Re: Recent Updates to Sugar Almanac
Hi Faisal, sorry to get to this so late, are there any outstanding doubts? The DS is right now a big mess due to long-standing bugs and basic features still unimplemented. I hope that during the next release significant resources will be allocated to this, if in the meantime you could suggest better APIs (both high and low level), that would be awesome. Thanks, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Updates This Week to the Sugar Almanac - Using the Datastore and More
On Fri, Jun 27, 2008 at 5:17 PM, Faisal Anwar [EMAIL PROTECTED] wrote: Hi Tomeu, Thanks so much for the clarifications. I understand now the abstraction intended for metadata accessed through DSMetadata and DSObject and will try to write that up a little more forcefully. I guess the main thing I was concerned about was having a consistent and lasting interface to the datastore (which can certainly be implemented in the form of a dictionary if that is most appropriate). I hope this abstraction will hold moving forward so that activity developers are confident that their hooks in to datastore will work with the same behavior over time (this is part of the purpose of the documentation I'm working on). Yes, I'm confident that we'll be able to maintain compatibility with activities that use the current API, although I also hope that we'll come up with something much better soon. With regards to the lower level DBus calls in datastore and elsewhere, what I'm really looking out for is what levels of the tech stack an activity developer will need to learn to build a fairly robust and complex activity. Of course, for some things lower level Dbus calls may be unavoidable, but I'm assuming that there should be a substantive interface in python that abstracts away the Dbus functionality for most developers. As I go through more of the code in datastore, presence, and elsewhere, I'll try to identify cases where perhaps a low-level Dbus call could be abstracted to some standard python calls. Well, the low level API isn't really that hard to use. The dbus-python bindings already do a great work of making easier to use DBus services from python. If you find any capability in the low level bindings that could be useful to activities and is not yet in the python wrapper, please tell. Thanks, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Questions, OLPC-Caldas
On Tue, Jul 1, 2008 at 01:14, Bobby Powers [EMAIL PROTECTED] wrote: 2008/6/30 Carlos Dario Isaza Zamudio [EMAIL PROTECTED]: 2) What's the password when i try to access the XO through SSH? you need to set the root password before you can do this. if you're in the Terminal activity: sudo passwd should give you a prompt to set the new root password. I've heard that changing the user olpc's password can have unintended consequences. (not sure where this is documented, tho I think its on the wiki somewhere I usually set a password for user olpc and ssh in as olpc. I haven't noticed any unintended consequences... Regards Morgan ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: boot timings
Am 27.06.2008 um 21:55 schrieb [EMAIL PROTECTED]: hello -- yesterday, as much for an exercise in using the serial port, manipulating the kernel commandline, and doing a little exploring, i resurrected an old tool of mine which timestamps lines received over a serial port (or a socket), and i used it to get a trace of XO startup and shutdown. Nice! Thanks. So hardware-wise, USB and the touchpad seem to be time hogs, but it seems unlikely we can do anything about that. On the software-side, sshd takes close to 3 secs to start, would make sense to disable in user builds - without a password it is not useful anyway. But the one thing that really stands out as silly is that we could shave a whopping 20 secs off the shutdown by not trying to unmount the missing /security/.private ... - Bert - there's no scientific method here, nor any great conclusions, but since we don't usually even see the boot logs, i thought i'd pass along the results, for general interest. oh -- this was done using build 708, on a unit with the dev key installed, of course. paul =- paul fox, [EMAIL PROTECTED] 0.000| 0.000|Forthmacs 0.000| 1.216|Type 'i' to interrupt stand-init sequence 0.000| 0.000|First stand-init: 0.209| 0.208|Exceptions 0.209| 0.000|CIF 0.209| 0.000|DHCP init 0.209| 0.000|memory node 0.209| 0.000|gpio 0.209| 0.000|Probing memory 0.209| 0.000|MMU 0.209| 0.000|Reclaim dictionary 0.209| 0.000|PCI host bridge 0.209| 0.000|ISA 0.209| 0.000|RTC 0.209| 0.000|CPU nodes 0.209| 0.000|CMOS 0.209| 0.000|Null-NVRAM 0.209| 0.000|board revision 0.209| 0.000|Date to EC 0.240| 0.031|Wireless reset 0.257| 0.016|PCI properties 0.257| 0.000|Manufacturing data 0.257| 0.000|USB setup 0.561| 0.303|Suspend/resume 0.561| 0.000|Century 0.577| 0.015|Init SHA-1 variables 0.577| 0.000| 0.577| 0.000|USB2 devices: 2.241| 1.664|/pci/[EMAIL PROTECTED],5/[EMAIL PROTECTED],0 2.529| 0.288|USB1 devices: 2.529| 0.000|OLPC C2, 256 MiB memory installed, S/N CSN748056BF 2.685| 0.156|OpenFirmware CL1 Q2D16 Q2D 2.685| 0.000| 2.685| 0.000|Type the Esc key to interrupt automatic startup 2.701| 0.015|5 3.708| 1.006|4 3 5.724| 2.016|2 1 7.740| 2.016| Boot device: /nandflash:\boot\olpc.fth Arguments: Boot device: /nandflash:\boot\olpc.fth Arguments: 15.632| 7.891|Boot device: /pci/[EMAIL PROTECTED]:\boot\vmlinuz Arguments: ro root=mtd0 rootfstype=jffs2 console=ttyS0,115200 fbcon=font:SUN12x22Boot device: /pci/[EMAIL PROTECTED]:\boot\vmlinuz Arguments: ro root=mtd0 rootfstype=jffs2 console=ttyS0,115200 fbcon=font:SUN12x22 16.609| 0.976|Loading ramdisk image from /pci/[EMAIL PROTECTED]:\boot \olpcrd.img ... 21.341| 4.732|[0.00] Linux version 2.6.22-20080523.1.olpc. 28f4cb6e780db07 ([EMAIL PROTECTED]) (gcc version 4.1.2 20070925 (Red Hat 4.1.2-27)) #1 PREEMPT Fri May 23 03:06:12 EDT 2008 23.481| 2.139|[0.00] BIOS-provided physical RAM map: 23.481| 0.000|[0.00] BIOS-e801: - 0009f000 (usable) 23.481| 0.000|[0.00] BIOS-e801: 0010 - 0edfd000 (usable) 23.481| 0.000|[0.00] 237MB LOWMEM available. 23.482| 0.000|[0.00] Zone PFN ranges: 23.482| 0.000|[0.00] DMA 0 - 4096 23.482| 0.000|[0.00] Normal 4096 -60925 23.482| 0.000|[0.00] early_node_map[1] active PFN ranges 23.524| 0.041|[0.00] 0:0 -60925 23.524| 0.000|[0.00] DMI not present or invalid. 23.524| 0.000|[0.00] Allocating PCI resources starting at 1000 (gap: 0edfd000:f1203000) 23.524| 0.000|[0.00] Built 1 zonelists. Total pages: 60450 23.524| 0.000|[0.00] Kernel command line: ro root=mtd0 rootfstype=jffs2 console=ttyS0,115200 fbcon=font:SUN12x22 23.525| 0.000|[0.00] Initializing CPU#0 23.525| 0.000|[0.00] CPU 0 irqstacks, hard=c0714000 soft=c0713000 23.525| 0.000|[0.00] PID hash table entries: 1024 (order: 10, 4096 bytes) 23.567| 0.041|[0.00] Detected 431.221 MHz processor. 23.567| 0.000|[ 23.150198] Console: colour EGA 80x25 23.567| 0.000|[ 23.150328] Dentry cache hash table entries: 32768 (order: 5, 131072 bytes) 23.567| 0.000|[ 23.150936] Inode-cache hash table entries: 16384 (order: 4, 65536 bytes) 23.567| 0.000|[ 23.188408] Memory: 228104k/243700k available (2142k kernel code, 15040k reserved, 810k data, 168k init, 0k highmem) 23.568| 0.000|[ 23.188473] virtual kernel memory layout: 23.568| 0.000|[ 23.188483] fixmap : 0xd000 - 0xf000 ( 8 kB) 23.611| 0.042|[ 23.188496] vmalloc : 0xcf80 - 0xb000 ( 775 MB) 23.611| 0.000|[ 23.188508] lowmem : 0xc000 - 0xcedfd000 ( 237 MB) 23.611| 0.000|[ 23.188520] .init : 0xc06e4000 -
Re: Can't upload a .po file to Pootle
Sayamindu Dasgupta [EMAIL PROTECTED] writes: Another thing that may have caused the problem - http://lumiere.ens.fr/~guerry/etoys.po has a formatting error at line 2683. Fixed: http://lumiere.ens.fr/~guerry/etoys.po You can upload and overwrite the current etoys.po file. Thanks very much! PS: I tried to upload the fixed version of this file and the server rejected it, so it's really a matter of file size, not of corrupted file.. -- Bastien ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: my XO has difficulty with f9
On Mon, Jun 30, 2008 at 9:02 PM, Mikus Grinbergs [EMAIL PROTECTED] wrote: thereafter every attempt at booting would stop somewhere. [If I used the 'check' key to get out of pretty boot, the booting process would always stop after the console message Starting anacron:. If I used manual intervention to get to the text console, after the Starting anacron: message there would be Well, until the sugar guys fix: http://dev.laptop.org/ticket/7357 then it will always seem like the F9 builds freeze with pretty boot. =( But I don't really know what to make of the rest of your problem report. If you could carefully boot the latest joyride with pretty boot disabled (if you forget to press 'check' at boot you can also just switch to the console before the clock animation is complete) and tell us exactly what the last message is -- after anacron, X is supposed to start, and currently sugar takes a *long* time to start up fully. Assuming X is starting, what's the last screen you see look like? --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [PATCH] Install customization packages left for us by a USB key.
