OLPC:News
Laptop News 2007-07-07
1. Arahuay, Peru: Carla Gomez Monroy has been helping the ministry of education with a school trial at the Institución Educativa Apóstol Santiago, a combination primary and secondary school in a small town in the Cordillera de la Viuda, 2600 m. above sea level.
2. Taipei: Mary Lou Jepsen gave a keynote at a display manufacturing conference in Taipei where she highlighted the fact that the XO display took only six months to go from specification to full certification, ready for high-volume mass production product. This rapid development is unheard of in the display industry where 10-to-20 years is more the norm for a new display—and thus the XO was the center of much discussion at the conference. Other large display manufacturers are expressing interest in helping with Gen2 development as well as providing a second source of the Gen1 display. Of course Mary Lou explained our loyalty and strong relationship with ChiMei. The reception was quite different from 18 months ago, when getting buy-in from the display manufacturers was one of the largest challenges for OLPC.
3. OLPC was nominated and selected as a potential beneficiary of the American Express Card's Members Project (http://www.membersproject.com). Projects are voted for through several rounds and the winners are eligible for a grant of between $1–$5 million. OLPC is listed under project number 07229.
4. XO testing: Quanta has put an enormous effort into testing the laptops in each of the five builds to date (A Test and B1–B4). The approximately 7000 prototypes have undergone temperature, electrical, mechanical, durability, and environmental testing:
Temperature: 55C/40% relative humidity (RH) operation test, 32C/50% RH operation test, 30C/85% RH operation test, –0C operation test, 85C storage test, –40C storage test, thermal shock and profile test (60C to –20C);
Electrical: AC power, BIOS flashing, Open Firmware, power management, USB 2.0, NAND flash, Wireless LAN, camera, memory, battery, LED indicator, stress test, ESD, battery discharge, LCD module verification, line-voltage and frequency test, power-on/off test, altitude test, wave-form measurement, frequency response, speaker performance, touch-pad performance, S0 state, S3 states, driver level, frequency accuracy, oscillation allowance, negative resistance, load capacitance, DDR1, critical trace, power-rail ramp, voltage level and noise, USB 1.1, Radiation of EN 55022, EN 61000-4-4, skin/case temperature, etc.;
Mechanical environmental test: operating and non-operating vibration test, operating and non-operating shock test, package drop test, package storage test, tilt drop test, free drop test, LCD stress test, base pressure test, LCD-pressure vibration test, switch-protection test, LCD-twist test, connector-tension test, adapter-cable bending test, spill test, water test;
Durability test: hinge 65K cycles, battery 10K cycles, buttons 1M cycles, power button 700K cycles, touch-pad buttons 3M cycles, USB ports 10K cycles, DC in 20K cycles, DC-in, line-out, and mic-in 10K cycles, wireless-antenna 5K cycles, SD card 16K cycles;
Special environmental test: sand and dust test, salt-fog test, solar- radiation test, rain test;
Abusive test: free-throw test, USB and SD card reverse test, tumbling test, water-sprinkle test, hinge max-angle test, tablet-mode max-angle test, antenna max-force test.
These are torture tests. Most of these tests are harder—by far—than that required for conventional laptops Currently, XO has passed the vast majority of tests. Provisions to pass the tough ESD, salt-fog, power-on/off test, and operating shock test are under way; as are even higher free-drop tests, more stringent hinge-torque tests; as well as stronger set of testing underway at UL as we enter C-Test phase in August.
5. Mechanicals: Bret Recor of Fuse Project was in Shanghai this week to work closely with Frank Lee of Quanta on finalizing the texturing for plastic-housing parts. Bean texture will appear on the exterior white housing and a matte—“satin”—texture on the interior and the green parts.
6. Trial-2 software: At the beginning of the week we added the latest Marvell firmware to both the XO and the school server software images to get past “flag day” in Cambridge: builds before Build 486 (or Build 406.16) will not work with these later builds. Henceforth, as we send products out to the field, we need to know if the recipient already has XOs; any older machines will need to be upgraded so the new ones and old ones will work together. (The test group continues to keep a log of Release Notes for each build, which can be accessed at Test Group Release Notes.
