Re: Translation refresh
FYI: The tracking of activities, and marking which ones were in the best state to ship, was part of the build debate earlier this year. The proposal was for OLPC F. to mark some activities as mature enough to consider for deployment and to push them to have a consistent branch name for each time based deployment. OLPC F. couldn't hit the minimum buy-in (put a year number in the versioning) so I dropped all the build fixes. Try again next year. http://wiki.laptop.org/go/2008_Debate_of_Build_and_Release -- Charles On Tue, Apr 22, 2008 at 4:58 AM, Bernie Innocenti [EMAIL PROTECTED] wrote: Kim Quirk wrote: Collaboration is really important to any release... so we need to include some activities that collaborate as part of formal testing. Similarly, Journal is much more than just an activity... so that will have to be part of systematic testing. Browse has to work as it is our connection to the outside world and to our local or school library. So that will have be part of any good test plan. What version(s) of Browse and other important activities are we going to test each OS release with? Here's an example: I installed the G1G1 activity pack some time ago, and I don't even know what versions of activities I'm using. Will this random bunch of activities keep working when Update.2 comes out? Vice-versa, can we expect activities released next year to work with my build 703 system or will I be forced to upgrade at some point? The complexity of an N-to-M compatibility testing is the reason why Linux distributors tend to bundle all the existing applications with the OS (either on installation media or on online repositories). Not addressing these dependencies now will lead to the same compatibility hell that has swamped a well known desktop OS. -- \___/ |___|Bernie Innocenti - http://www.codewiz.org/ \___\ CTO OLPC Europe - http://www.laptop.org/ ___ 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: Translation refresh
Just so I can get it into the wiki correctly, the primary release steps you mention: 1. Check-in: Making sure all the relevant changes are checked into GIT on proper branches, and that the result is only completed functionality that seems to work together. 2. Testing: Building it, getting different people to run smoke tests, running some (t.b.d.) more formal QA, iterating as tickets are filed and closed. 3. Sign the build: in a dark room with secret commands, the build is signed. 4. Transfer the build image to manufacturing, make sure they pick it up, make sure it gets into the production queue for new units. Am I missing anything? Charles Merriam From check-ins, to packaging, to build creation, to testing, to the signing of the build, to getting a build properly into manufacturing... there are many steps and each step has many points of failure. Some of these we can eliminate over time through automation... but we aren't there yet. Kim On Sun, Apr 20, 2008 at 3:24 PM, Bernie Innocenti [EMAIL PROTECTED] wrote: On Sun, 2008-04-20 at 18:54 +0530, Sayamindu Dasgupta wrote: I don't think it makes sense to make seperate releases _only_ for translations. Why does rolling a release seem to be such a big thing? Generating a new OS image takes 10 minutes of machine time and this is as simple as it can get. All the other steps our release procedures that create overhead assume some amount of testing is necessary before a new OS hits users. But if the only thing you change is translations, it doesn't matter whether you're doing it with a new OS image or through a separate language pack. The resulting system will in both cases be the old one plus these new strings. And this is what you have to retest in both cases. What we need is a fastpath in our procedures for this case. I think we had something adequate for security updates. Michael? sidenote Our friends here told me that we must urgently translate the word Pippy because apparently it has a very inappropriate meaning in Turkish :-) /sidenote I am currently working on a language-pack builder for deployers and testers, which would generate language packs for different releases (eg: Update-1, or Joyride), etc. This should separate the release process substantially from the translations, and deployers can add enhanced language packs for the deployed systems as the translations evolve. This would add yet another degree of implicit dependencies to our system. The way I see it is that we already have a very dangerous situation where Sugar and the activities can freely vary with respect to each other with no robust dependency tracking. If we also add translations to the equation, we're making it much worse. Then you get bug reports such as I don't get a string translated to Turkish, you'd have to ask the user: - what OS release? - what activity version? - what language pack? Unless we plan to switch to a true package manager, we can't modularize things too much. However, to make this work we also need to follow some kind of branching policy wrt the releases (eg: once Update-1 is released, bugfixes targetted for subsequent minor releases f'd uor Update-1 should be committed to the Update-1 branch only, while development for Update-2 should continue in the devel branch). This has to be done for _all_ activities (and of course, the components of Sugar as well). Yes, this is what is being done already for Sugar and many other packages hosted on dev.laptop.org (although there's no policy that mandates it). -- \___/ |___| Bernie Innocenti - http://www.codewiz.org/ \___\ CTO OLPC Europe - http://www.laptop.org/ ___ 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 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Jobs (was Re: Chilling Effects paper at USENIX UPSEC)
Weren't you just posting bitter rantings how OLPC was all lost yesterday? On Sat, Apr 12, 2008 at 1:20 AM, Edward Cherlin [EMAIL PROTECTED] wrote: On Fri, Apr 11, 2008 at 6:16 AM, C. Scott Ananian [EMAIL PROTECTED] wrote: ...OLPC is hiring again, which means that hopefully soon we will only be underappreciated, not quite so much overworked. We're more than doubling our devel team, hiring QA folk (finally!), and I'm excited. If y'all have high quality candidates, send them our way! --scott -- ( http://cscott.net/ ) I see five jobs listed at http://laptop.org/en/jobs.shtml. It sounds like you have heard of others. Any chance of a Doc Lead to organize hardware and software manuals, training materials, and textbooks? or some paid Volunteer Coordinators? -- Edward Cherlin End Poverty at a Profit by teaching children business http://www.EarthTreasury.org/ The best way to predict the future is to invent it.--Alan Kay ___ 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: Jobs (was Re: Chilling Effects paper at USENIX UPSEC)