On Mon, Jun 30, 2008 at 10:19 PM, Michael Stone [EMAIL PROTECTED] wrote: On Mon, Jun 30, 2008 at 04:58:33PM -0400, C. Scott Ananian wrote: + yum -yt --nogpgcheck install $pkgs 1. As an earlier commenter hinted, you want localinstall because otherwise yum may try to talk to the network in order to download its header cache and to look for newer versions of packages that satisfy dependencies of the installation set. k, thx. @@ -212,6 +228,11 @@ __EOF__ echo olpc-configure: replaying rainbow spool... /usr/sbin/rainbow-replay-spool fi + + # developer customizations. + if /usr/bin/olpc-test-devkey -q ; then + install_customization_packages 2. Do we need a guard for the existence and executability of olpc-test-devkey? it's in the Requires clause for the package, but also bash does not echo 'yes' given: $ if /asdasd ; then echo yes ; fi 3. Why do we care whether there's a devkey? We would actually be better off checking that all the RPMs we're installing are owned by uid 0, this being the exact privilege that we're attempting to safeguard. because we're also trying to enforce P_SF_RUN and a whole bunch of other random things; all of which everyone seems to agree can be subsumed under you're a developer, you can shoot yourself in the foot if you want to. The loosey-goosey but this is highly likely to break when you upgrade between major releases objection, for instance, is answered by the foot-shooting permission. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: boot timings
On Tue, Jul 1, 2008 at 5:15 AM, Bert Freudenberg [EMAIL PROTECTED] wrote: Am 27.06.2008 um 21:55 schrieb [EMAIL PROTECTED]: On the software-side, sshd takes close to 3 secs to start, would make sense to disable in user builds - without a password it is not useful anyway. I think one near-term solution is to try to use upstart to defer more service startups, so that we launch X (and sugar) just as soon as possible. sshd and anacron (for instance) would be perfectly happy starting 10 seconds or a minute later. (help wanted here) But the one thing that really stands out as silly is that we could shave a whopping 20 secs off the shutdown by not trying to unmount the missing /security/.private ... Yes, that's part of the olpcrd security work that's on my plate. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: boot timings
bert wrote: Am 27.06.2008 um 21:55 schrieb [EMAIL PROTECTED]: hello -- yesterday, as much for an exercise in using the serial port, manipulating the kernel commandline, and doing a little exploring, i resurrected an old tool of mine which timestamps lines received over a serial port (or a socket), and i used it to get a trace of XO startup and shutdown. Nice! Thanks. So hardware-wise, USB and the touchpad seem to be time hogs, but it seems unlikely we can do anything about that. the only suspicious USB-related thing that i noticed was the 1.6 seconds in the firmware, apparently to discover the wireless. in the normal boot case i wouldn't think we'd need this, but perhaps in the truly normal case (no dev key) we don't do it. i didn't re-run the test with no dev key. On the software-side, sshd takes close to 3 secs to start, would make sense to disable in user builds - without a password it is not useful anyway. right. this is exactly the kind of thing this tool is good for spotting. just as scott wants to put a stake in the ground about not letting the next release get bigger, i think it would also be worth not letting it get slower. paul =- paul fox, [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: OLPC-Update + RPMs WAS:Re: OLPC XO Opera browser as Sugar activity
On Mon, Jun 30, 2008 at 9:03 PM, [EMAIL PROTECTED] wrote: what I would really like to see is for OLPC to not just release the snapshots, but to have a way for developers to get the rest of the build environment, complete with either the scripts, or command logs of what is done to go from the fedora build to the OLPC build. (This may already be available and I just don't know where to look) http://wiki.laptop.org/go/Pilgrim http://wiki.laptop.org/go/Building_custom_images I would then like to see someone maintain another base-level distro that can run on the OLPC, but not be based on Sugar so that people who want a normal distro can use one, and also so that various performance and usability issues can be identified as being caused by the software vs being caused by the limited hardware. there have been a few people who have made single-shot builds, but AFAIK nobody has maintained/improved the image after the initial 'here, see, it boots' announcement Partly this is because, once you have a traditional system in place and booting, you can use the package manager on that system to keep it continually up-to-date, so (IMO) there's no much need for me to keep re-releasing http://wiki.laptop.org/go/Installing_Debian_as_an_upgrade (for example). Red Hat developers used to routinely run full Fedora on the machines, but the very limited NAND space make this a tricky proposition, unless you plan on running on an external SD card. I did create a full Edubuntu installation on an SD card, but the result wasn't what I'd call child-friendly, mostly due to lack of a good activity chooser. As Martin points out, what's really been missing is a dedicated maintainer. Perhaps this is because (a) it's pretty easy to get started put a new distro on, but (b) it's really really hard to make something which will satisfy all those who want a non-Sugar distro. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: boot timings
On Tue, Jul 1, 2008 at 8:21 AM, [EMAIL PROTECTED] wrote: bert wrote: Am 27.06.2008 um 21:55 schrieb [EMAIL PROTECTED]: the only suspicious USB-related thing that i noticed was the 1.6 seconds in the firmware, apparently to discover the wireless. in the normal boot case i wouldn't think we'd need this, but perhaps in the truly normal case (no dev key) we don't do it. i didn't re-run the test with no dev key. No, I suspect that's the ~1.5s it takes to verify the signature on the dev key. For many deployments that will be 1.5s to check a signature on an activation lease. The original design was to cache that check in some secure manner, but there's not really any appropriate protected space on the G1G1 design. (SPI flash would be it, but it's too risky to write to routinely.) --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: First Draft Development Process Proposal
On Sun, Jun 29, 2008 at 11:20, Tomeu Vizoso [EMAIL PROTECTED] wrote: 3) Stable collaboration :) I know this is a hard one. We just put Cerebro into joyride. We think that some activities, such as Read, will be easy to modify to use it. You might try it and see. Which activities do you care about most in this regard? (If you want to play with Cerebro on your existing image, then just install the RPM and poke Polychronis if you need help.) I thought the plan was to find a way to use Cerebro without having to rewrite activities. Has this changed or are you just suggesting a short term solution? Cerebro currently has a different API to Telepathy. I chatted to Polychronis while I was at 1CC, and we are considering (as a short term project) porting Read to the Cerebro API as a test case: to investigate/demonstrate how well collaboration performs using Cerebro, using an activity that has shown problems on the current setup. This will also give an idea between the difference between the functionality offered by telepathy-salut and Cerebro. The medium term plan is to integrate Cerebro into a Telepathy connection manager. This may result in building abstractions (Clique) on top of Cerebro in an inefficient way if the models are very different, as they appear to be - and also without using features that Cerebro offers such as file transfer. In any case it may well be an improvement over salut over a mesh network. Based on this, the long term may involve abstracting out the API that activities and Sugar need, to a general API that can use either the Telepathy API or the full Cerebro API - assuming we can't merge these enough. That's my understanding, which will be adjusted based on further communication :) Regards Morgan ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: First Draft Development Process Proposal
On Sun, Jun 29, 2008 at 11:26, Tomeu Vizoso [EMAIL PROTECTED] wrote: Bryan Berry wrote: 3) Stable collaboration :) I know this is a hard one. We just put Cerebro into joyride. We think that some activities, such as Read, will be easy to modify to use it. You might try it and see. Which activities do you care about most in this regard? We care most about Write. I will have to test out Cerebro. Maybe I can get Pradosh, our new intern to work on it this week. The collaboration component in Abiword (AbiCollab) is written in C++, and AFAIK Cerebro currently only offers a Python API. I think that AbiCollab is designed to have different network backends, so that may help in writing a new one that used Cerebro (if there was a C-callable API). AFAIK Cerebro offers a D-Bus API, as Telepathy and Presence Service do. But we're still a way off from having Write work on Cerebro. Morgan ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: olpc-dm, olpc-utils, olpc-session, rainbow etc
Hi Michael, On Tuesday 24 June 2008 02:17, Michael Stone wrote: Where is olpc-session kept? (Of course I can copy it from my XO, but... :) On-XO, rpm -qif `which olpc-session` thanks for that rpm command! I've added it to my rpm-tips+tricks notes :) Is this the latest olpc-util package: http://people.redhat.com/sundaram/olpc-utils-0.15-1.fc7.src.rpm ? Nope. You'd want to run something like: koji latest-pkg dist-olpc2 olpc-utils Thanks, I now found git.laptop.org/projects (again ;) or, on a recent build, rpm -qi olpc-utils (Also, if koji isn't available on your OS, then please consider packaging it; it's very handy for investigating the Fedora build system.) Hm. koji is not available in Debian. Do you know where the upstream sources are kept? Let me know if you require any assistance with rainbow or olpc-utils. olpc-utils is basically ready for upload, the debian package source is available at svn://svn.debian.org/svn/debian-olpc/olpc-utils (yeah, svn, we'll switch to git soonish..) rainbow is next on my list, I guess I will ask you then.. :) I haven't uploaded (and fully polished..) olpc-utils yet, as I'm busy bringing other packages in shape for lenny, as I dont see a realistic chance to support the XO in lenny with a sugar desktop experience like with fedora. Thanks, Holger pgpAzzhxNTbDa.pgp Description: PGP signature ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: my XO has difficulty with f9
