Re: Fedora 10 for 9.1.0?
I agree with Marco, no distro-wars, pretty, please. Where I sit, Ubuntu's advantages have decreased, rather than increased over the last couple years. Even Mark Shuttleworth, when I last chatted with him early this year, said it didn't make a significant difference. The wider Sugar and software appropriate for kids is available, the better we all are. - Jim On Wed, 2008-10-15 at 13:41 +0200, Marco Pesenti Gritti wrote: On Wed, Oct 15, 2008 at 10:57 AM, Bert Freudenberg [EMAIL PROTECTED] wrote: Am 15.10.2008 um 01:19 schrieb [EMAIL PROTECTED]: the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. Ubuntu also seems a much better fit in spirit than RedHat. No distribution wars plase :) Let's stay on topic... Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel -- Jim Gettys [EMAIL PROTECTED] One Laptop Per Child ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, Oct 15, 2008 at 5:02 PM, Tomeu Vizoso [EMAIL PROTECTED] wrote: Don't know much about spirits, but AFAIHS, RedHat has contributed enormously to Sugar. Do we really know if a switch from RH to Canonical would have worked better? AFAIK before they tried to do the OLPC thing but eventually cozied up to Intel with the Classmate but nothing much prospered there I think. Although since Canonical has a program for Netbooks they already have some infrastructure for it. -- Jerome G. Website: http://www.gotangco.com Blog: http://engage.wordpress.com ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Tue, Oct 14, 2008 at 18:21, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish Then it seems we should have the expectation of rebasing during every major OLPC release, since we are doing time based releases and so is Fedora. This means we need a strategy of to handle this, and not just find someone for this time, but get better at doing it so that it becomes less of an issue every time. Ubuntu rebases off Debian unstable every six months. When a new development cycle opens, they first update the toolchain - compilers etc - then have a period to suck in debian packages, then a stabilisation period, then a feature freeze, then progressively more freezes before release. See https://wiki.ubuntu.com/IntrepidReleaseSchedule for an example. Regards Morgan ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, Oct 15, 2008 at 11:30 AM, James Cameron [EMAIL PROTECTED] wrote: On Wed, Oct 15, 2008 at 10:57:45AM +0200, Bert Freudenberg wrote: Ubuntu also seems a much better fit in spirit than RedHat. Agreed. xodist could easily be used, and the OLPC specific RPMs could be converted to .deb with not a lot of effort. Launchpad could also offer some benefit in terms of package branches and private package archives. But really, it comes down to what the current developers are familiar with. The technical reasons don't beat not-invented-here. I don't think it comes down to that. I have used debian-based distros since the first debian appeared and today still use Ubuntu on all my personal computers. Needed to learn new stuff to do my work on OLPC like packaging rpms, using koji, etc but nothing that supposed a big effort. What it comes down is the support that you get. RH has invested very talented engineers' time to the project (starting with Marco, what would be Sugar today if he wasn't around?). I'm not sure if Canonical shares the Sugar vision or if they would have invested so much on OLPC stuff. But about the community, I must say that we have had awesome support from both the Fedora and Debian+Ubuntu communities. I'm saying this as an observer, I'm just a contracted developer and don't have any insider information about OLPC deals with other companies. My point is only that RH has already done a great deal of work in Sugar even if it hasn't been advertised widely. Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
Am 15.10.2008 um 11:02 schrieb Tomeu Vizoso: On Wed, Oct 15, 2008 at 10:57 AM, Bert Freudenberg [EMAIL PROTECTED] wrote: Am 15.10.2008 um 01:19 schrieb [EMAIL PROTECTED]: the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. Ubuntu also seems a much better fit in spirit than RedHat. Don't know much about spirits, but AFAIHS, RedHat has contributed enormously to Sugar. Do we really know if a switch from RH to Canonical would have worked better? I did not mean to belittle RedHat's contributions and much less that of the Fedora community. That's not what I was suggesting, and I also would not lump the RedHat developers with RedHat as a company, besides that is history anyway so we cannot change it to see where things would stand if this or that did / did not happen. My point was that overall, RedHat is targeting businesses, where Ubuntu aims for individuals. The latter seems to be better aligned with the OLPC goals. That's all. Anyway, I'll desert the rwar before it begins ... - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