Sorry, that was meant to be a reply not reply to all. mea culpa On Sat, Apr 12, 2008 at 6:47 AM, Charles Merriam [EMAIL PROTECTED] wrote: Weren't you just posting bitter rantings how OLPC was all lost yesterday? On Sat, Apr 12, 2008 at 1:20 AM, Edward Cherlin [EMAIL PROTECTED] wrote: On Fri, Apr 11, 2008 at 6:16 AM, C. Scott Ananian [EMAIL PROTECTED] wrote: ...OLPC is hiring again, which means that hopefully soon we will only be underappreciated, not quite so much overworked. We're more than doubling our devel team, hiring QA folk (finally!), and I'm excited. If y'all have high quality candidates, send them our way! --scott -- ( http://cscott.net/ ) I see five jobs listed at http://laptop.org/en/jobs.shtml. It sounds like you have heard of others. Any chance of a Doc Lead to organize hardware and software manuals, training materials, and textbooks? or some paid Volunteer Coordinators? -- Edward Cherlin End Poverty at a Profit by teaching children business http://www.EarthTreasury.org/ The best way to predict the future is to invent it.--Alan Kay ___ 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: how to let activities write to file without risking security
FYI, this might crop up. http://wiki.laptop.org/go/Talk:Bitfrost#org.freedesktop.DBus.Error.AccessDenied Basically, don't choose the solution of making both users have the same uid. -- Charles Merriam On Thu, Apr 10, 2008 at 12:24 AM, John Gilmore [EMAIL PROTECTED] wrote: At OLE Nepal we need to let our etoys image allow writing to disk, however under rainbow the image is executed under another user id. What's the way to give an/our activity permission to write to certain directories without just making them world writable, which is surely not the way to go. Make them world writeable. I don't know why the Nepal team wants to insist on ultimate super high security all the time. Security is not there to make your life miserable. In many cases it isn't there for any reason at all; somebody did it for their own situation, which doesn't match your situation. If it gets in the way, turn it off! That's why there is a switch. John ___ 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: Build Debate: Followup on Build Naming
Thanks for formalising this, I would also strongly suggest that the organisation is moved to the far right, and that we get rid of year. component major minor bugfix organisation I strongly suggest we keep the year. Yes, really, OLPC should release new software at least once per year. It should dump support for software two or more years old. It should release based on time, not feature. Also, why add a minor-minor (bugfix) number? Charles ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Build Debate: Followup on Build Naming
Do you expect to make a major change to the API more than once per year? Would you like major changes to the server API to release contemporaneously with other components? Do you want subtle, minor changes to the API made over a year ago to be the cause of difficult to diagnose problems? Do you want both you and customers to have a context in which only one year of development need be considered? How bad is it if all minor bug-fixes and minor API changes are given a new major version once per year? - ask interesting questions ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: XENified images for XO
There's been some talk about building for multiple platforms: Aside from the XO-1 hardware, various other builds with advocates include Linux builds: Ubuntu (widely used for Actitivies development), Fedora 7 jh-build variant (widely used for OS and systems development), Gentoo, Cebian, and other Linuv variants, advocated by adherents to those operating systems. and also virtualization builds: VMware - which has free (gratis) server software for a variety of platforms. VMware players have had commercial success and the players tend to be stable. Q/Emu - An open source emulator. Used a lot. I suppose XEN could be added to the list. There is a debate going on about the build and release cycle: http://wiki.laptop.org/go/2008_Debate_of_Build_and_Release#Multiple_Targets.2FTimely_Builds It's currently bogged down in the minimum buy-in for time based releases (years in the release name). Charles On Thu, Apr 10, 2008 at 12:16 PM, Marcus Leech [EMAIL PROTECTED] wrote: Has anyone done any work on building XENified images for XO? I'm interested in this for building a large-scale virtualized XO environment for testing purposes. The other option is to run the XO image in HVM mode, but that limits which processors I can use to host such a thing. Cheers -- Marcus LeechMail: Dept 1A12, M/S: 04352P16 Security Standards AdvisorPhone: (ESN) 393-9145 +1 613 763 9145 Strategic Standards Nortel Networks [EMAIL PROTECTED] ___ 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: Cutting a slice of wikipedia - CDPedia
Ah, the old days were cutting out the images and putting the whole thing on one's cell phone. Seriously, one might consider: http://simple.wikipedia.org/wiki/Main_Page Simple is about 10,000 articles written in simple English, aimed at children trying to learn English. FYI, Charles. On Tue, Apr 8, 2008 at 5:21 PM, Martin Langhoff [EMAIL PROTECTED] wrote: Yesterday we had a mini-sprint with argentinian pythonistas and we discussed Alecu's CDPedia which is a Python toolchain that does are good job of cutting a slice of wikipedia and cutting off the least interesting parts to make it fit. His project is here http://code.google.com/p/cdpedia/ and it would be great if Alecu could explain a bit more what it does -- I am sure I didn't do it any justice above ;-) So - Alecu, meet the list, list, say hi to Alecu ;-) I would love to see this progress -- we definitely need something like this to assist the localization teams to build a good content package for the XS. Are there related projects? I thought I had seen one but I cannot find anything now, so it was perhaps discussion about desired functionality? cheers, martin -- [EMAIL PROTECTED] [EMAIL PROTECTED] -- School Server Architect - ask interesting questions - don't get distracted with shiny stuff - working code first - http://wiki.laptop.org/go/User:Martinlanghoff ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: RPMs and Activity bundles
Hmm.. More details? ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Chilling Effects paper at USENIX