7. Hardware management: Richard Hughes continued his work on our hardware policy manager. He is in the final stages of getting it into our builds by running it through the Fedora review process. Matthias Clasen has been helping him run through the process and has been pointing out improvements to be made. He also integrated open hardware manager (OHM) with the new X IDLETIME alarm interface. He also added support for the battery and AC adapter to the hardware abstraction layer (HAL) so that it is more easily exported to programs. Using this interface, Marco Gritti has added support to Sugar to extract the battery information from HAL.
8. Mesh collaboration: The Collabora team had a busy week working on bug-fixes in the dbus-python bindings, debugging interactions between telepathy and the other modules, working through details of the tubes API, fixing issues with the peer-to-peer extensible messaging and presence protocol (XMPP), working on the multicast protocol for the mesh, finishing up the “hellomesh” activity, and fixing the chat activity.
9. Building builds: John Palmieri continued pushing builds out the door. He has been splitting his time between trying to get Trial-2 work done and also working on a stable 406 build (the latter build includes power management and other low-level features, but not the new collaboration and Journal features). He also worked on getting QEMU (an open source machine emulator and virtualizer) and VirtualBox (a commercial virtualizer) working with our images again.
The basic Vmware image conversion process outlined at in the OLPC wiki (See Emulating the XO/UsingVMware) is also working. Scott Devine has been working on VMware mesh-network emulation and the XO images are now part of Vmware’s daily regression testing.
10. Sugar: Marco Gritti and Tomeu Vizoso continue to chase bugs in Sugar and the Journal as part of the Trial 2 work. Walter Bender and Eben Eliason revisited the XO color schemes; more vivid and legible color dyads will be available in the upcoming builds. Dan Williams worked on Network Manager, fixed an avahi bug, fought through numerous packaging issues as part of the Fedora Core 7 move, and also worked on some gstreamer issues around camera support.
11. Updates: Scott Ananian wrote a recursive binary diff tool based on bsdiff/bspatch in order to investigate the lower limits of upgrade sizes. Comparisons with rsync and the Dan Williams patched version of “updatinator” showed that our various upgrade strategies are now not too far from optimal.
12. X Window System: Bernie Innocenti spent much of the week working on bug #1837, the “fancy color on text” bug, which turned out to be pretty tough. The more he dug into it, the more he keep getting evidence pointing in opposite directions. We see similar text corruption on OLPC, F8 and Ubuntu Gutsy on the Geode LX (both 16bpp and 32bpp) and Radeon R200 (but only in 16bpp) in both with EXA and XAA. Disabling acceleration cures it. The plot thickens.
Another serious bug that had been making the X server on the GX (B2 and B1) unusable in the development builds was found through work of Jordan Crouse and Adam Jackson late Friday afternoon.
13. Kernel: Andres Salomon worked on getting code upstream this week. The Geode support in master is mostly in Andrew Morton's tree; some is trickling into Linus's tree as well. Andres also did the standard merge dance (nothing exciting there, though we should be approaching a final 2.6.22 soon). Marcelo Tosatti spent time merging his USB changes to master.
Marcelo and Chris Ball looked into some suspend/resume failures. We can reliably suspend and resume wireless so long as the wireless module remains connected to the USB; Marcelo and Cozybit continue to investigate issues suspending and resuming with the wireless disconnected entirely.
The “pop” that we receive from the amplifier is barely audible on the B4 machines, but we will continue to investigate it—if the pop is caused by a hardware bug, now is the time to fix it.
The microphone LED was turning on regardless of whether we were using the audio device for playback or recording. Chris came up with logic to keep it turned off when we aren't recording, implemented it, and will send his patch to Jaya Kumar for review.
Scott imported glibc into git and (after some fighting) got it to build, so he is setup to write the glibc RDNSS-over-RA patches.
14. School server: Mitch Bradley got a preliminary version of PowerPC Open Firmware running on the school-server development machine. Mitch also released the core PowerPC Open Firmware code to openbios.org under an MIT license.
