. If more geeks like us did this with their favorite Windows user, I believe there would be more people asking why Windows does not install as nicely as Linux. Want proof? http://windows-get.sourceforge.net/ Has anyone stopped to think that in our quest to solve bug #1, that the answer is not to make Linux behave like Windows, but instead, show Windows users a taste of what Linux does well. Linux already does package management well... very well. Now can we get onto other problems -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Announcement: One Click Installer
On Mon, 2007-08-06 at 13:03 -0600, Conrad Knauer wrote: > On 8/6/07, Krzysztof Lichota <[EMAIL PROTECTED]> wrote: > > > I would like to share with you the project I have been working for some > > time now which I think could help solving bug #1. > > > > The problem: > > - Users coming from Windows (and in general beginners) want installation > > of applications to be as easy as possible. Download, Next, Next, Done > > kind of experience. > > Individual DEB files installed with Gdebi provide this sort of thing > currently (e.g. try http://www.getdeb.net/) Or even better: http://digg.com/linux_unix/Ubuntu_Install_applications_from_Internet_with_a_single_click?t=7711876 This solution works first of all within the existing distribution tools, and second without trying to turn Linux into Windows. We all know how well the first Windows turned out, and the OP does not get that point. People are seeking something better than Windows. Many will tell you otherwise, but when pushed, I think overall satisfaction with Windows is right up their with the cell phone companies... Nobody is happy, but don't feel that Linux or Mac are realistic choices... Resolving that misconception about the usability of Linux is how we solve Bug #1. Bug #1 is NOT, and I can not emphasize this enough, _NOT_ going to get fixed by offering the same old broken solution. Many of the problem with the Windows environment is caused by their software installation process (beyond the scope of this thread). The article above in my opinion offers the hope of a better solution, though this cookie is not completely baked. It allows users to seek better solutions in the forums, where they are more likely to be looking for one. It allows software distribution and support to be integrated, so that support leads to solution. This is not the same old broken system. It also encourages uses to discuss these things in forums, where developers can get a getter indication of what people are looking for. There are lots of wins here. I would love to see a similar setup for the repositories also. However, we need to proceed with caution here. If a user has to maintain too many repositories in order to get all their software, that becomes more difficult than anything on the market now. Imagine Oracle maintains their own repository, so does google, so does apache, etc. Having a few repositories actually decreases problems. Look at the Fedora situation... different sets of repositories that are incompatible with one another. If we really want to move backwards on bug #1, lets encourage lots and lots of people to go out and build their own repositories without some sort of approval process. Just my thoughts on this matter. -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Single CD for Server & Desktop?
On Tue, 2007-07-31 at 11:14 +0200, Wouter Stomp wrote: > On 7/30/07, Kevin Fries <[EMAIL PROTECTED]> wrote: > > > > Actually, what would be ideal in the corporate or business environments > > would be a very minimal install with all updates coming from a central > > repository (i.e. not from the CD). Something similar to the old Debian > > or RedHat FTP based install. > > > > A netboot cd image already exists: > http://archive.ubuntu.com/ubuntu/dists/feisty/main/installer-i386/current/images/netboot/ > > Wouter. I have done at least 30 installs for different network setups in the past few months, and did not know that... Maybe that information should be a little easier to find on the Ubuntu website. Not only would that information have helped me on several occasions, but I think it would have prevented the OPs original post. He was just looking for a way of installing from a single media. A net install could offer a easy option. But other than you posting in a newsgroup for developers, how would the average corporate network admin know this? -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Single CD for Server & Desktop?
On Sat, 2007-07-28 at 18:34 -0400, Bryan Haskins wrote: > Yea, server isn't just a subset of desktop, as it installs minimal and > a LAMP setup, so it would significantly bump up the size. We're > pushing close to 700mb now. Plus most users would have no use in > this. It would only be convenient enough for the people who could > handle a server and downloading a new ISO... plus a liveCD for a > server install would just be silly =D In the single case you are correct. But in CORPORATE you could not be more wrong! A single CD would actually be preferable. Actually, what would be ideal in the corporate or business environments would be a very minimal install with all updates coming from a central repository (i.e. not from the CD). Something similar to the old Debian or RedHat FTP based install. I think what would serve the OPs original request, and could be a great asset to businesses large and small would be a setup such as this: You first install a server on your network. Add another option to the install along side the DNS and LAMP options for Domain Master. The domain master would install LDAP as the central user store; GOsa for user management; apt-cacher for package management; ssh server, and create an install user w/ssh key authentication; and a program to create a basic USB key used to install clients. Then use that key (complete with the server's ssh keys) to initiate an install off the server. Or, as an alternative, have the CD ask on boot if this is a server or client install (though I like the key idea much better). Call this distro the Corporate Edition. One media to keep up to date! Also, it could be further extended with the use of kickstart to create multiple install types (Install program contacts server via ssh; home directory of install user has a list of kickstart files; install program offers list of predefined install types; option for a roll your own would be a configuration item). And best of all, it would be a great way of chipping away at bug #1. Try installing that easily in Windows even with SMS, lol. Just my $0.02 -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Using standardized SI prefixes
On Wed, 2007-06-13 at 15:06 +0100, Scott James Remnant wrote: > On Wed, 2007-06-13 at 15:01 +0100, Alex Jones wrote: > > > 1 TB is not rounded. It means precisely 1 × 10^12 bytes, no more and no > > less. > > > No it doesn't. > > The meaning of 1 TB depends on the context, and has always done so. Your correct, 1TB always meant 1 x 2^20 And before you start in again... No that is not context sensitive. The B denotes bytes, which denotes a computer context, and therefore the metric context is no longer applicable. At least historically. However, now that the SI has become standard, we can now allow TB to take on the metric 1 x 10^12 and use TiB to mean 1 x 2^20. Now we eliminate context, and can say what we mean... Unless you are trying to deceive? I hope that is not what you are proposing. Ubuntu is not the only distro having these conversations lately. It is time to start using the standard. Standards are created to get everyone on the same page. I find it unsettling that the one guy arguing against the standard set aside by IEC, IEEE, NIST, CENELEC, and the EU, has an @ubuntu.com address. I would expect someone of that high level to embrace not reject the standards of the industry. -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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: Using standardized SI prefixes
On Wed, 2007-06-13 at 14:29 +0100, Scott James Remnant wrote:
> On Wed, 2007-06-13 at 12:51 +0200, Christof Krüger wrote:
>
> > On Tue, 2007-06-12 at 15:52 +0100, Ian Jackson wrote:
> > > shirish writes ("Using standardized SI prefixes"):
> > > > Please look at http://en.wikipedia.org/wiki/Binary_prefix .