I'm a bit slow, being a bugbear of very little brain. I read the paper, and it seems to summarize as: 1. The BitFrost Specification is documentation, not detailed implementation. The author does not read code. 2. BitFrost does not promise anonymity. 3. BitFrost does not cover how to secure the Country Key Store. 4. If used as a specification, and all packets are signed and the Country's Key Store is compromised, then bad things can happen. It seems like OLPC F. should issue an immediate (preemptive) response saying: 1. BitFrost is an open-source implementation. The BitFrost Specification is a high level document and not an engineering specification. Engineers can read the implementation source code. 2. BitFrost does not promise anonymity to school children. [If factcheck says HTTP packets are not generally signed then add] However, it does not enable the pervasive montoring the author suggests. 3. BitFrost does not specify general security measures for the country wide servers. 4. It is unfortunate that a respected conference did not do a better job at vetting this paper. Below is my blow-by-blow. If no one else writes a Wiki page on it by next week, I may do it. Charles Merriam. Concerns seem to be: 2.2 - BitFrost has poor documentation and is not on standards track. Could someone let me know if *all* the BitFrost implementation is opensource? 2.3 - ECC Keypair does not specify keysize Anyone shed light on this? 2.3 - Long lived photograph/name/laptop pairing is made. Um, yes. Author questions, but does not support reasoning for question, this linkage. Also, is this Photograph transmitted as the P in her tuple? Or is P a crypto P? If the photo is not transmitted, then her assertion of being linkable falls down. I hate it reviews let an article publish without checking all the terms. The author incorrectly lumps this under Compromising Privacy. The Compromising Privacy under Bitfrost 7.2, 8.16, 9.2 addresses stealing documents from a user; anonymity is not part of the BitFrost specification or goals. The author also starts a poor researcher's tool here: It's not said why this happens, but if it is because of X then it is wrong. 2.4 - Keys/User This appears to summarize as BitFrost doesn't tell you how to protect your country's key store. 2.5 Bitfrost does not specify anonymous communication. If done like X, you can't get anonymous communication. 2.6 Is it true that calling home for an XO does not include the local School Server? If it does include the local School Server, the author's assertion of remote villages bricking until Internet Access is restored is incorrect. Also points out that an authority could turn off a child's laptop at will. (part of the spec.) 2.7 Spec doesn't cover some bios implementation details. 3.1 The lack of anonymity makes this a bad tool for overthrowing corrupt regimes. 3.2 If author is correct about how packets are signed and an oppressive government monitors all traffic and overtly punishes children for saying anti-government things online, then it could hurt the child's esteeem. Again, would someone in the code answer if all HTTP packets are signed? 3.3 If government monitors all communication, children may be surprised that things said within their school are monitored. 4.0 Conclusions Finds BitFrost doesn't support anonymity, and believes it to be in the spec. Brings up spec addresses user space programs, not the implementing operating system. Footnotes, etc: Didn't check to see if shipping version have a led on the camera. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Thread Summary. to date.
Can't tell your players without a program Micheal stone: no problem Andres Salomon: hmm. Apple Blueberry (named alphabetical) Gary Martin: No, official-703.. No to OLPC2 thats hardware Dennis Gilmore: OLPC2. Oh, an the next hardware is XO-2 and should have same releases. Simon: 802.month or 802.season to push exact time. OLPC-2 type naming for feature based. Morgan: use internal names without exact ship times in case we missed. Arron Konstom: outward consistency counts. No update-1-703, even if we did it before Walter Bender: Seasons are out. Feature based naming will slip. XO-2 is hardware. OLPC-2, er Sugar-2, is software. Or, OLPC-Fedora 1, or..er, names are hard. Well, ship based on time. Paul Fox: OLPC doesn't sound like software. Start with high numbers. Tomeu Vizoso: Sugar sounds like software. Kent Loobey: Schools really want predictable dates. Let's use solstices which aren't. Polychronis Ypodimatopoulos: Prefixes shouldn't get to long. Richard Smith: How about feature based? hardware version.major software.minor software Mitch Bradley: What are we releasing? OLPC component Generation Ordinal Jim Gettys: Note that OS protocol changes may or may not change all Activity binaries. Martin Langoff: Feature based, major software (API).minor software (Stability).bugfix - country, with some interaction with ISO numbers. Let's start with 0. something since the API isn't stable. Mitch Bradley: Feature based with letters, .10 doesn't work too well. Morgan Collect: Right 7.10 is said as 7.1 and 7.04 and 7 -- Charles Merriam ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Build Debate: Followup on Build Naming
Here's may proposal:
OLPC Year components major:minor [- special_build]
OLPC 2008 OS 1:0 - Mexico
OLPC 2009 Activity Bundle 2:14
SPE 2009 Student Bundle 1:0 - Approved by Sec. Mota
OLPC = Built by OLPC. If the Secretariat of Public Education builds a
custom, they name it SPE or anything not OLPC.
Year = The year (). This provides a simple, human readable, first
classification. It does encourage upgrading once a year and lets OLPC
easily drop support for versions two years old.
Components = The components included, e.g., School Server, OS,
Activity Bundle, Great Books, etc.
Major = Version numbers that restart every year. That is, 2008 OS
1.0 and 2009 OS 1.0 are different. As currently stated, OLPC F. is
pushing for two major updates per year 1.x and 2.x. Components
with the same major versions are generally expected to play together.
Minor = Yes, there will be patches and bug fixes. People should
decide if this should start at 0 for each {Component, Major Version}
or just for each {Component}. The latter would mean that one couldn't
tell how many patches were applied to the OS component, but would
know that 2008 OS 1:14 was built after 2008 Activity Bundle 1:13.
I'm in favor of just this latter scheme, because the shorthand 1:14
becomes unambiguous.
Special Build = A special build for a market or reason. So, - ISO
3166 CountryName or - G2G1 Build or whatever. While it may seem
redundant to the minor version, it makes it easy to parse.
People will use shorthands to describe this:
* OLPC 2008 1 means the first (April-ish) release of everything.
* OS 1:15 or 1:15 means the specific version for the current year.