On Tue, Jul 1, 2008 at 9:52 AM, Mikus Grinbergs [EMAIL PROTECTED] wrote: Scott, thank you for your response. By saying after anacron, X is supposed to start, you've clarified things for me. My current thinking is that X was not starting for me, and that I may have caused that by (at that first boot) making some sort of change that worked with f7 but not with f9 (I install some rpms, etc.). That seems likely, but the olpc-update system should ensure that you've got a clean system (no extra RPMs) when you revert. I would hope that even if X doesn't start (especially if X doesn't start) you can still get a console and/or use alt-boot (hold down 'O' at boot) to fix things. That's the whole point, after all. and tell us exactly what the last message is I believe I gave adequate information in my post. When I use the 'check' key, the last message is 'Starting anacron'. When instead I just switch to the console, some additional messages show up after the 'Starting anacron' message (each message contains a header, then 'msh0: link becomes ready'). [By the way, I have no wireless at my house - the XO radio has nothing it can talk to.] I've waited 10+ minutes for additional messages - none came. Hmm, I was hoping that when you said no more messages what you means was, X started, and showed various things before hanging, but there was nothing more written to the console during that process. It seems like you're hanging at X startup. Do you remember what RPMs you might have installed to make this happen? I was concerned about randomly seeing jffs2 checksum error messages on the text console (but not as the last message). By saying X is supposed to Yes, http://dev.laptop.org/ticket/7177 Mostly harmless. With apologies, I think I'll wait a bit. Didn't like having to do a complete reflash to get back to a working f7 system - so I'll hold off on f9 Did you try alt-boot (holding down the O gamepad key during boot)? Did it not work? until I again get impatient to see how well does it work?. f9 did not see removable USB storage devices (except my permanent SD card), and there are times when I need to use two of them on my XO (in addition to my USB keyboard and USB mouse). The current Joyride (2089) does *not* as yet provide me the complete working environment that I am used to with Update.1 or Joyride-2056. Yes, this is true. http://dev.laptop.org/ticket/7357 and http://dev.laptop.org/ticket/7348 looks like the two biggest blockers to more people trying joyride right now. I *think* that USB/SD not being mounted when you are looking at the console (as opposed to X/Sugar) is a feature. Do folks think this needs to be fixed? --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: my XO has difficulty with f9
On Tue, Jul 1, 2008 at 2:00 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Mon, Jun 30, 2008 at 9:02 PM, Mikus Grinbergs [EMAIL PROTECTED] wrote: thereafter every attempt at booting would stop somewhere. [If I used the 'check' key to get out of pretty boot, the booting process would always stop after the console message Starting anacron:. If I used manual intervention to get to the text console, after the Starting anacron: message there would be Well, until the sugar guys fix: http://dev.laptop.org/ticket/7357 then it will always seem like the F9 builds freeze with pretty boot. =( We are missing olpc-hardware-manager. I requested the OLPC-3 branch and I'll build it when it's created. Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: my XO has difficulty with f9
On Tue, Jul 1, 2008 at 11:28 AM, Marco Pesenti Gritti [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 2:00 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Mon, Jun 30, 2008 at 9:02 PM, Mikus Grinbergs [EMAIL PROTECTED] wrote: thereafter every attempt at booting would stop somewhere. [If I used the 'check' key to get out of pretty boot, the booting process would always stop after the console message Starting anacron:. If I used manual intervention to get to the text console, after the Starting anacron: message there would be Well, until the sugar guys fix: http://dev.laptop.org/ticket/7357 then it will always seem like the F9 builds freeze with pretty boot. =( We are missing olpc-hardware-manager. I requested the OLPC-3 branch and I'll build it when it's created. Thanks. Can we make a scratch build and throw it in public_rpms while we're waiting? --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [PATCH] Install customization packages left for us by a USB key.
On Tue, Jul 01, 2008 at 08:05:46AM -0400, C. Scott Ananian wrote: 3. Why do we care whether there's a devkey? We would actually be better off checking that all the RPMs we're installing are owned by uid 0, this being the exact privilege that we're attempting to safeguard. because we're also trying to enforce P_SF_RUN and a whole bunch of other random things; all of which everyone seems to agree can be subsumed under you're a developer, you can shoot yourself in the foot if you want to. And, as you will observe here http://dev.laptop.org/git?p=security;a=blob;f=rainbow.txt;hb=HEAD#l101 in my opinion, the cheapest way to implement P_SF_CORE + P_SF_RUN is by turning the root password into a developer key, then by applying a CoW layer such as we recently discussed. The loosey-goosey but this is highly likely to break when you upgrade between major releases objection, for instance, is answered by the foot-shooting permission. It's also answered by the fact that RPM checks dependencies, no? Michael ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [PATCH] Install customization packages left for us by a USB key.
On Tue, Jul 1, 2008 at 11:36 AM, Michael Stone [EMAIL PROTECTED] wrote: http://dev.laptop.org/git?p=security;a=blob;f=rainbow.txt;hb=HEAD#l101 in my opinion, the cheapest way to implement P_SF_CORE + P_SF_RUN is by turning the root password into a developer key, then by applying a CoW layer such as we recently discussed. Right, we'll revisit this when all that is in place. In particular, that means locking down sudo (and things which use it, like sugar-control-panel) far more than we do currently. The loosey-goosey but this is highly likely to break when you upgrade between major releases objection, for instance, is answered by the foot-shooting permission. It's also answered by the fact that RPM checks dependencies, no? Only if you assume that (a) nothing in the RPM set is going to be considered 'critical', and (b) all RPMs are well-behaved. I don't think we can enforce either. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Touchpad issues on joyride
On Mon, 2008-06-30 at 16:24 -0700, Deepak Saxena wrote: Hi, For those of you running into crazy pointer behavior on the latest Joyride builds (2080+ with latest kernel), can you please try the following in the terminal/console and report back on if this helps at all after some extended usage: echo 120 /sys/modules/psmouse/parameters/ignore_delta You meant echo 120 /sys/module/psmouse/parameters/ignore_delta (module, not modules) I am still seeing the crazy mouse behavior quite often. It is perhaps occurring less often, but if there is any improvement it is only a small one. Daniel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Recent Updates to Sugar Almanac
Hi Tomeu, No problem about the late reply ... I am documenting improvements as I document the current state of the system. So I'll try to suggest some more coherent api as well. Let's keep in touch about this. Faisal On Tue, Jul 1, 2008 at 4:49 AM, Tomeu Vizoso [EMAIL PROTECTED] wrote: Hi Faisal, sorry to get to this so late, are there any outstanding doubts? The DS is right now a big mess due to long-standing bugs and basic features still unimplemented. I hope that during the next release significant resources will be allocated to this, if in the meantime you could suggest better APIs (both high and low level), that would be awesome. Thanks, Tomeu ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
GLX on XO questions
I'm trying to get Panda3D running on an XO, as a feasibility study, more than for any specific goal. I've successfully loaded Panda3D and Mesa-OpenGL. When I try to run a sample program, Panda3D complains about not finding GLX. Last month there was a discussion here about OpenGL and GLX, but there did not seem to be any resolution of how to get them in. Is there a known path to get GLX in? In the build notes for Joyride 2072 (which seems to be fc9) there is the following line: --- Changes for vnc 4.1.2-30.fc9 from 4.1.2-20.fc7 --- + rebuild against new mesa to fix GLX under anaconda (#443635) Does this mean that GLX exists in these builds, and that I should either try with a fc9 Joyride or wait for this to reach the main branch? Thanks, Bob ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Touchpad issues on joyride
On Jul 01 2008, at 11:58, Daniel Drake was caught saying: On Mon, 2008-06-30 at 16:24 -0700, Deepak Saxena wrote: Hi, For those of you running into crazy pointer behavior on the latest Joyride builds (2080+ with latest kernel), can you please try the following in the terminal/console and report back on if this helps at all after some extended usage: echo 120 /sys/modules/psmouse/parameters/ignore_delta You meant echo 120 /sys/module/psmouse/parameters/ignore_delta (module, not modules) Yep. I am still seeing the crazy mouse behavior quite often. It is perhaps occurring less often, but if there is any improvement it is only a small one. Daniel, Can you do: echo 1 /sys/module/psmouse/parameters/tpdebug And send me the data log for the time you start moving around to a few seconds after you start seeing the bad behavior? Easiest way to do this is edit /etc/rsyslog.d and update the kern.* line to: kern.* /var/log/messages Then restart rsyslogd. When done, just email me /var/log/messages. Thanks, ~Deepak -- Deepak Saxena [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: my XO has difficulty with f9
I may have caused that by (at that first boot) making some sort of change that worked with f7 but not with f9 That seems likely, but the olpc-update system should ensure that you've got a clean system (no extra RPMs) when you revert. I would hope that even if X doesn't start (especially if X doesn't start) you can still get a console and/or use alt-boot (hold down 'O' at boot) to fix things. That's the whole point, after all. Close, but no cigar. The first time (after X did not start) I booted with 'O'. The previously-running version came up, but without Journal or any Activities. I looked everywhere (/var/log, /home/olpc/.sugar/default/logs, etc.) for a clue as to how come -- but found no footprints anywhere. And could not start anything meaningful from the text console - kept getting messages like no display. To be able to continue, did a nand reflash. And that wiped having something for 'O' to load. Don't remember what all customizations I had tried to apply to f9, that could have screwed up X starting. http://dev.laptop.org/ticket/7357 and http://dev.laptop.org/ticket/7348 looks like the two biggest blockers to more people trying joyride right now. To me they're not. If I can bypass 7357 with 'check', that's good enough for me. And I actually LIKE 7348 -- I normally do not use the Journal to access removable storage devices, so having no Journal icon is no problem for me. In fact, I wrote http://dev.laptop.org/ticket/6584 to complain about the Journal scanning removable storage devices for their content; that ticket has now been closed with this notation about removable storage devices: USB sticks will stop being available from the journal, instead being accessible in a simpler view in the shell. I am perfectly willing to do outside-of-Journal mounts. If I remember correctly (things were not going right), when in Joyride 2089 I plugged an USB stick directly into the XO, nothing of any note happened. In particular, the drivers did not create (recognize?) a /dev DEVICE that I could manually use in a 'mount' command. mikus p.s. IIRC, the time I tried plugging my USB hard drive directly into the XO (Joyride 2089, Q2D16), OFW complained mightily about partition-type being Linux on a DOS or something device. [OFW in f9 was not able to access that hard drive - telling me 'Can't open disk label package'. OFW in f7 has no problem, even though it's the same firmware !] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
New joyride build 2094
http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2094 Changes in build 2094 from build: 2092 Size delta: -0.79M -fonts-thai-ttf 0.4.4-1olpc1.2 -- This mail was automatically generated See http://dev.laptop.org/~rwh/announcer/joyride-pkgs.html for aggregate logs See http://dev.laptop.org/~rwh/announcer/joyride_vs_update1.html for a comparison ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Inappropriate use of private meetings lists.