Dan Margo completed the scripts that will let us build OLPC-specific RPM configuration packages. These packages follow a fairly complex install-and-update policy that: (A) organizes OLPC-specific configurations into a central directory, etc.olpc; and (B) respects user's configuration changes, while clobbering RPM defaults. These scripts are in want of more debugging on large, serious use cases, but are concept-complete.
Next week we plan to start building Fedora Core 7 livecd build scripts from a clean slate (and merge the work Holger has done based on Fedora 6) Once that is straightened out, we can start determining and then exporting the XS-specific configurations.
Scott also fought the QEMU/KVM (kernel-based virtual machine) fight and got XO and (sort of) XSX images built and working under emulation. The ground is laid for Dan and Scott to create XSX builds based on Fedora Core 7 images next week using Dan's rpm-configuration code.
15. Security: Mitch and Ivan Krstić hashed out firmware security issues around activation, firmware updates, and developer keys. Ivan, Mitch, and Scott defined three code-paths: (1) the chain from booting a signed kernel though invoking activation or pivoting to one of two base file-systems (upgrade or backup); (2) the anti-theft server, anti-theft client, and “invoke-upgrade-now” interactions; and (3) the fetch-upgrade hand-off to the security kernel, which validates the upgrade and tweaks some bits to signal the boot code (closing the circle). Implementation has begun: Ivan has built a skeleton antitheft server and client, and Scott has an initrd that takes control immediately after open firmware boots our (signed) kernel.
16. Embedded controller: Mitch helped Joel Stanley (Richard Smith's Summer of Code intern) get started on embedded controller (EC) development and in the process discovered a fix for the flakiness of the EC recovery process.
To help with the ongoing debugging efforts of Trac Bug #1752 Richard created a test version of the EC code that asserts the WLAN wakeup pin before issuing the SCI to the host. This allows the WLAN to have its USB bus up and running prior to the host. Experiments with WLAN firmware that does not detach from the USB bus works with suspend/resume. The EC test firmware was a possible workaround but Chris found that it still had issues.
David Woodhouse and Richard worked on making the PCB temp sensor work and verify that the resistor divider is set correctly. However, they found that the EC code is not working correctly. The voltage A/D readings do not appear to be updating. With a bit of Forth code under OFW to read the A/D directly and then they looked at the range of the sensor. The hardware is OK and there is enough resolution for our purposes
17. Content jams: The Commonwealth of Learning (COL) is working actively on open content for educators and students, especially those in regions newly introduced to computers. They are interested in starting a Summer of Content project along with OLPC, supporting existing open content efforts and focusing on short-term goals that will lead to literacy and more content creation. They are starting work on a Spanish Wikieducator, and planning content-creation workshops in each of the 53 Commonwealth states. Mel Chua met with Wayne Mackintosh of COL and worked out some initial details. Summer of Content would provide visibility to open content projects that need help and a unified way for interested students and young teachers to intern for these projects, focusing on potential interns in the Southern Hemisphere and running two summers a year, starting at the end of this calendar year.
18. Our Stories: UNICEF has presented “Our Stories” to their country offices in Uruguay, Brasil, Argentina, Nigeria, Rwanda, and Ethiopia, and are starting to collect stories that can be published and shared from their existing youth networks. Google's OurStories team now includes two engineering leads and website and interface leads.
19. Educational activity guidelines: Lauren Klien has been working on templates for learning activities and guidelines for people writing them (See Educational activity guidelines, Sample learning activities, and Learning activities). If you are working on activities of your own, please post them to the wiki and add/suggest guidelines of your own that you have found useful.
20. Distributed content: Thibaut Lamadon is working on the next iteration of his MeshBoard activity—a community bulletin board that runs serverlessly on the mesh network. Thibaut is testing out a new implementation that uses Tubes APIs.
21. Wiki: A weekly review of the OLPC wiki changes (now ~1000 a week) began last week. Xavier Alvarez’s latest round of categorization and organization of the wiki includes a proposal for a standard set of information boxes covering licensing and subject matter and audience, where to get more information, and what features or WikiProjects something is associated with. Comments and suggestions welcome (See OBX proposals).
MaMaMedia has started to move content into the OLPC wiki (See World Wide Workshop Foundation).