* 2008 - Mexico means the build Mexico choose for the yearly deployment.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel
Re: Latest news from Intel
FYI, HP also announced a lower cost ($500) laptop aimed at classrooms today. http://www.mercurynews.com/ci_8848583 2008/4/8 Prakhar Agarwal [EMAIL PROTECTED]: Sorry for cross posting. Could not resist myself. Please, visit the link below. Some of you might have read it already. There's a substantial mention of OLPC. http://news.bbc.co.uk/2/hi/technology/7334518.stm Regards, -- Prakhar Agarwal Technical Head - Library RD Team 3rd Year B.Tech, IT JIIT University,Noida Life is the greatest teacher ___ 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
2008 Debate of Build and Release
Here's my write-up from the April Fools memo (http://wiki.laptop.org/go/April_Fool_2008_Build_Process) and Mini-conference presentation. I'll get something on the wiki shortly, probably at http://wiki.laptop.org/go/2008_Debate_of_Build_and_Release. I'd appreciate any comments. Thanks, Charles About The Debate The 2008 Debate on Build and Release started when Charles Merriam (l) wrote an April Fool's document named April Fool 2008 Build Process (l). The document is short and should be read before reading this page. The ideas were then presented as a set of slides (l) at the April 4 Spring Miniconferce (l) Much debate ensued. Build and Release processes and tools change over time as software matures. You can find references to the build and release debates of 2007 (L), which more more concerned with getting any process to work. Debating Debating occurs in several venues: the talk page of this article (l); calls and conferences to be determined; and the devel and testing mailing lists (l). Please start *all* debate posts with Build Debate:. If your posts relate to a section, mention the section. For example, a good subject line might be Build Debate: Time Based Release has another problem. Relevant information, conclusions, and unsolved issues will be eventually be added into this article. If you want to track the progress and conclusions, you may want to watch this page (l) on your watchlist (l). Social vs Technical Problems Thinking about build and release management causes arguments everywhere because it dredges up long standing concerns about corporate strategy and implementation. Build and release management is not a solution to social and business problems, no matter how much you wish it were. Nor will build and release management squeeze seven months of engineering into a six month cycle. Be cognizant of discussions that catapult from release management to general management. Also, recognize that the OLPC project is transitioning rapidly from realizing an impossible dream to managing many thousands of laptops in diverse deployments. People need time to adjust expectations for this change. Social Problems To Address Adopt a Time Based Release System * See Ubuntu's Wiki for an explanation (l) and release schedule [l]. Time Based Release is a practice and procdure to ship based on the calendar instead of based on features completed. The release ships on-time and features still stabilizing will slip forward to ship in the following release. This change in philosophy provides a regular heartbeat of new versions, so that slipped features have a reliable next ship date. The ultimate aim is to always be able to build a nearly shipping version, This means an exceptional circumstance requiring a special release would only need Disaster Insurance final quality testing. New features still gaining quality are developed in separate branches until they can be incorporated into a solid branch. This is a contentious discussion and took up most of the MiniConference discussion time: * Every project starts with feature based releases until it hits a minimum functionality. Stability and predictability become issues after there are users relying on the project's functionality. OLPC now has many schoolchildren, teachers, and developers relying on releases. * Delaying releases to serve one market breaks other markets: many bug fixes and improvements will have been tested and ready each release date. Some countries will only pick up one release per year and may not be able to adopt any changes if release dates slip. * Time-based releases also trigger contemplation of the changing management directives for each release. Having specific release dates may make the discussion easier. Specifically, an inflexible ship-date grounds discussions in reality and helps to evaluate trade-offs of important features. The build manager creates a release on a date from whatever engineering is completed. Build Manager Not Necessarily On Site in Cambridge This idea brings more contentious debate and contemplation. The role of the build manager is to make sure releases happen on specific dates and to balance adherence to a release process with the flexibility of exceptions to that process. He or she does not get features by going down the hall to beat up engineers for patches. Also, a growing percentage of development happens in the open source community, which is off-site. Buy-in From OLPC Foundation Before Work Buy-in facilitates work that is used. Work proceeds with only partial commitment: full adoption will take a year of successes to cement. Participants at the Mini Conference believed the minimum buy-in for work to commence was doable. OLPC Foundation must agree to rename Update-1 to something with a year designation. The exact name is flexible, e.g., 2008 Release 1 and XO-08, April Edition are both fine. This commits to releases annually, closer
Re: [Testing] New OLPC Process and Rules for Builing Activities, Releases, and Firmware Builds
April Fool's was a good way to present what does eventually need to be done, and what is done in some other open source projects. It's a good way of doing it, and would require some work. I would set up a set of stages with the highest and easiest pay-offs first, e.g. building versions with and without activities and with and without source would be easy. I don't know how to implement some of the features, especially code coverage metrics. That said, it is likely to remain an April Fool's prank unless OLPC Foundation folk are willing to head towards tested and timely releases where experimentation takes place on branches. This change in mentality is work for some developers and impossible for others. There are some fix-ups on the proposal; the stages of water metaphor needs better explanation and the spring and summer metaphors need renaming (thanks Andrew). Does anyone believe OLPC Foundation is up to this style of change? Charles On Wed, Apr 2, 2008 at 10:29 AM, Grig Gheorghiu [EMAIL PROTECTED] wrote: I'm a bit