> > >
> > > Urgh, these things are ugly and an abomination. We should avoid them.
> >
> > I'd really like to hear some real arguments against SI prefixes, besides
> > being ugly or funny to pronounce or just because "it has always been
> > like that". Advantages of using SI prefixes has been mentioned in this
> > thread. Please tell me the disadvantages so there can actually be a
> > constructive discussion.
> >
> User Confusion.
>
> Most users do not know what a "tebibyte" is, and they do not care. They
> know that "a terabyte" is "about a million million bytes", and that is
> sufficient.
No it is not!!!
As larger and larger sizes are used, what was once an minor difference,
is starting to become significant. It almost reminds me of that old
scam of taking the rounded portions of a penny in financial calculations
and putting into an account. It adds up fast.
As we move from kilobytes to megabytes to etc the percentage of error
becomes greater and greater. It is no longer approximate, it is out and
out incorrect.
I will be the first to admit that even though I am a computer scientist,
I did not understand the difference between Ki and K. I just always
thought of them as 1024. It actually annoys me to explain to the
uneducated the difference all the time due to what I always thought was
laziness of marketing people. I know realize it is a metric vs binary
thing.
Personally, when Joe Sixpack sees that a drive is 300 GiB or 300 GB he
will likely think it is the same. We all know that is not true.
Knowing it is not true, and purposely ignoring the the fact that the guy
buying the 300GiB drive is getting 7.4% more drive than the guy buying a
300GB drive is being complacent in the fraud.
As the size of drives grow, so does the error. Here is a chart:
kilobyte -> kibibyte -> +2.4%
megabyte -> mebibyte -> +4.9%
gigabyte -> gibibyte -> +7.4%
terabyte -> tebibyte -> +10%
petabyte -> pebibyte -> +12.6%
exabyte -> exbibyte -> +15.3%
zettabyte -> zebibyte -> +18.1%
yottabyte -> yobibyte -> +20.9%
With drives coming out in terabyte sizes soon, this is a 10% difference!
With raids approaching and in some cases exceeding exabyte sizes, this
is more than a 15% difference.
This goes way beyond "approximate". Imagine if you went to the gas
station to by a gallon (litre for our Europeans) of gas for your car,
but only received 85% of the gas you thought you paid for. Now how do
you feel about how close enough the differences are?
As unethical and immoral as it is to sell gas without accurately
reflecting the units/price, it is just as bad for us in the computer
industry to perpetrate this fraud by misreporting the size of memory,
drives, etc.
If Joe Sixpack sees that 300GiB drive on the shelf next to a 300GB
drive, and both are marked accurately, and he assumes they are both the
same size, that is not our problem. He should become a more educated
consumer. If the salesman does not explain the difference (ever been to
Best Buy, lol) that is his fraud. But to purposely report that the
300GB and 300GiB drive are the same, because I am too lazy for field
questions is just immoral, and wrong. And to do so because I am too
lazy to field a few questions while people get used to the difference...
Unethical does not even begin to describe it.
The standard is in place. Use it. Be honest and reflect sizes
correctly. Kb is not the same as KiB. We all know that. If the public
gets confused, there is always Wikipedia. If you are too lazy to
explain it yourself, send them to:
http://en.wikipedia.org/wiki/Binary_prefix
I for one will begin using the SI units, now that I understand the
difference. "Approximate" is just an excuse to be lazy, and is
purposely giving wrong information to people. People look to me to
build data reporting tools... That information should always reflect
accurately unless I place a disclaimer on the data. Laziness should
never be an excuse.
--
Kevin Fries
Senior Linux Engineer
Computer and Communications Technologies, Inc.
a division of Japan Communications, Inc.
signature.asc
Description: This is a digitally signed message part
--
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
Embuntu
dress that effect the devices we produce. How about Japanese language input on a small screen. Every company selling small devices in Japan has to deal with that one. Or how about PHS networking, or the pending change to 3G. Every reader of this list can probably come up with several issues made more difficult, or are just flat different, on an embedded or mobile platform, that I did not mention. Place your own 10th item here. As I hope I have clearly pointed out, and embedded environment has some very different trade-offs from a desktop or server environment. The Ubuntu Mobile group is completely sidestepping these issues in order to build Mameo. They are planing on deploying it on a "Tablet" device. Much of what they are working on will also apply to embedded devices, such as touch screens, and onscreen keyboards. But they are ignoring anything not Mameo related, and are not building for an embedded device. Therefore, I would like to see who else is working on these issues, and work toward getting a real embedded project going. I will take the point if need be. Thanks in advance -- Kevin Fries Senior Linux Engineer Computer and Communications Technologies, Inc. a division of Japan Communications, Inc. signature.asc Description: This is a digitally signed message part -- 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