Am 15.10.2008 um 01:19 schrieb [EMAIL PROTECTED]: the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. Ubuntu also seems a much better fit in spirit than RedHat. - Bert - ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, Oct 15, 2008 at 10:57:45AM +0200, Bert Freudenberg wrote: Ubuntu also seems a much better fit in spirit than RedHat. Agreed. xodist could easily be used, and the OLPC specific RPMs could be converted to .deb with not a lot of effort. Launchpad could also offer some benefit in terms of package branches and private package archives. But really, it comes down to what the current developers are familiar with. The technical reasons don't beat not-invented-here. -- James Cameronmailto:[EMAIL PROTECTED] http://quozl.netrek.org/ ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, Oct 15, 2008 at 10:57 AM, Bert Freudenberg [EMAIL PROTECTED]wrote: Am 15.10.2008 um 01:19 schrieb [EMAIL PROTECTED]: the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. Ubuntu also seems a much better fit in spirit than RedHat. No distribution wars plase :) Let's stay on topic... Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, 15 Oct 2008, Jim Gettys wrote: I agree with Marco, no distro-wars, pretty, please. Where I sit, Ubuntu's advantages have decreased, rather than increased over the last couple years. Even Mark Shuttleworth, when I last chatted with him early this year, said it didn't make a significant difference. from a single-shot conversion point of view I agree that there is no significant difference. however, what started this conversation was the desire to figure out what fedora builds were going to be the base of a long-term-supported RHEL release. That information is not available for the redhat family, but is available for Ubuntu. That could be a significant difference. The wider Sugar and software appropriate for kids is available, the better we all are. definantly. David Lang - Jim On Wed, 2008-10-15 at 13:41 +0200, Marco Pesenti Gritti wrote: On Wed, Oct 15, 2008 at 10:57 AM, Bert Freudenberg [EMAIL PROTECTED] wrote: Am 15.10.2008 um 01:19 schrieb [EMAIL PROTECTED]: the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. Ubuntu also seems a much better fit in spirit than RedHat. No distribution wars plase :) Let's stay on topic... Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Tue, Oct 14, 2008 at 6:49 PM, Tomeu Vizoso [EMAIL PROTECTED] wrote: On Tue, Oct 14, 2008 at 6:21 PM, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish gtk 2.14 has some good stuff, would be nice to start to rebase on F10 ASAP so we don't have so much stress as with the F8 rebase. Ahem, I meant the F9 rebase. Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
Marco - I think this is one of the important questions we should be discussing in the near term. I'm not advocating either for or against it, but simply that it is something we should consider seriously. That includes identifying all the consequences/implications of rebasing on F10, along with any advantages. - Ed On Oct 12, 2008, at 7:34 AM, Marco Pesenti Gritti wrote: Hello, do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... http://lists.laptop.org/pipermail/sugar/2008-October/009194.html Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Tue, Oct 14, 2008 at 6:21 PM, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish gtk 2.14 has some good stuff, would be nice to start to rebase on F10 ASAP so we don't have so much stress as with the F8 rebase. Regards, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, Oct 15, 2008 at 5:21 AM, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish Yeah - the Fedora lifecycle does not end up being a good fit for us. There is no clear (supported) path to go from a Fedora (bleeding edge) release to a LTS path with RHEL or CentOS. Is there any hints as to how that could be improved? Understanding how RH picks where to base RHEL would be a start... On one hand we need the latest freshest code as we're driving quite a few changes in the stack -- but we also need LTS. (I don't mean to complain or flame -- the focus is on lifecycle in the first para. Some things have been fantastic, including the beginning of the lifecycle - IOWs how quickly Fedora picks upstream changes...) cheers, m -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
We went through this early on in the development of the OLPC software stack. It became clear that we were not far enough along to be able to settle in on RHEL. Maybe we'll be at that point after another turn or two of the crank. Maybe the XS will be there sooner. But too much is in flux. -walter On Tue, Oct 14, 2008 at 7:06 PM, Martin Langhoff [EMAIL PROTECTED] wrote: On Wed, Oct 15, 2008 at 5:21 AM, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish Yeah - the Fedora lifecycle does not end up being a good fit for us. There is no clear (supported) path to go from a Fedora (bleeding edge) release to a LTS path with RHEL or CentOS. Is there any hints as to how that could be improved? Understanding how RH picks where to base RHEL would be a start... On one hand we need the latest freshest code as we're driving quite a few changes in the stack -- but we also need LTS. (I don't mean to complain or flame -- the focus is on lifecycle in the first para. Some things have been fantastic, including the beginning of the lifecycle - IOWs how quickly Fedora picks upstream changes...) cheers, m -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel -- Walter Bender Sugar Labs http://www.sugarlabs.org ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