confused at this point as to whether Charles's message was an April Fool's prank or the real deal. Charles -- you got all of us here, now can you shed some light? :-) Grig --- C. Scott Ananian [EMAIL PROTECTED] wrote: On Tue, Apr 1, 2008 at 6:55 PM, Charles Merriam [EMAIL PROTECTED] wrote: New OLPC Process and Rules for Building Activities, Releases, and Firmware Builds I. Introduction It's an exciting time at the OLPC Foundation! In the next few weeks we will be releasing Update 1 and holding our first Mini-Conference for developers at 1 Cambridge Center. Also, we are announcing our new processes for streamlining the development process. Process and rules make it easier to create quality deployments to the children world wide that now depend on their XOs. We will be releasing high-quality, regularly scheduled deployments timed to coincide with the school year in most countries. These changes will help developers concentrate on high quality software and have their changes make it out to children more quickly. The major changes outlined in this document include: Time-based Release Schedules Developer Changes: Better GIT web interface standard project metrics Useful and predictable build targets II. Time-based Release Schedule OLPC is moving to time-based release schedule. A growing number of open source projects have standardized on this approach including the Ubuntu and Gnome projects. See https://wiki.ubuntu.com/TimeBasedReleases for one explanation of this system. Major updates will be signed and released on May 15 and November 15 each year. This will allow ample time for review, teacher training, modified lesson plans, and deployment. The version numbers will be in the form YY.season, so our next two releases will be 08.Spring near May 15, 2008 and 08.Autumn on November 15, 2008. This year, because of the transition, Update-1 may be released on a different date than May 15, 2008. It will still be officially called the 08.Spring version. Getting a stable build out to all corners of the globe can be hard. A branch grows in stability over time and stablity of the release and field testing the final release candidate takes time. We plan to finalize the exact schedule for 08.Autumn shortly, but expect the following: 45 days until release Feature Freeze 30 days until release User Interface Freeze, Sugar OS freeze, Imports Freeze 15 days until release Translation packages freeze, Final freeze and start final testing 0 days Release on schedule. +30 daysAnnounce schedule, priority, and tool chain changes for next release at developers conference. III. Developer Changes These changes should help developers by making it easier to get their changes into regular builds. Changes are minimal: most developers will only need to name a new GIT branch. The biggest change for developers will be to provide named branches for the stable version and for each release version. The OLPC Foundation may create a named branch for inactive and completed projects that should be a release. Also, the OLPC Foundation may create an as shipped branch when we finish a release cycle. We recommend that projects try to develop new features be in separate branches and merge them back into a 'stable' branch as they are completed; just our advice. See the wiki for the latest branch names and explanations. Another exciting change is a new look to the online GIT
Re: http access to git repository
Try man git-http-push, man git-http-pull. -- Charles 2008/4/1 Ravi Kondamuru [EMAIL PROTECTED]: Hi, I have problem accessing git as the firewall seems to be disallowing git port. Are there any alternative ways to accessing git repository? thanks, Ravi. ___ 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 OLPC Process and Rules for Builing Activities, Releases, and Firmware Builds
, formats, and bundles. There are three versions in the build process: product, pre-product, and joyride. Product releases are the two per year tested releases for deployment. Pre-product releases are the alpha, beta, and release candidate builds that replace the current 'stable' designation. Joyride is the nightly build with all the stability of driving fast on slippery mountain roads hoping not to crash the same place twice. Only joyride builds will pull the latest library versions from off-site servers. There will also be multiple formats produced in the build process, which should make it more accessible to Activity developers. Each build will be available in Ubuntu (currently Gutsy version) packages, Ext3 files for Q/EMU, a virtual drive for VMWare servers, jffs2 (flash/Nand), .zip/.xo, and continued support for sugar-jhbuild users. We are still fleshing out the details for a testing harness version and for a Macintosh developer version. We hope that any programmer who wants to develop for the OLPC will be able to do so with under thirty minutes of set-up. Finally, there will be multiple bundles for each build: Sugar/OS without Activities; Sugar/OS with Solidity=Ice Activities; Sugar/OS with Solidity=Ice Activities Source Code; and Sugar/OS with All Activities Source Code. The two bundles with source code may exceed the standard storage ability of XO-1 laptops. There will also be some special bundles. Live CD builds, based on the Sugar/OS with Solidity=Ice Activities bundle for the latest product release will be aimed at potential donors and press. We are still exploring making custom deployment bundles to provide configuration help such as selecting activities, the Jabber host, default languages, and security settings. That's a lot of builds! At a minimum we are expecting (3 versions) x (4 formats) x (4 bundles) + Live and deployment builds. The extra complexity will be worth it to make available the torrent or download that best suits your needs. V. Conclusions Summary The OLPC Build Process is getting cleaner, faster, and more reliable. The new time-based release cycle provides the regularlity needed by teachers and regional groups; the new developer interface and branch structure speeds up development; and the new build targets make it easier to acquire and test the latest builds. Any challenges with the transition will be rewarded with higher quality software. Charles Merriam April 1st OLPC Build Administrator [EMAIL PROTECTED] Note for Non-United States Readers: The United States has a tradition of pranks and jokes each April 1st. See Wikipedia for an explanation of this tradition. This is not a new policy of the OLPC Foundation, nor do I represent them. The OLPC Foundation remains committed to retro style build processes emphasizing the artistic experience of administration over product quality. Smile. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [Testing] Automated testing, OLPC, code+screencasts.