When Mozilla went public, the first item on their list of design principles was: External development counts more than convenience or ease-of-habit for internal-to-Netscape developers. The Netscape X-heads, for example, have moved all of their mail usage except for I'm-out-sick-today and any truly-proprietary messages to the mozilla.unix newsgroup. Likewise with NGLayout hackers and the mozilla.layout group. So it shall be for all development. http://www.mozilla.org/roadmap/roadmap-26-Oct-1998.html I thought we achieved broad consensus a few weeks ago that this principle should be adopted by OLPC, and it was indeed heartening to see more engagement on the devel@ lists and a shift away from private ad-hoc mailing lists. We created a list of 'truly-proprietary' messages, and occasionally even successfully moved conversations to devel@ when the topic strayed away from the proprietary and confidential on that list. I also thought I was successful in convincing management of the pressing need for a community liason, to help ensure that our openness was persistent, and to take personal responsibility for prodding people to use appropriate public fora. I was away in Europe for almost two weeks, and while I've been gone I'm sad to say it seems OLPC has been backsliding. On the truly proprietary list I have received messages about OFW2 status, even though it was made public at a press-invited event back in May, on our public mailing lists by our CEO himself (http://lists.laptop.org/pipermail/sugar/2008-May/005752.html), and on sites such as OLPCNews. I've also received many many other messages that don't pass any sort of confidentiality bar. Part of the problem, of course, is (as I raised earlier), without a community liason with authority, no one can definitely say what is safe to disclose and what is not, so people are erring on the side of caution and forgetting their prime directive of transparency. Further, many meetings and discussions that used to happen on public IRC channels, so as to better include our many non-local contractors and employees, not to mention interested members of the community, have reverted to face-to-face meetings. Expediency is the rationale given -- which of course is exactly the rationale rejected by the principle as stated above. Often transcription or call-in access is offered as a poor substitute to equal access for the community and external developers. Perhaps transparency is not actually a goal of OLPC. But if it is, OLPC has stopped making progress towards this goal. I am wondering if it is appropriate that I unsubscribe from the truly proprietary group and refuse to take part in face-to-face meetings, to encourage the sort of openness OLPC claims to desire. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: New joyride build 2094
On Tue, Jul 1, 2008 at 2:31 PM, Build Announcer v2 [EMAIL PROTECTED] wrote: http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2094 Changes in build 2094 from build: 2092 Size delta: -0.79M -fonts-thai-ttf 0.4.4-1olpc1.2 Was this intentional? I think Bryan Berry's like to notice this gone... --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: New joyride build 2094
On Tue, Jul 1, 2008 at 10:47 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 2:31 PM, Build Announcer v2 [EMAIL PROTECTED] wrote: http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2094 Changes in build 2094 from build: 2092 Size delta: -0.79M -fonts-thai-ttf 0.4.4-1olpc1.2 Was this intentional? I think Bryan Berry's like to notice this gone... Not intentional, I put it back in public_rpms. Probably we should be pulling thaifonts-scalable from Fedora 9 instead though. The rpm is ~ 1 MB bigger but still. Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: New joyride build 2094
On Tuesday 01 July 2008, C. Scott Ananian wrote: On Tue, Jul 1, 2008 at 2:31 PM, Build Announcer v2 [EMAIL PROTECTED] wrote: http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2094 Changes in build 2094 from build: 2092 Size delta: -0.79M -fonts-thai-ttf 0.4.4-1olpc1.2 Was this intentional? I think Bryan Berry's like to notice this gone... --scott I did not do it. I guess someone cleaned up somebodies public_rpms i was going to ask you if you did it. I did find an upstream for it. we will need to create a new set of rpms. we only ever had a rpm no src.rpm -- Dennis Gilmore signature.asc Description: This is a digitally signed message part. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
re: Release Status Meeting - 8.2.0 - Tomorrow, 2:00 PM EDT, various venues - Notes
Hi All, Marco, Greg, Kim, Joe, Paul, Eben, Chris, Scott, Jim, Denis, Michael and others people met on Tuesday July 1 at 2PM US ET via IRC, phone and in person. Sorry for the long e-mail but it was a very productive meeting and I want to keep everyone in the loop. Agenda is at: http://lists.laptop.org/pipermail/devel/2008-June/015961.html Here are a few basic notes and a list of action items. Notes: Lesson learned: Don't hold a meeting via phone and IRC at the same time. Choose one, preferably IRC if you want the broadest audience. We opened the meeting mostly in person and on the phone then moved to IRC when we realized we couldn't do both simultaneously. The first part talked about What is Release 8.2.0. Worked from the definition at: http://wiki.laptop.org/go/8.2.0 We talked agreed its a time based release and talked about what that means (it goes when it reaches the right quality regardless of what features are in). Further definition at: http://wiki.laptop.org/go/Release_Process_Home#Time-based_Releases We talked about target customers, features and the definition of support. We agreed that the goal is to get it to G1G1 if its available in time. Then we started walking through the Release Contracts page: http://dev.laptop.org/report/18 AKA Target Feature Set Including Status page linked from: http://wiki.laptop.org/go/8.2.0 We agreed that the relationship between the release contract and the roadmap is fuzzy. One is what we wanted originally: http://dev.laptop.org/milestone/8.2.0%20(was%20Update.2) the other is what we plan to do: http://dev.laptop.org/report/18 Its not 100% clear how they map top each other and that needs to be explained better. Focusing on the Target Feature Set Including Status page we covered a definition of what is a release contract. Paraphrasing, its an understanding between Michael and a developer that they can and will deliver a feature in time for the release. Target Feature Set Including Status page (http://dev.laptop.org/report/18) lists everything which has a chance to make the release. We agreed that if you want to get something else on that page you should e-mail Michael the details. We talked about what this page offers that we don't already have. The conclusion was that its a high level view of the main features in the release. Each item on this page should include a list of relevant bug id that give the next level of granularity. Greg asked if this is too much process and no one said it was, definitively. Greg also got agreement that each item on the Target Feature Set Including Status page needs a link to some documentation saying what the item includes. Consumers of this feature description are developers to comment on design, QA to write test cases, and product management to share with the plan with users. It can be a link to an existing web page (e.g. new Sugar UI). Greg noted that we do not have much time to redesign the items listed so design changes on them may be deferred at the discretion of the feature owner. The meeting mostly to IRC at this point and a discussion of where we are in the process and what freeze means. There was reference to the new release process: http://wiki.laptop.org/go/Release_Process_Home and to other process draft pages. Kim tried to drive consensus on how know what in this list will actually make the release: http://dev.laptop.org/report/18 There was discussion of the need to build a release candidate ASAP just to see where we are. and other in draft processes and a picture: http://teach.laptop.org/~mstone/d5.svg Also a lot of discussion of how to pick a candidate build, when to pick one, how to build one and other good engineering details. The result was encapsulated in two action items below. I pasted all the IRC info I captured here: http://wiki.laptop.org/go/8.2.0#Paste_of_part_of_IRC_Status_Meeting_Held_on_July_1 I missed much of it :-( Anyone else who has a log please post it. Aside from the communication challenges. It was a great meeting from my perspective. Thanks everyone! When I say we agreed above it means that the people in the meeting agreed. If you feel otherwise its never too late to speak up. Any feedback, comments, complaints, issues, edits, revisions of the notes or other input gratefully accepted, as always. * Action item - Joe to help create and send out for review the release criteria by July 14. - Kim will list all relevant builds in the 8.2.0 wiki page (http://wiki.laptop.org/go/8.2.0). - Kim will write a definition of support including the meaning of backward compatibility in http://wiki.laptop.org/go/8.2.0 - Greg will improve of workflow of the web pages by July 15 - Kim will check with Peru and Greg will check with Uruguay teams to ensure that they do not plan to upgrade to 8.2.0. Action item due by July 20. - Greg will check Kim's whiteboard for list of future customers and check if they are going to use the
pedal and crank (dev) slowness tonight due to backups 20080701-2008-0702
I am backing up crank and pedal to the new machines owl and swan today. Doing this during business hours was causing slowness in git today so I have desisted until midnight tonight, when I will restart the backups. There may be periodic slowness due to backups and moving things around over the holiday and weekend as well. Henry [EMAIL PROTECTED] ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
New joyride build 2095
http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2095 Changes in build 2095 from build: 2094 Size delta: -0.13M +olpc-hardware-manager 0.4.2-1.fc9 --- Included olpc-hardware-manager version 0.4.2-1.fc9 --- -- This mail was automatically generated See http://dev.laptop.org/~rwh/announcer/joyride-pkgs.html for aggregate logs See http://dev.laptop.org/~rwh/announcer/joyride_vs_update1.html for a comparison ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: First Draft Development Process Proposal
Hi All, Thanks for all the comments on the Development Process. http://wiki.laptop.org/go/Release_Process_Home A few gentle suggestions on managing the input. A - My intention is that this page (http://wiki.laptop.org/go/Release_Process_Home) will be the final page. So please put comments and discussion in the talk section. Feel free to make signed edits to the page if there is consensus. Any typo fixing or additional links and references are also welcome (e.g. does someone have a link and explanation of the OLPC-3 build which they can add to the builds section?). I want to manage comments on the Talk page and on this list if possible. B - The best way to change a section is to offer alternative text and get consensus for it. Write exactly what you think the text should say, post it here and/or on the talk page. Once there are enough +1s we can call it final. A couple of people at 1CC need to sign off eventually but if the community agrees that's pretty certain to be the final word. C - The very best way have your input adopted is to write a section. No takers on the open items yet and there are some major areas ... I should have explained my plan for collecting comment before, sorry. I have no complaints about any of the input so far, keep it coming. Here are my responses to a few of the issues raised: 1 - Translations input GS - I agree we need a better definition of that. I added it to the to do list. 2 - Synching with Fedora schedule GS - No opinion right now. Is there consensus? How long do we need after a fedora release comes out before our release is ready? 3 - Core OS vs Core + Activities GS - My intention was that this doc is for Core OS. I added a to do list item for activities and removed on offending comment. We need a definition of what constitutes the Core OS. I prefer a URL with all relevant SW modules, but whatever developers need is OK. Do we have consensus that this doc is for Core OS only? GS - That said, I think we should keep with current naming convention on Releases used in the field which include activities. The fewer times you change the naming convention the better. Also, I think we should document the naming convention down to the OS + Activities level even if we don't have a process for including activities yet. 4 - Support time frame. GS - I agree that release should be supported until the second subsequent release is out (ala Fedora). Do we have consensus on that? 5 - Code names and community roadmap. GS - I agree with the code name idea and the community roadmap idea. Just type of the text you want on the page including where you want it to go. Post it to the talk section and/or send it to this list, get consensus and its in as far as I'm concerned. 6 - Types of builds, meaning of freezes, definition of what requires a minor release. GS - I agree that those could all be improved. Just type of the text you want on the page including where you want it to go. Post it to the talk section and/or send it to this list, get consensus. Thanks for the review and suggestions. I didn't see anyone commenting on whether this is useful or not. Are there any open source developers reading this who are on the fence about working with OLPC? Does this help explain how we work and does it help motivate you to chip in? Is it useful for the rest of you already working on the project? FYI I have a pre-planned vacation I need to take starting tomorrow. I will be back online Thursday July 10. I will collect all comments and edits then and make another major revision. Thanks, Greg S ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
fonts-thai-ttf has been abandoned!