Hi, gtk 2.14 has some good stuff, would be nice to start to rebase on F10 ASAP so we don't have so much stress as with the F8 rebase. Once we decide to do this, we'll need to talk about who would do it -- for the F7 rebase we had J5, and then Dennis for F9, and it's not clear who could take the lead on F10. Once we know we intend to go ahead with a rebase, we could ask the fedora-olpc-list if anyone's interested in volunteering to spearhead it? Jeremy seems like an obviously great candidate, but I wouldn't expect the Fedora-on-XO work to be finished up for another month or so, and we ran into troubles with rebasing too late in the release cycle last time. Just some thoughts, - Chris. -- Chris Ball [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Tue, 14 Oct 2008, Walter Bender wrote: We went through this early on in the development of the OLPC software stack. It became clear that we were not far enough along to be able to settle in on RHEL. Maybe we'll be at that point after another turn or two of the crank. Maybe the XS will be there sooner. But too much is in flux. I don't want to start a distro flamefest, but the question needs to be asked. the distro landscape has changed a bit in the last few years, is it worth considering a jump to Ubuntu if it has a better fit for your release cycle? at the very least it telegraphs the long-term support versions. it's not like you really depend on the underlying distro for very much. it's mostly a convienient codebase to start from in developing your own distro. David Lang -walter On Tue, Oct 14, 2008 at 7:06 PM, Martin Langhoff [EMAIL PROTECTED] wrote: On Wed, Oct 15, 2008 at 5:21 AM, Jeremy Katz [EMAIL PROTECTED] wrote: On Sun, 2008-10-12 at 13:34 +0200, Marco Pesenti Gritti wrote: do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... If not, then you're going to be basing on a Fedora release which will be EOL'd[1] very soon after the OLPC release... Jeremy [1] Fedora 9 EOL will be 1 month after Fedora 11's release, so say June-ish Yeah - the Fedora lifecycle does not end up being a good fit for us. There is no clear (supported) path to go from a Fedora (bleeding edge) release to a LTS path with RHEL or CentOS. Is there any hints as to how that could be improved? Understanding how RH picks where to base RHEL would be a start... On one hand we need the latest freshest code as we're driving quite a few changes in the stack -- but we also need LTS. (I don't mean to complain or flame -- the focus is on lifecycle in the first para. Some things have been fantastic, including the beginning of the lifecycle - IOWs how quickly Fedora picks upstream changes...) cheers, m -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Wed, 2008-10-15 at 12:06 +1300, Martin Langhoff wrote: Yeah - the Fedora lifecycle does not end up being a good fit for us. There is no clear (supported) path to go from a Fedora (bleeding edge) release to a LTS path with RHEL or CentOS. Is there any hints as to how that could be improved? Understanding how RH picks where to base RHEL would be a start... RHEL is based off of the Fedora release which is most fitting for the desired RHEL schedule. Which makes it a bit of a black art and not entirely replicable. Past RHEL releases... RHEL 2.1 - based on Red Hat Linux 7.2 RHEL 3 -- based on something between Red Hat Linux 9 and what became FC1 RHEL 4 -- based on Fedora Core 3 RHEL 5 -- based on Fedora Core 6 The only real hint I have is that it's a sliding scale of which is more important -- having distro support for an extended period of time or having the most recent stuff. OLPC to this point, as Walter alludes, has definitely leaned more towards the latter. It's possible, perhaps even likely, that this will start to skew towards the former as OLPC matures as a platform. I think that this would be a great topic for some more in-depth discussion at the planning meeting previously mentioned On one hand we need the latest freshest code as we're driving quite a few changes in the stack -- but we also need LTS. The problem comes in that even if Fedora _were_ to pick hey, let's take Fedora n and choose it for longer term, it would only help a subset of the people. And the cost in doing so would end up negatively impacting the quickness with which we do other things. In addition, the update policies within Fedora (rebase early, rebase often? :) don't really mesh well with the needs of someone really looking at any sort of long-term support from my experience But, that's a flamewar which is ongoing on another list and which I stayed out of there, so I'm going to let it go after this here too :-) Jeremy ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Fedora 10 for 9.1.0?
On Tue, 2008-10-14 at 19:12 -0400, Chris Ball wrote: gtk 2.14 has some good stuff, would be nice to start to rebase on F10 ASAP so we don't have so much stress as with the F8 rebase. Once we decide to do this, we'll need to talk about who would do it -- for the F7 rebase we had J5, and then Dennis for F9, and it's not clear who could take the lead on F10. Once we know we intend to go ahead with a rebase, we could ask the fedora-olpc-list if anyone's interested in volunteering to spearhead it? There's the start of a listing of what some of the deltas from Fedora 10 to the current OLPC software stack is in bugzilla (https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=FedoraOLPCDelta iirc) which would be where I'd recommend anyone interested to start with trying to get some of the changes resolved before Fedora 10 is released. Longer-term, I think at least from my ideal point of view, an OLPC software release can be viewed largely as just a spin of Fedora bits plus perhaps a few things which are not suitable for Fedora, thus making the amount of rebasing needed slim to none. One thing that would help there is having more OLPC people involved in maintaining/co-maintaining packages which are critical to OLPC in Fedora. If anyone's interested in stepping up here and needs pointers on how to get started, let me know. Jeremy ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Fedora 10 for 9.1.0?
Hello, do we plan to rebase to F10 for 9.1.0? I'm asking because I'd need to know if I can depend on gtk 2.14... http://lists.laptop.org/pipermail/sugar/2008-October/009194.html Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