Hello Titus, Good to see you at PyCon. Here's a random idea I think would be worthwhile, if it could work well into the Sugar-gui or even the other subsystems of Sugar. Tooting my own horn today, Charles Merriam === http://www.charlesmerriam.com/blog/?p=106 Problem: Testing GUIs tends to be hard. Writing a reasonable test for GUIs usually involves contortions to find the correct widget and values. For example, a test might want to confirm that the font is now bold for the third text field, named system_danger_level, in the hbox in the floating frame in the second panel in the third tab bar in the dialog box. Figuring out how to tell if the test passed is usually difficult. Solution: Add a function in the GUI framework that returns a single large data structure for the state of the GUI. Use standard Python programming to navigate it. The test becomes: assert(gobject.dump(system_danger_level)[font][style] == bold) This is one of my poorly researched ideas: it came up while talking with Shandy before Mark Shuttleworth's talk last night. All GUI frameworks are inherently a bit crufty and hard to navigate. On the other hand, the data types in Python are rich: dictionaries, arrays, nested structures, etc. While handing such a data structure to PyGTK to modify the GUI might require a lot of writing, asking PyGTK to disgorge such a data structure is far easier. Consider adding it the the GObject functions. Call a new gobject.dump_main_context() and get a huge Python data structure back. It might have lots of redundant methods of finding the same data. For example, a tree of all the contexts or dialog boxes and the usual tree objects inside that are grabbed by tools like Guitar's GUI Ripper. It might also have a handy hash of object id's and their associated sub-records, like an index into the big tree. While some may decry the memory and time cost of creating this tree might have the legitimate wish for a second function known simply as gobject.dump(). It would take a single identifier tag and return a single object from that tag 'downwards' in detail. A well published heuristic would have it do the right thing when given a tag that exists in multiple places. This feels like a implement it first and then see if its useful type of hack. On Wed, Mar 26, 2008 at 12:31 AM, Titus Brown [EMAIL PROTECTED] wrote: Hi all, I'm delurking with a vengeance here, I think :). ... I would very much like to try to introduce systematic and robust GUI testing into the OLPC and I will be working towards that end as time and resources permit. Constructive comments and genuine interest are welcome! cheers, --titus - Dr. C. Titus Brown, [EMAIL PROTECTED] Asst Prof., Michigan State U. http://ged.cse.msu.edu/ ___ Testing mailing list [EMAIL PROTECTED] http://lists.laptop.org/listinfo/testing ___ 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: [Testing] Automated testing, OLPC, code+screencasts.
I just glanced at DogTail:
+ GUI testing written in Python
+ Allows some querying/setting of actual fields.
- Not maintained for past couple years
- Most documentation and examples missing
- Test cases appear verbose, but hard to tell with all examples missing.
- Testing uses 'tree' approach requiring manaully setting focus and such.
- Relies on all sorts of stuff, including the disability libraries and
CORBA of all things to send messages.
Here's a good document:
http://www.redhat.com/magazine/020jun06/features/dogtail/
Here's a sample test script:
# Focus gedit's text buffer.
25 focus.text()
26
27 # Load the UTF-8 demo file. Use codecs.open() instead of open().
28 from codecs import open
29 from sys import path
30 utfdemo = open(path[0] + '/data/UTF-8-demo.txt')
31
32 # Load the UTF-8 demo file into the text buffer.
33 focus.widget.text = utfdemo.read()
34
35 # Click gedit's Save button.
36 click('Save')
37
38 # Focus gedit's Save As... dialog
39 focus.dialog('Save as...')
40
41 # click the Browse for other folders widget
42 activate('Browse for other folders')
43
44 # Click the Desktop widget
45 activate('Desktop', roleName = 'table cell')
Overall, DogTail spends most of its effort getting around the PyGTK
limitations about exposing the entire tree. It uses the accessibility
UI as a substitute for having an API into PyGTK. Most of the test
script is mucking around getting focus, etc.
Anyone actually use DogTail still?
Charles
On Thu, Mar 27, 2008 at 5:55 AM, Chris Ball [EMAIL PROTECTED] wrote:
Hi Charles,
Consider adding it the the GObject functions. Call a new
gobject.dump_main_context() and get a huge Python data structure
back.
You're describing at-spi¹, which is an accessibility framework that
provides introspection on GUI widgets. There's an existing GUI test
framework called Dogtail², written in Python, that uses the widget data
from at-spi in order to allow easy addressing of widgets for writing
GUI tests for GTK apps.
It would be a great project for someone to look into what it takes
to get Dogtail working on the XO. It has some gnarly dependencies
(bonobo, corba..) but there wouldn't be a need to put them in every
build, just builds that we want to run a test harness against.
Thanks,
- Chris.
¹: http://developer.gnome.org/projects/gap/
²: http://people.redhat.com/zcerza/dogtail/
--
Chris Ball [EMAIL PROTECTED]
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel
Re: Project submission: FiftyTwo
Hmm.. This should be in the list.
Mike Fletcher used this as the subject of his PyCon Tutorial this year
and has complete running code
including mesh networking. It should be put in the current
activities, and the wiki should be updated.
Mike? You handle or make me handle?
Charles
On Wed, Mar 26, 2008 at 7:06 PM, [EMAIL PROTECTED] wrote:
1. Project name : Tic-Tac-Toe
2. Existing website, if any :
3. One-line description : Tic-Tac-Toe for the XO
4. Longer description : A Tic-Tac-Toe game for the XO, using the
actual X
: and O from the front of the XO.
:
:
5. URLs of similar projects :
6. Committer list
Please list the maintainer (lead developer) as the first entry. Only list
developers who need to be given accounts so that they can commit to your
project's code repository, or push their own. There is no need to list
non-committer developers.
Username Full name SSH2 key URLE-mail
- --
#1 kawk kawk Already submitted [EMAIL
PROTECTED]
7. Preferred development model
[X] Central tree. Every developer can push his changes directly to the
project's git tree. This is the standard model that will be familiar to
CVS and Subversion users, and that tends to work well for most
projects.
[ ] Maintainer-owned tree. Every developer creates his own git tree, or
multiple git trees. He periodically asks the maintainer to look at one
or more of these trees, and merge changes into the maintainer-owned,
main tree. This is the model used by the Linux kernel, and is
well-suited to projects wishing to maintain a tighter control on code
entering the main tree.
If you choose the maintainer-owned tree model, but wish to set up some
shared trees where all of your project's committers can commit directly,
as might be the case with a discussion tree, or a tree for an individual
feature, you may send us such a request by e-mail, and we will set up the
tree for you.