We added a package named 'fonts-thai-ttf' to our builds a while ago for thai font support. However, no one here now remembers where this font came from, or where the upstream came from. Can someone familiar with thai support help out? Ideally we'd like to confirm the licensing and then grow a maintainer for this package in fedora. Thanks! --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: fonts-thai-ttf has been abandoned!
C. Scott Ananian wrote: We added a package named 'fonts-thai-ttf' to our builds a while ago for thai font support. However, no one here now remembers where this font came from, or where the upstream came from. Can someone familiar with thai support help out? Ideally we'd like to confirm the licensing and then grow a maintainer for this package in fedora. I *think* this was provided by behdad, adding him. Am I wrong to think that thaifonts-scalable should replace it? Marco Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release Status Meeting - 8.2.0 - Tomorrow, 2:00 PM EDT, various venues - Notes
On Tue, Jul 1, 2008 at 5:13 PM, Greg Smith [EMAIL PROTECTED] wrote: We talked about what this page offers that we don't already have. The conclusion was that its a high level view of the main features in the release. Each item on this page should include a list of relevant bug id that give the next level of granularity. Greg asked if this is too much process and no one said it was, definitively. I also expressed concern that this view of the process doesn't include an explicit means for tracking blocking bugs and regressions. I think the idea was generally accepted that a feature-driven list like this is most useful early in the release cycle and at decision times when decisions to cut features have to be made, but that later in the process features are expected to be done and the most important release driver is blocking bugs. (Now switching back to expressing personal opinion:) For the moment, I'm personally concerned with how close are we right now to release which (to me) means, how many blocking bugs and regressions are left in joyride. Taking the extreme view, I don't care how many features are complete in it -- I'm perfectly willing to cut some features if that's the shortest path to fixing a bug and getting most of the features out on time. (Unfortunately, many of our current blocking bugs are caused by big already-landed features in a way that would be more work to back out the feature as it is to simply fix the bug.) Maybe I'm premature in switching to a blocker-oriented view, but I certainly want to ensure that we don't lose sight of the big bugs as we congratulate ourselves on landing or partially-landing features. IMO we made the feature view decisions several weeks ago, when we (among other things) committed to basing 8.2 on F9. Now that we've done so, the blocker view deserves to be foregrounded to drive the bugs out. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release Status Meeting - 8.2.0 - Tomorrow, 2:00 PM EDT, various venues - Notes
On Tue, Jul 1, 2008 at 5:13 PM, Greg Smith [EMAIL PROTECTED] wrote: - Kim will check with Peru and Greg will check with Uruguay teams to ensure that they do not plan to upgrade to 8.2.0. Action item due by July 20. Why don't we want them to use 8.2? I suspect some words were left out, and what you really meant was, their schedules for adopting 8.2 are not pressing? If we don't expect our two largest deployments to adopt our release, why are we making it? Something's not right here. Incidentally, on the blocking bug front, I notice that Uruguay's wireless problems with 703/708 were nowhere to be found on the roadmap for 8.2. This is a blocker to our producing something useful for the kids. --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: fonts-thai-ttf has been abandoned!
On Tue, Jul 1, 2008 at 5:52 PM, Marco Pesenti Gritti [EMAIL PROTECTED] wrote: C. Scott Ananian wrote: We added a package named 'fonts-thai-ttf' to our builds a while ago for thai font support. However, no one here now remembers where this font came from, or where the upstream came from. Can someone familiar with thai support help out? Ideally we'd like to confirm the licensing and then grow a maintainer for this package in fedora. I *think* this was provided by behdad, adding him. Am I wrong to think that thaifonts-scalable should replace it? From http://koji.fedoraproject.org/koji/buildinfo?buildID=34572 it looks like you are right, considering the first changelog entry is from Behdad and explicitly mentions OLPC. But I'd like some confirmation from someone doing work in Thailand, if possible. Is there a test case I can run to find out if Thai support works? --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: fonts-thai-ttf has been abandoned!
On Tue, Jul 1, 2008 at 6:19 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 5:52 PM, Marco Pesenti Gritti [EMAIL PROTECTED] wrote: C. Scott Ananian wrote: We added a package named 'fonts-thai-ttf' to our builds a while ago for thai font support. However, no one here now remembers where this font came from, or where the upstream came from. Can someone familiar with thai support help out? Ideally we'd like to confirm the licensing and then grow a maintainer for this package in fedora. I *think* this was provided by behdad, adding him. Am I wrong to think that thaifonts-scalable should replace it? From http://koji.fedoraproject.org/koji/buildinfo?buildID=34572 it looks like you are right, considering the first changelog entry is from Behdad and explicitly mentions OLPC. But I'd like some confirmation from someone doing work in Thailand, if possible. Is there a test case I can run to find out if Thai support works? While we're at it: why are we including libthai-devel, consisting mostly of a whole bunch of .h files? Is there some need for that I'm missing (and can I test it)? --scott -- ( http://cscott.net/ ) ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Freezing features this week.