22. Reading: David Teller reports that the Lector book-reading project, which is javascript-based and runs inside the browser without its own UI, has a new version. They are just now in touch with Ian Bicking and Stephen Thorne about integrating their work on an in-browser interface for XO. Ian has a simple booksplitter working to split large files into small readable chunks. Josh Gay is working on a proper annotation spec to send around for how a generalized stet implementation will work; for the coming week.
23. Games: Julius Lucks and team continued to tweak their number munchers game, making sure it was properly localized and can share tilesets with the Memonumbers game (See Kuku).
More News
Laptop News is archived at Laptop News. Also on community-news.
You can subscribe to the OLPC community-news mailing list by visiting the laptop.org mailman site.
Press requests: please send email to press@racepointgroup.com
Milestones
Latest milestones:
Nov. 2007 | Mass Production has started. |
July. 2007 | One Laptop per Child Announces Final Beta Version of its Revolutionary XO Laptop. |
Apr. 2007 | First pre-B3 machines built. |
Mar. 2007 | First mesh network deployment. |
Feb. 2007 | B2-test machines become available and are shipped to developers and the launch countries. |
Jan. 2007 | Rwanda announced its participation in the project. |
All milestones can be found here.
Press
Laptop News 2007-07-07
1. Arahuay, Peru: Carla Gomez Monroy has been helping the ministry of education with a school trial at the Institución Educativa Apóstol Santiago, a combination primary and secondary school in a small town in the Cordillera de la Viuda, 2600 m. above sea level.
2. Taipei: Mary Lou Jepsen gave a keynote at a display manufacturing conference in Taipei where she highlighted the fact that the XO display took only six months to go from specification to full certification, ready for high-volume mass production product. This rapid development is unheard of in the display industry where 10-to-20 years is more the norm for a new display—and thus the XO was the center of much discussion at the conference. Other large display manufacturers are expressing interest in helping with Gen2 development as well as providing a second source of the Gen1 display. Of course Mary Lou explained our loyalty and strong relationship with ChiMei. The reception was quite different from 18 months ago, when getting buy-in from the display manufacturers was one of the largest challenges for OLPC.
3. OLPC was nominated and selected as a potential beneficiary of the American Express Card's Members Project (http://www.membersproject.com). Projects are voted for through several rounds and the winners are eligible for a grant of between $1–$5 million. OLPC is listed under project number 07229.
4. XO testing: Quanta has put an enormous effort into testing the laptops in each of the five builds to date (A Test and B1–B4). The approximately 7000 prototypes have undergone temperature, electrical, mechanical, durability, and environmental testing:
Temperature: 55C/40% relative humidity (RH) operation test, 32C/50% RH operation test, 30C/85% RH operation test, –0C operation test, 85C storage test, –40C storage test, thermal shock and profile test (60C to –20C);
Electrical: AC power, BIOS flashing, Open Firmware, power management, USB 2.0, NAND flash, Wireless LAN, camera, memory, battery, LED indicator, stress test, ESD, battery discharge, LCD module verification, line-voltage and frequency test, power-on/off test, altitude test, wave-form measurement, frequency response, speaker performance, touch-pad performance, S0 state, S3 states, driver level, frequency accuracy, oscillation allowance, negative resistance, load capacitance, DDR1, critical trace, power-rail ramp, voltage level and noise, USB 1.1, Radiation of EN 55022, EN 61000-4-4, skin/case temperature, etc.;
Mechanical environmental test: operating and non-operating vibration test, operating and non-operating shock test, package drop test, package storage test, tilt drop test, free drop test, LCD stress test, base pressure test, LCD-pressure vibration test, switch-protection test, LCD-twist test, connector-tension test, adapter-cable bending test, spill test, water test;
Durability test: hinge 65K cycles, battery 10K cycles, buttons 1M cycles, power button 700K cycles, touch-pad buttons 3M cycles, USB ports 10K cycles, DC in 20K cycles, DC-in, line-out, and mic-in 10K cycles, wireless-antenna 5K cycles, SD card 16K cycles;
Special environmental test: sand and dust test, salt-fog test, solar- radiation test, rain test;
Abusive test: free-throw test, USB and SD card reverse test, tumbling test, water-sprinkle test, hinge max-angle test, tablet-mode max-angle test, antenna max-force test.