8. Set up a project mailing list:
[ ] Yes, named after our project name
[ ] Yes, named __
[X] No
When your project is just getting off the ground, we suggest you eschew
a separate mailing list and instead keep discussion about your project
on the main OLPC development list. This will give you more input and
potentially attract more developers to your project; when the volume of
messages related to your project reaches some critical mass, we can
trivially create a separate mailing list for you.
If you need multiple lists, let us know. We discourage having many
mailing lists for smaller projects, as this tends to
stunt the growth of your project community. You can always add more lists
later.
9. Commit notifications
[ ] Notification of commits to the main tree should be e-mailed to the list
we chose to create above
[ ] A separate mailing list, projectname-git, should be created for
commit
notifications
[X] No commit notifications, please
10. Shell accounts
As a general rule, we don't provide shell accounts to developers unless
there's a demonstrated need. If you have one, please explain here, and
list the usernames of the committers above needing shell access.
11. Translation
[X] Set up the laptop.org Pootle server to allow translation commits to be
made
[ ] Translation arrangements have already been made at ___
12. Notes/comments:
I noticed this page: http://wiki.laptop.org/go/Tic_tac_toe , and decided to
do something about it. This is the result.
Thanks,
KAWK
___
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: [Olpc-open] Nortel LearniT animations (Seth Woodworth)
Hi All, I think I added all the substance from this thread into the wiki (http://wiki.laptop.org/go/Gnash). It's late, so I would apprecate Rob et al doing a quick read. Also, can someone add more information about the specific gnash version/codecs being installed on which XOs and confirm that the primary issue in developing Flash for Gnash is picking open codecs? Have a great day! or evening! Charles Merriam On Mon, Mar 24, 2008 at 10:41 AM, Rob Savoye [EMAIL PROTECTED] wrote: Steve Holton wrote: Gnash will *never* be fully compatible with Flash because the closer Gnash gets to being a viable free Flash replacement, the more incentive there is for Adobe to change the Flash specification in a way to break compatibility. They've already changed the format in a big, hence all our hard work to reverse engineer SWF v9. ActionScript 3 is finally ECMAScript compatible, same as JavaScript, so I doubt that'll change much in the future. Also all the changes in SWF v9 were performance oriented, and that required a new VM. Gnash now does support the SWF v9 format changes, that was easy. It's implementing the ActionScript class libraries that's much of the work left. SWF has evolved very slowly, so I don't feel we'll be chasing Adobe for long. Two decades in the Microsoft format wars should have taught that lesson to everyone by now. Look how long (and how much) it's taken ODF to get where it's at. Yes, but as far as I can tell, OpenOffice works well enough with M$ Office, compatibility wise, that I haven't had to use M$ Office for many years. Not everything converts in OO 100% all the time, but what doesn't work I can easily live with. OTOH, the XO offers us an opportunity to create a new standard among an audience which has no investment in the old. But this is a limited opportunity. New standards still don't solve the problem of playing existing content (often proprietary), which is what I though we were discussing. Also playing SWF files in the future is not something we worry about, since that will only effect new content, which doesn't exist yet. :-) My point is that we want people to work with us. Most of the time all I hear is Gnash sucks, it's not 100% compatible yet. We know that already... What we want to do is identify what sucks, produce test cases, and then fix the problems. Bitching about the problem and dumping Gnash does not solve the problem, it merely ignores it. It's the easy way out. Yes, it can take some time for an end user with a problem to work with us till we identify what is wrong. As none of us can use the Adobe player due to clean room problems, it's our end users that help us work on testing compatibility. Many people have helped contribute to the development of Gnash merely by helping answer questions about what's wrong, and trying patches, and most of them were not professional engineers. All we are asking for is help beyond just griping, and patience as our small team pushes forward. - rob - ___ 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: Google Summer of Code and OLPC
No problem. Wiki now has a link. -- Charles On Thu, Mar 20, 2008 at 8:04 PM, Martin Langhoff [EMAIL PROTECTED] wrote: On Thu, Mar 20, 2008 at 10:11 PM, Roberto Fagá [EMAIL PROTECTED] wrote: Try One Laptop Per Child :) http://code.google.com/soc/olpc/about.html Ooops! I'm a tired fool it seems. Sorry about the noise! martin ___ 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: PlayGo
Hello Ed, The error message could be better. The project is projects/PlayGo. I'll double check when I get back to my laptop, but try: git clone git://dev.laptop.org/projects/PlayGo It's a common problem. Add it to the wiki pls? Charles On Tue, Mar 18, 2008 at 4:06 PM, Edward Cherlin [EMAIL PROTECTED] wrote: I just now tried to git-clone PlayGo, and couldn't. Do you know what is wrong? https://dev.laptop.org/git?p=projects/PlayGo;a=tree is fine, but git clone git://dev.laptop.org/PlayGo Initialized empty Git repository in /home/mokurai/dev/OLPC/git/PlayGo/.git/ fatal: The remote end hung up unexpectedly fetch-pack from 'git://dev.laptop.org/PlayGo' failed. It doesn't say that the directory doesn't exist, or that it is not a git repository. In fact, it doesn't actually say that anything is wrong. It just fails. -- Edward Cherlin End Poverty at a Profit by teaching children business http://www.EarthTreasury.org/ The best way to predict the future is to invent it.--Alan Kay ___ 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: OLPC seeks a CEO -- who was your favorite CEO elsewhere?