ear world, We're FREEZING OUR FEATURE-SET THIS WEEK, which means that features targeted at the 8.2.0 release should be IN A JOYRIDE BUILD BY THIS SUNDAY, JULY 6TH. If the outlines of your feature aren't a) in Joyride by the end of the week, [1] b) documented in a release contract in [2] http://dev.laptop.org/ticket/18 then we'll probably defer you if you ask us to include your work in 8.2.0 at any later date. If your work isn't going to be ready to test by Sunday, then please explain your situation on [EMAIL PROTECTED] Thanks, Michael [1]: Joyride is our development build stream. Please contact us if you have questions about how to use Joyride. [2]: Release contracts are an organizational tool that I'm trying out for this release. See http://wiki.laptop.org/go/User:Mstone/Scheduled_software_release_process#Release_Contracts for my thoughts on what they represent. If you think you need a release contract, best come talk to me. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
New joyride build 2096
http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2096 Changes in build 2096 from build: 2095 Size delta: 2.23M +thaifonts-scalable 0.4.9-3.fc9 -libthai-devel 0.1.9-4.fc9 --- Included thaifonts-scalable version 0.4.9-3.fc9 --- -- This mail was automatically generated See http://dev.laptop.org/~rwh/announcer/joyride-pkgs.html for aggregate logs See http://dev.laptop.org/~rwh/announcer/joyride_vs_update1.html for a comparison ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
custom kernel problems
Hi, In regards to: http://wiki.laptop.org/go/Rebuilding_OLPC_kernel I gave up on the git head kernel as there seems to be a lot of bugs there, and tried to build the origin/stable kernel. I'm running a Fedora 9 (i386) system for doing the build. I get this set of messages and the kernel fails to build: kernel/built-in.o: In function `timespec_add_ns': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' kernel/built-in.o: In function `do_gettimeofday': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/kernel/time/timekeeping.c:121: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/kernel/time/timekeeping.c:122: undefined reference to `__umoddi3' kernel/built-in.o: In function `timespec_add_ns': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:172: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:172: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' make: *** [.tmp_vmlinux1] Error 1 error: Bad exit status from /var/tmp/rpm-tmp.47209 (%build) I enabled various bluetooth modules and a few additional USB modules (for bluetooth, serial and wireless interfaces) using make ARCH=i386 gconfig. Is this a known issue? ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Release 8.2.0 -- pls add critical features
I know that Nepal is a small potato compared to Peru and Uruguay but there are two features that we really need. Furthermore, I think these features would immensely benefit the bigger deployments. 1. Need to be able to launch activities such as Scratch, EToys, Pippy, etc. by clicking on a hyperlink in browse. The activity wouldn't run in the browser. We need this functionality in order to effectively use Moodle in our schools. Here's the ticket I opened on the http://dev.laptop.org/ticket/6958 2. Need to be able to remove activities via the GUI, including installed activities, to make room for new ones. We are working on Offline Moodle activity bundles for courses. Each course will be subdivided into weekly modules. these modules will quickly fill up the ssd. kids will need a way to remove activities themselves to make room for new ones. http://dev.laptop.org/ticket/7071 Our development team in Nepal is moving from developing just activities to developing entire courses that include activities, lesson plans for the teachers, and supplementary materials. This in response to feedback from the teachers at the pilots and department of education. Bryan Berry ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release 8.2.0 -- pls add critical features
On Wed, Jul 2, 2008 at 3:23 AM, Bryan Berry [EMAIL PROTECTED] wrote: 2. Need to be able to remove activities via the GUI, including installed activities, to make room for new ones. We are working on Offline Moodle activity bundles for courses. Each course will be subdivided into weekly modules. these modules will quickly fill up the ssd. kids will need a way to remove activities themselves to make room for new ones. http://dev.laptop.org/ticket/7071 You can delete those from the Journal. Is that insufficient for some reason? Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release 8.2.0 -- pls add critical features
You can delete those from the Journal. Is that insufficient for some reason? yeah, if the activity comes pre-installed you can't remove it. We include a lot of activities in our custom build as I imagine Peru and Uruguay do ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release 8.2.0 -- pls add critical features
On Wed, Jul 2, 2008 at 3:34 AM, Bryan Berry [EMAIL PROTECTED] wrote: You can delete those from the Journal. Is that insufficient for some reason? yeah, if the activity comes pre-installed you can't remove it. We include a lot of activities in our custom build as I imagine Peru and Uruguay do Where/how are these activities installed? I thought customization keys was putting them in ~/Activities like user installed activities. Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release 8.2.0 -- pls add critical features
the activities are in ~/Activities . I spoke w/ Bernie about this and he told me that you couldn't remove activities unless they had been originally installed by the user. I may have misunderstood him and be wasting your time w/ a moot issue. I will test it myself when I get into the office this morning. If I am mistaken I will apologize profusely. for the last rollout to the schools I used the customization key to build a custom image which I copy-nand'ed to the XO's. For the refresh of Sugar I will probably try to use something like Puritan to build a custom image that includes Nepal's E-Paath activities and additional rpms for gnuchess and flash player -Original Message- From: Marco Pesenti Gritti [EMAIL PROTECTED] To: Bryan Berry [EMAIL PROTECTED] Cc: OLPC Developer's List devel@lists.laptop.org, Gregsmitholpc [EMAIL PROTECTED] Subject: Re: Release 8.2.0 -- pls add critical features Date: Wed, 2 Jul 2008 03:36:16 +0200 On Wed, Jul 2, 2008 at 3:34 AM, Bryan Berry [EMAIL PROTECTED] wrote: You can delete those from the Journal. Is that insufficient for some reason? yeah, if the activity comes pre-installed you can't remove it. We include a lot of activities in our custom build as I imagine Peru and Uruguay do Where/how are these activities installed? I thought customization keys was putting them in ~/Activities like user installed activities. Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release 8.2.0 -- pls add critical features
On Wed, Jul 2, 2008 at 3:45 AM, Bryan Berry [EMAIL PROTECTED] wrote: the activities are in ~/Activities . I spoke w/ Bernie about this and he told me that you couldn't remove activities unless they had been originally installed by the user. I may have misunderstood him and be wasting your time w/ a moot issue. I will test it myself when I get into the office this morning. If I am mistaken I will apologize profusely. No, I think you are right actually. Now I see what you mean. Marco ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Deploy] fonts-thai-ttf has been abandoned!
For testing, Scott, we are growing a set of each new keyboard/language laptop that comes out of manufacturing. The 'ultimate' test for fonts, translations, keyboard integration is to load a build on these laptops. I have two of each new SKU and I have tried to label one as 'WP' (write protected for final test), and one is not write-protected to accept earlier builds. I don't want these leaving my office (until they have a more permanent home) since there are only 2 of each (I'm hoping they will mulitply while sitting on the ark). The slightly painful, but do-able test case with any MP laptop requires setting mfg data and re-imaging, and is documented here: http://wiki.laptop.org/go/Tests/Keyboard_mappings Kim On Tue, Jul 1, 2008 at 6:43 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 6:19 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 5:52 PM, Marco Pesenti Gritti [EMAIL PROTECTED] wrote: C. Scott Ananian wrote: We added a package named 'fonts-thai-ttf' to our builds a while ago for thai font support. However, no one here now remembers where this font came from, or where the upstream came from. Can someone familiar with thai support help out? Ideally we'd like to confirm the licensing and then grow a maintainer for this package in fedora. I *think* this was provided by behdad, adding him. Am I wrong to think that thaifonts-scalable should replace it? From http://koji.fedoraproject.org/koji/buildinfo?buildID=34572 it looks like you are right, considering the first changelog entry is from Behdad and explicitly mentions OLPC. But I'd like some confirmation from someone doing work in Thailand, if possible. Is there a test case I can run to find out if Thai support works? While we're at it: why are we including libthai-devel, consisting mostly of a whole bunch of .h files? Is there some need for that I'm missing (and can I test it)? --scott -- ( http://cscott.net/ ) ___ Deployment mailing list [EMAIL PROTECTED] http://lists.laptop.org/listinfo/deployment ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Release Status Meeting - 8.2.0 - Tomorrow, 2:00 PM EDT, various venues - Notes
My thoughts in-line... Kim On Tue, Jul 1, 2008 at 6:16 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Jul 1, 2008 at 5:13 PM, Greg Smith [EMAIL PROTECTED] wrote: - Kim will check with Peru and Greg will check with Uruguay teams to ensure that they do not plan to upgrade to 8.2.0. Action item due by July 20. Why don't we want them to use 8.2? Two reasons for Peru to stay with 8.1: 1 - Peru has 'blessed' their build for the next 75k laptops and we got it into production for them. It is 703+peru activities (which you knew, but may not have thought about the reason we ECO'd it into production was so they don't have to upgrade before giving them out to students). 2 - Peru has created and printed their User Manuals based on the UI of 8.1. We should expect and encourage them to continue on this path. We will support 8.1 until we ship 9.1, which should work for them. That's that part that I will confirm when I visit them. Greg has agreed to check in with Uruguay on where they are in their roll out to teachers and students. I suspect some words were left out, and what you really meant was, their schedules for adopting 8.2 are not pressing? If we don't expect our two largest deployments to adopt our release, why are we making it? KQ - We have many, many more deployments, trials, pilots, possibly G1G1, who will be just getting their laptops when 8.2 is ready or soon there after. This release is for them. Even in the case of G1G1, if those laptops go out with 8.1.1, they can MUCH more easily be upgraded to 8.2 than was possible with earlier releases. Something's not right here. Incidentally, on the blocking bug front, I notice that Uruguay's wireless problems with 703/708 were nowhere to be found on the roadmap for 8.2. This is a blocker to our producing something useful for the kids. KQ - Do you have a specific bug in mind? Let's make sure it gets listed when we start listing/triaging blocking bugs (which I agree we can start doing at any time); and make sure it is getting addressed. --scott -- ( http://cscott.net/ ) ___ 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: First Draft Development Process Proposal
On Tue, Jul 1, 2008 at 5:39 PM, Greg Smith [EMAIL PROTECTED] wrote: Hi All, Thanks for all the comments on the Development Process. http://wiki.laptop.org/go/Release_Process_Home A few gentle suggestions on managing the input. A - My intention is that this page (http://wiki.laptop.org/go/Release_Process_Home) will be the final page. So please put comments and discussion in the talk section. Feel free to make signed edits to the page if there is consensus. Any typo fixing or additional links and references are also welcome (e.g. does someone have a link and explanation of the OLPC-3 build which they can add to the builds section?). I want to manage comments on the Talk page and on this list if possible. KQ - feel free to move my comments to the talk page. (If I haven't already gotten to it) B - The best way to change a section is to offer alternative text and get consensus for it. Write exactly what you think the text should say, post it here and/or on the talk page. Once there are enough +1s we can call it final. A couple of people at 1CC need to sign off eventually but if the community agrees that's pretty certain to be the final word. C - The very best way have your input adopted is to write a section. No takers on the open items yet and there are some major areas ... I should have explained my plan for collecting comment before, sorry. I have no complaints about any of the input so far, keep it coming. Here are my responses to a few of the issues raised: 1 - Translations input GS - I agree we need a better definition of that. I added it to the to do list. 2 - Synching with Fedora schedule GS - No opinion right now. Is there consensus? How long do we need after a fedora release comes out before our release is ready? 3 - Core OS vs Core + Activities GS - My intention was that this doc is for Core OS. I added a to do list item for activities and removed on offending comment. We need a definition of what constitutes the Core OS. I prefer a URL with all relevant SW modules, but whatever developers need is OK. Do we have consensus that this doc is for Core OS only? KQ - I think a 'release' consists of everything needed to put it behind us: core OS, signed core OS with all the parts needed for all the upgrade capabilities (fs.zip, .crc, .img, .md5, .usb?,...); images+activities for all customizations (G1G1, Peru, possibly AL); documentation GS - That said, I think we should keep with current naming convention on Releases used in the field which include activities. The fewer times you change the naming convention the better. Also, I think we should document the naming convention down to the OS + Activities level even if we don't have a process for including activities yet. 4 - Support time frame. GS - I agree that release should be supported until the second subsequent release is out (ala Fedora). Do we have consensus on that? KQ +1 5 - Code names and community roadmap. GS - I agree with the code name idea and the community roadmap idea. Just type of the text you want on the page including where you want it to go. Post it to the talk section and/or send it to this list, get consensus and its in as far as I'm concerned. 6 - Types of builds, meaning of freezes, definition of what requires a minor release. GS - I agree that those could all be improved. Just type of the text you want on the page including where you want it to go. Post it to the talk section and/or send it to this list, get consensus. Thanks for the review and suggestions. I didn't see anyone commenting on whether this is useful or not. Are there any open source developers reading this who are on the fence about working with OLPC? Does this help explain how we work and does it help motivate you to chip in? Is it useful for the rest of you already working on the project? FYI I have a pre-planned vacation I need to take starting tomorrow. I will be back online Thursday July 10. I will collect all comments and edits then and make another major revision. Thanks, Greg S ___ 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: custom kernel problems
Sorry to reply to my own post... This patch to linux-2.6.22/arch/i386/Makefile enables the stable kernel to build on F9: - diff --git a/arch/i386/Makefile b/arch/i386/Makefile index bd28f9f..790e378 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -34,7 +34,9 @@ CHECKFLAGS+= -D__i386__ CFLAGS += -pipe -msoft-float -mregparm=3 -freg-struct-return # prevent gcc from keeping the stack 16 byte aligned -CFLAGS += $(call cc-option,-mpreferred-stack-boundary=2) +CFLAGS += $(call cc-option,-mpreferred-stack-boundary=2,) + +CFLAGS += $(call cc-option,-fno-tree-scev-cprop,) # CPU-specific tuning. Anything which can be shared with UML should go here. include $(srctree)/arch/i386/Makefile.cpu --- Based on work around mentioned here: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32044#c28 On Tue, 2008-07-01 at 20:56 -0400, Scott Douglass wrote: Hi, In regards to: http://wiki.laptop.org/go/Rebuilding_OLPC_kernel I gave up on the git head kernel as there seems to be a lot of bugs there, and tried to build the origin/stable kernel. I'm running a Fedora 9 (i386) system for doing the build. I get this set of messages and the kernel fails to build: kernel/built-in.o: In function `timespec_add_ns': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' kernel/built-in.o: In function `do_gettimeofday': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/kernel/time/timekeeping.c:121: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/kernel/time/timekeeping.c:122: undefined reference to `__umoddi3' kernel/built-in.o: In function `timespec_add_ns': /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:172: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:172: undefined reference to `__udivdi3' /archive/fedora/olpc/git/olpc-2.6-rpm/BUILD/kernel-2.6.22/linux-2.6.22.i586/include/linux/time.h:177: undefined reference to `__umoddi3' make: *** [.tmp_vmlinux1] Error 1 error: Bad exit status from /var/tmp/rpm-tmp.47209 (%build) I enabled various bluetooth modules and a few additional USB modules (for bluetooth, serial and wireless interfaces) using make ARCH=i386 gconfig. Is this a known issue? ___ 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
New joyride build 2097
http://xs-dev.laptop.org/~cscott/olpc/streams/joyride/build2097 Changes in build 2097 from build: 2096 Size delta: -0.13M -kernel 2.6.25-20080630.1.olpc.4ae580e3a9597a7 +kernel 2.6.25-20080701.1.olpc.5001ddd18d37eee -cdparanoia-libs alpha9.8-30 +cdparanoia-libs 10.0-2.fc9 -glib2 2.16.3-5.fc9 +glib2 2.16.3-7.fc9 -libXfont 1.3.1-4.fc9 +libXfont 1.3.2-1.fc9 -mesa-libGL 7.1-0.31.fc9 +mesa-libGL 7.1-0.35.fc9 -mesa-libGLU 7.1-0.31.fc9 +mesa-libGLU 7.1-0.35.fc9 -xorg-x11-server-utils 7.3-3.fc9 +xorg-x11-server-utils 7.4-1.fc9 --- Changes for glib2 2.16.3-7.fc9 from 2.16.3-5.fc9 --- + Fix a directory ownership oversight (#449885) + Fix a race in class initialization that causes assertion --- Changes for libXfont 1.3.2-1.fc9 from 1.3.1-4.fc9 --- + libXfont 1.3.2 --- Changes for xorg-x11-server-utils 7.4-1.fc9 from 7.3-3.fc9 --- + sessreg-1.0.4 + xhost-1.0.2 + xrandr-1.2.3 + xrdb-1.0.5 + xset-1.0.4 + xsetpointer-1.0.1 -- This mail was automatically generated See http://dev.laptop.org/~rwh/announcer/joyride-pkgs.html for aggregate logs See http://dev.laptop.org/~rwh/announcer/joyride_vs_update1.html for a comparison ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
How do we manage translation effort in Release process/roadmap?
Hi, all. I have read though Greg's release process draft of OLPC (http://wiki.laptop.org/go/Release_Process_Home) and ReleaseTeam/Roadmap of SugarLabs (http://wiki.sugarlabs.org/go/ReleaseTeam/Roadmap). But both draft documents haven't explained translation of software (including activity) and others. Until midst of update.1 development, development of activities and translation had been aligned to the road map of XO software. it was straightforward; we were notified when window for translation of whole project was opened/closed. Now our collaboration has become complex, because of SugarLabs's split. Translators are still working with one unified portal (i.e Pootle), but I can't understand how and when each PO will be pulled to build. Without those knowledge it would be difficult for translation community to manage their schedule. Could you please explain about this? For instance, scheduled build of Terminal activity with pulling newer translation was announced recently. (http://lists.laptop.org/pipermail/localization/2008-June/001138.html) So we could easily manage the effort. But could we expect similar announcement for every activities, or will the window for translation of activity aligned to development road map of sugarlab or OLPC? Maybe I missed important thing, though... Thanks in advance /Korakurider ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Inappropriate use of private meetings lists.
Scott, I think we all agree that communications can improve and constructive ideas on how to do that are always welcome. I'm not sure how you decided that we had consensus on following Mozilla's design principles. I don't remember being part of that discussion. I'm not sure how we define consensus, which brings me to one of the problems that consensus-driven decision making often faces -- how do we know when we've reached consensus? What strikes me as more fundamental or underlying in your comments is the disconnect between how some people think or want OLPC to be managed versus how we are actually managing and making day to day decisions. There are top-down decisions being made by a few people that drive the direction of OLPC. These decisions are not waiting for consensus, and they are made by a small number of people. I don't believe this is going to change (at least not in the short term). At the same time, there are many decisions that are driven by the community that come from the bottom up. This seems to work pretty well to involve the community in many areas of OLPC operations. I believe this 'business model' is intentional and that OLPC is not trying to be an organization run by consensus. The interesting discussions come from the areas where the top down meets the bottom up. We have a lot of decisions and discussions that need to happen in this middle ground. I would argue that this is where we are making our efficiency over consensus trade offs. Sometimes efficiency wins, and sometimes consensus wins. You (and many others) are helping to identify once we've made an efficiency trade off if there are better ways to communicate and how to make the information public. This is very helpful. There is an analogy here with pushing code upstream. It is often a good idea, but there are many reasons why every patch does NOT go upstream (you've argued quite a few yourself). Let's keep having the discussions, but recognize that these two decision making models exist at OLPC. Those who are employed by OLPC and need to carry out OLPC goals (sometimes in conflict with community goals) are asked to help make decisions in this middle ground. Kim On Tue, Jul 1, 2008 at 2:54 PM, C. Scott Ananian [EMAIL PROTECTED] wrote: When Mozilla went public, the first item on their list of design principles was: External development counts more than convenience or ease-of-habit for internal-to-Netscape developers. The Netscape X-heads, for example, have moved all of their mail usage except for I'm-out-sick-today and any truly-proprietary messages to the mozilla.unix newsgroup. Likewise with NGLayout hackers and the mozilla.layout group. So it shall be for all development. http://www.mozilla.org/roadmap/roadmap-26-Oct-1998.html I thought we achieved broad consensus a few weeks ago that this principle should be adopted by OLPC, and it was indeed heartening to see more engagement on the devel@ lists and a shift away from private ad-hoc mailing lists. We created a list of 'truly-proprietary' messages, and occasionally even successfully moved conversations to devel@ when the topic strayed away from the proprietary and confidential on that list. I also thought I was successful in convincing management of the pressing need for a community liason, to help ensure that our openness was persistent, and to take personal responsibility for prodding people to use appropriate public fora. I was away in Europe for almost two weeks, and while I've been gone I'm sad to say it seems OLPC has been backsliding. On the truly proprietary list I have received messages about OFW2 status, even though it was made public at a press-invited event back in May, on our public mailing lists by our CEO himself (http://lists.laptop.org/pipermail/sugar/2008-May/005752.html), and on sites such as OLPCNews. I've also received many many other messages that don't pass any sort of confidentiality bar. Part of the problem, of course, is (as I raised earlier), without a community liason with authority, no one can definitely say what is safe to disclose and what is not, so people are erring on the side of caution and forgetting their prime directive of transparency. Further, many meetings and discussions that used to happen on public IRC channels, so as to better include our many non-local contractors and employees, not to mention interested members of the community, have reverted to face-to-face meetings. Expediency is the rationale given -- which of course is exactly the rationale rejected by the principle as stated above. Often transcription or call-in access is offered as a poor substitute to equal access for the community and external developers. Perhaps transparency is not actually a goal of OLPC. But if it is, OLPC has stopped making progress towards this goal. I am wondering if it is appropriate that I unsubscribe from the truly proprietary group and refuse to take part in face-to-face