That would be Alan Baratz, former CEO of JavaSoft, last seen at Cisco after his latest company was acquired. Rumor has that Cisco is choosing not to integrate NeoPath gracefully and Alan may be available. Anyone keep closer relationships with him? I haven't talked to him since JavaSoft. Charles Merriam On Tue, Mar 11, 2008 at 1:47 PM, John Gilmore [EMAIL PROTECTED] wrote: http://www.businessweek.com/technology/content/mar2008/tc2008035_429837.htm OLPC is looking for a CEO. Nicholas is more of an idea man, and he plans to continue as Chairman and cheerleader. But he appears to have realized that with its current management, the organization can't outgrow its early chaos. (For this I give him every credit; most founders who aren't suited to manage a larger, more structured organization resist installing a steady hand at the wheel.) There are probably a few people on the devel list who are actually qualified to be CEO of a nonprofit tech company like OLPC. I encourage them to apply (it's not clear how, which shows you how far things have degenerated). But I'm more interested in asking the software developers on the list: == Who's the best manager or CEO you ever worked for? Suggest to that person that they consider the job. OLPC has plenty of resources, and also plenty of challenges. We on the outside have only seen a fraction of them (like schedules sliding out of control; botched distribution; support handled only by the skin of the teeth; key people dragged around to fill big holes, leaving other big holes behind them; diminished expectations in both sales and technical achievement). OLPC has already changed the world in a small way, by teaching us that there's a vibrant world market for low cost, high function portable computers, and reminding us how much leverage there is in third world educational improvement. OLPC still has a chance to change the world in a big way, by satisfying that market, rather than leaving it to commercial companies to half-assedly pick up the pieces. Steering OLPC back on to the rails before it crashes and burns will be a job your favorite CEO or manager will never forget. Give 'em a call... John ___ 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: Microsoft? (was Re: OLPC seeks a CEO -- who was your favorite CEO elsewhere?)
Is there *any* suggestion that the entire Microsoft on OLPC story is anything other than: 1. A small group of experimenters at Microsoft playing around in the slack time. 2. FUD stories to downplay OLPC. The OLPC corporate needs to respond with a one liner that we have no plans to now or in the future. The story could change, but the current stance does need to be known. FUD works kind of like nuisance law-suits: failure to say anything is an automatic loss. Nothing mean, cruel, or chilling. Just OLPC does not help and support this effort. MS is on their own. Charles 2008/3/11 Todd Cranston-Cuebas [EMAIL PROTECTED]: I'm waiting to hear about this one also. On the one hand the OLPC can't be shipped with the Flash plug-in but the whole project is going to go to Microsoft? Talk about moving between extremes. I'm not sure why a more balanced approach couldn't work but then again, I'm more of a supporter (bought unit through the G1G1 program) and advocate. I'm not a coder, etc. but have been really encouraged by the dramatic grassroots support for the unit to take up the need for support, etc. I have to say that I'm a little surprised that the actual shipping OS, Sugar interface, activities etc. are still very much a work in progress (some bugs, keys not enabled, no reveal code key, networking problems, etc.), but that's not necessarily bad as long as there is healthy support for refinement. It's this titanic shift that is catching me off-guard. Let's face it, the OLPC has been both enhanced by, and perhaps held back by, a hard line support of just OS and a strict constructivist educational approach. Then again, fervent adherence to a philosophy and cause has pushed the world to see what could be possible through this little green machine. Todd On Tue, Mar 11, 2008 at 2:26 PM, victor [EMAIL PROTECTED] wrote: More worrying is this bit from the article in the link OLPC will hand more of the development and support of its XO laptop and its core software to technology companies, (...), and Microsoft (MSFT), which is just now putting the finishing touches on a version of Windows for the XO machine. I didn't know Microsoft and Windows were going to be there. So why all the effort if in the end a closed OS is going to be used? Is this true? Victor - Original Message - From: John Gilmore [EMAIL PROTECTED] To: [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Tuesday, March 11, 2008 8:47 PM Subject: OLPC seeks a CEO -- who was your favorite CEO elsewhere? http://www.businessweek.com/technology/content/mar2008/tc2008035_429837.htm OLPC is looking for a CEO. Nicholas is more of an idea man, and he plans to continue as Chairman and cheerleader. But he appears to have realized that with its current management, the organization can't outgrow its early chaos. (For this I give him every credit; most founders who aren't suited to manage a larger, more structured organization resist installing a steady hand at the wheel.) There are probably a few people on the devel list who are actually qualified to be CEO of a nonprofit tech company like OLPC. I encourage them to apply (it's not clear how, which shows you how far things have degenerated). But I'm more interested in asking the software developers on the list: == Who's the best manager or CEO you ever worked for? Suggest to that person that they consider the job. OLPC has plenty of resources, and also plenty of challenges. We on the outside have only seen a fraction of them (like schedules sliding out of control; botched distribution; support handled only by the skin of the teeth; key people dragged around to fill big holes, leaving other big holes behind them; diminished expectations in both sales and technical achievement). OLPC has already changed the world in a small way, by teaching us that there's a vibrant world market for low cost, high function portable computers, and reminding us how much leverage there is in third world educational improvement. OLPC still has a chance to change the world in a big way, by satisfying that market, rather than leaving it to commercial companies to half-assedly pick up the pieces. Steering OLPC back on to the rails before it crashes and burns will be a job your favorite CEO or manager will never forget. Give 'em a call... John ___ 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 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel ___ Devel mailing list Devel@lists.laptop.org
Re: Microsoft? (was Re: OLPC seeks a CEO -- who was your favorite CEO elsewhere?)
Um, you guys do know how to use the search function on the Wiki, don't you? Please be civil. Yeah, Nicholas said pretty much the first half of that months ago. The issue is the conflicting Negroponte quotes: Windows on XO has not only been happening with our consent, but (also our) collaboration. Some of the first engineering models from any given build go to them, Negroponte said. http://www.news.com/Negroponte-Windows-key-to-OLPC-philosophy/2100-1016_3-6215837.html Negroponte says that a Windows operating system is in the process of being fine-tuned on the XO as we speak. Microsoft and OLPC are in discussion on how to release it, as well as how to announce, he said. Negroponte added that the Windows operating system should be available on the XO in less than 60 days. http://www.olpcnews.com/software/windows/xp_on_the_xo_in_60_days.html So, who knows? Charles ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