Re: [Server-devel] Wikiserver on XS
good thing there are brighter bulbs involved in olpc than myself :) cbj I will have to look at your code for wikislice and see how it harvests the images from wikipedia. I was having the most trouble w/ that part. Perhaps wikislice can easily be reused for wiktionary we like wiktionary because it is quite small and it has audio samples that say the word. Nepali kids seem really like the latter feature. -Original Message- From: Chris Ball [EMAIL PROTECTED] To: Bryan Berry [EMAIL PROTECTED] Cc: server-devel@lists.laptop.org Subject: Re: [Server-devel] Wikiserver on XS Date: Mon, 30 Jun 2008 12:06:51 -0400 Hi, I tried setting up a Wiktionary server on a shared server that our pilots have access to and it was a major pain in the ass. After 3 days I couldn't get it to work. I had a lot of trouble downloading and importing the images, perhaps Chris Ball's wikislices gets this right. I will have to look at the code and ask him. Oh, you mean using the wikislice technology to build a snapshot of the dictionary data from wiktionary, rather than the wikipedia data? That's an interesting idea. The wikislices are alright but they don't allow for searching for content which I think is an essential feature. Yes, they do; when you run the Wikipedia activity, there's a search toolbar that searches against the local index. (You might be confusing the Wikipedia activity with some hand-made Wikislice HTML content bundles that SJ made quite a while ago.) - Chris. ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
[Server-devel] VPN server for our project
Hi, OLPC France is plannig a project Windrose (http://wiki.laptop.org/go/Projects/WindroseOLPC ) consisting for the schools among the world participating in the project to share and publish their contents (text, images, videos, etc.) in a semi-private BLOG (likely EduBlogger). My question: has anyone experimented the fonction of a VPN network involving remote XS servers ? Our concern is to protect the privacy of the kids, allowing private areas and public areas and making the login process easier for the kids/teachers. The communication/authentification to the VNP server will possibly be organised via a local XS server. Thank you for your comments Best regards Samy ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] Want to package PostgreSQL for OLPC
On Wed, 2008-06-25 at 08:48 -0400, Martin Langhoff wrote: How hard is it to get a backport of Pg8.2/8.3, plus some key dependencies (php-pgsql, python-pgsql, pam-pgsql) recompiled to use the new libpq, all on F7? http://yum.pgsqlrpms.org ;) I have already built all PG combinations against Fedora 7-8-9 and RHEL 4,5. We have compat packages, and all pam-pgsql, etc in that repository. So, using those files won't be a problem. Also - I see you work at CommandPrompt -- so I'll throw a wishlist item I have on my list in your direction. Perhaps you, or someone at CommandPrompt has something similar. With Pg packaged, what we will need to come up with is ~3 sets of config files tuned for different memory footprints. The same XS image will be used in hosts with various memory configurations - 256MB RAM on XO hardware, 1GB on the recommended config, and high-end hosts may have more RAM. Sure, it is doable -- I can do it for you. Pg cannot take all of that memory, but perhaps 15%-20% is a reasonable footprint. The workload for Pg is mainly Moodle and MediaWiki. Ok. One of the key things for the XS is that we should not OOM, and our working set _must not_ end up in swap. And we have several services we have to run, so it's a bit of a tough diet on RAM usage. We gotta sweat every MB :-) You are right. So, please let me know when you need the config files, and also how can I submit those packages to OLPC repository. Regards, -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/ signature.asc Description: This is a digitally signed message part ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] Want to package PostgreSQL for OLPC
On Wed, 2008-06-25 at 09:44 -0400, Martin Langhoff wrote: We have compat packages, and all pam-pgsql, etc in that repository. So, using those files won't be a problem. Fantastic! So python-pgsql and php-pgsql are there too? python-psycopg2 is there, but not php-pgsql. The question is: Is php-pgsql already in OLPC package set? If yes, we have compat packages to satisfy dependencies. If not, I can give it a shot. ... So, please let me know when you need the config files, and also how can I submit those packages to OLPC repository. When? Yesterday ;-) - but perhaps there is no need for rebuilding the packages. What I am wondering about is what the best solution is to maintain those alternative configurations long-term. Perhaps we could have a postgresql-server-altinit package that provides an alternative init script + config files and disables the init script from postgresql-server. The alternative init check memory size, and starts with the appropriate config files. Maybe we can solve this issue during initdb process, and copy the preconfigured config file based on the memory to data directory after we run initdb. It may be easier to maintain... BTW... I must admit that I did not work for OLPC project before, so I don't know how to do things. Regards, -- Devrim GÜNDÜZ , RHCE PostgreSQL Replication, Consulting, Custom Development, 24x7 support Managed Services, Shared and Dedicated Hosting Co-Authors: plPHP, ODBCng - http://www.commandprompt.com/ signature.asc Description: This is a digitally signed message part ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] Laptop XS working
Martin/list To remind you, I am getting to know the XS using an installation on a laptop, with 2 XO-B4s and 1 XO-1, all running build 703 and G1G1 activity pack. I have followed the configuration instructions for a small school server, ejabberd starts on boot etc. I have tested the server with and without Internet access via Ethernet. I am using a prototype active antenna that obviously is working fine. I had initial success, finding that all three XOs immediately could access the server and Internet through the browser, and they could see each other in the neighbourhood, share and invite. I then registered all three, this was uneventful. The register option no longer appears on the XOs. However, the result (unless this is a red herring) is that they can all still access the server and Internet but do not see each other any more in the neighbourhood and cannot collaborate. Can you point to any lines of attack to diagnose the problem. I can see the school mesh and simple (olpc) mesh with my Windows laptop. David Leeming Technical Advisor, People First Network Tel: +677 76396(m) 24419(h) 26358 (w) www.leeming-consulting.com -Original Message- From: Martin Langhoff [mailto:[EMAIL PROTECTED] Sent: Tuesday, 1 July 2008 8:40 a.m. To: David Leeming Cc: Pia Waugh; Phill Hardstaff; Jeff Waugh; Ian Thomson; Barry Vercoe; Michael Hutak; OLPC Australia Board; XS Devel Subject: Re: Laptop XS working On Mon, Jun 30, 2008 at 4:14 PM, David Leeming [EMAIL PROTECTED] wrote: I got them collaborating (at the same time accessing Internet - so via the server) without even registering. How could that be? But I will do so. XOs are smart enough to collaborate even without the XS, and in this case they are using the AP but not using the ejabberd-provided services for collaboration. The main difference is that it scales much *much* better once they've registered and are using it. Can you explain the importance of the domain name. I just called mine oceania.org. Do we need to register domain names for external services? None whatsoever. I've copied this reply to [EMAIL PROTECTED] - let's have technical questions and answers archived *there* -- it is pretty hard for me to scale if the technical questions are asked in private. Doing it in the public list means they are archived and searchable :-) 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 ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] Laptop XS working
OK, done all that, I was fooled for a while, when logging onto ejabberd via the web interface (on a connected XO) I did not realise you need to enter the full username, i.e. [EMAIL PROTECTED] I bet that will fool a few other novices like me :) Two of the 3 laptops are now OK, but the third does not find the school mesh and reverts to the simple mesh. This happened after I registered it, it then locked up and had to be rebooted.. last question is, how does one remove the registered laptop, reset the XO (I can always reinstall it) and re-register it? David Leeming -Original Message- From: Martin Langhoff [mailto:[EMAIL PROTECTED] Sent: Wednesday, 2 July 2008 9:38 a.m. To: David Leeming Cc: Pia Waugh; Phill Hardstaff; Jeff Waugh; Ian Thomson; Barry Vercoe; Michael Hutak; OLPC Australia Board; XS Devel Subject: Re: Laptop XS working On Tue, Jul 1, 2008 at 6:07 PM, David Leeming [EMAIL PROTECTED] wrote: I then registered all three, this was uneventful. The register option no longer appears on the XOs. However, the result (unless this is a red herring) is that they can all still access the server and Internet but do not see each other any more in the neighbourhood and cannot collaborate. Can you point to any lines of attack to diagnose the problem. The output of the olpc-netstatus script is a key debugging tool. What do they say? A couple of questions that might help the matter - Did you restart the laptops after registration? - Did you configure ejabberd appropiately? Look in the configuration guide thesteps to setup the Online roster. 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 ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
Re: [Server-devel] Laptop XS working
John, Thanks, I have set it up now, including the roster, and it seems fine. Thanks all, David Leeming Technical Advisor, People First Network Tel: +677 76396(m) 24419(h) 26358 (w) www.leeming-consulting.com -Original Message- From: John Watlington [mailto:[EMAIL PROTECTED] Sent: Wednesday, 2 July 2008 3:10 p.m. To: David Leeming Cc: John Watlington; OLPC Australia Board; Pia Waugh; Barry Vercoe; Martin Langhoff; Michael Hutak; Phill Hardstaff; Ian Thomson; XS Devel; Jeff Waugh Subject: Re: [Server-devel] Laptop XS working On Jul 1, 2008, at 6:38 PM, Martin Langhoff wrote: On Tue, Jul 1, 2008 at 6:07 PM, David Leeming [EMAIL PROTECTED] wrote: I then registered all three, this was uneventful. The register option no longer appears on the XOs. However, the result (unless this is a red herring) is that they can all still access the server and Internet but do not see each other any more in the neighbourhood and cannot collaborate. Can you point to any lines of attack to diagnose the problem. The output of the olpc-netstatus script is a key debugging tool. What do they say? A couple of questions that might help the matter - Did you restart the laptops after registration? - Did you configure ejabberd appropiately? Look in the configuration guide thesteps to setup the Online roster. Symptom: olpc-netstatus shows all laptops running gabble to a server, the ejabberd admin page shows the laptop online, but the laptops can't see one another. This sounds like the online roster isn't setup properly. Another possibility is that there needs to be a DNS entry for conference.schoolserver resolvable on the school server. (An obscure riff on this is that ejabberd uses erlang name resolution, which tends to use alternate nameserver entries more aggressively than libc. Make sure that /etc/resolv.conf only points to the local nameserver.) Can we get more details about the laptop XS ? Is it on an XO wad ___ Server-devel mailing list Server-devel@lists.laptop.org http://lists.laptop.org/listinfo/server-devel