These are torture tests. Most of these tests are harder—by far—than that required for conventional laptops Currently, XO has passed the vast majority of tests. Provisions to pass the tough ESD, salt-fog, power-on/off test, and operating shock test are under way; as are even higher free-drop tests, more stringent hinge-torque tests; as well as stronger set of testing underway at UL as we enter C-Test phase in August.
5. Mechanicals: Bret Recor of Fuse Project was in Shanghai this week to work closely with Frank Lee of Quanta on finalizing the texturing for plastic-housing parts. Bean texture will appear on the exterior white housing and a matte—“satin”—texture on the interior and the green parts.
6. Trial-2 software: At the beginning of the week we added the latest Marvell firmware to both the XO and the school server software images to get past “flag day” in Cambridge: builds before Build 486 (or Build 406.16) will not work with these later builds. Henceforth, as we send products out to the field, we need to know if the recipient already has XOs; any older machines will need to be upgraded so the new ones and old ones will work together. (The test group continues to keep a log of Release Notes for each build, which can be accessed at Test Group Release Notes.
7. Hardware management: Richard Hughes continued his work on our hardware policy manager. He is in the final stages of getting it into our builds by running it through the Fedora review process. Matthias Clasen has been helping him run through the process and has been pointing out improvements to be made. He also integrated open hardware manager (OHM) with the new X IDLETIME alarm interface. He also added support for the battery and AC adapter to the hardware abstraction layer (HAL) so that it is more easily exported to programs. Using this interface, Marco Gritti has added support to Sugar to extract the battery information from HAL.
8. Mesh collaboration: The Collabora team had a busy week working on bug-fixes in the dbus-python bindings, debugging interactions between telepathy and the other modules, working through details of the tubes API, fixing issues with the peer-to-peer extensible messaging and presence protocol (XMPP), working on the multicast protocol for the mesh, finishing up the “hellomesh” activity, and fixing the chat activity.
9. Building builds: John Palmieri continued pushing builds out the door. He has been splitting his time between trying to get Trial-2 work done and also working on a stable 406 build (the latter build includes power management and other low-level features, but not the new collaboration and Journal features). He also worked on getting QEMU (an open source machine emulator and virtualizer) and VirtualBox (a commercial virtualizer) working with our images again.
The basic Vmware image conversion process outlined at in the OLPC wiki (See Emulating the XO/UsingVMware) is also working. Scott Devine has been working on VMware mesh-network emulation and the XO images are now part of Vmware’s daily regression testing.
10. Sugar: Marco Gritti and Tomeu Vizoso continue to chase bugs in Sugar and the Journal as part of the Trial 2 work. Walter Bender and Eben Eliason revisited the XO color schemes; more vivid and legible color dyads will be available in the upcoming builds. Dan Williams worked on Network Manager, fixed an avahi bug, fought through numerous packaging issues as part of the Fedora Core 7 move, and also worked on some gstreamer issues around camera support.
11. Updates: Scott Ananian wrote a recursive binary diff tool based on bsdiff/bspatch in order to investigate the lower limits of upgrade sizes. Comparisons with rsync and the Dan Williams patched version of “updatinator” showed that our various upgrade strategies are now not too far from optimal.
12. X Window System: Bernie Innocenti spent much of the week working on bug #1837, the “fancy color on text” bug, which turned out to be pretty tough. The more he dug into it, the more he keep getting evidence pointing in opposite directions. We see similar text corruption on OLPC, F8 and Ubuntu Gutsy on the Geode LX (both 16bpp and 32bpp) and Radeon R200 (but only in 16bpp) in both with EXA and XAA. Disabling acceleration cures it. The plot thickens.
Another serious bug that had been making the X server on the GX (B2 and B1) unusable in the development builds was found through work of Jordan Crouse and Adam Jackson late Friday afternoon.
13. Kernel: Andres Salomon worked on getting code upstream this week. The Geode support in master is mostly in Andrew Morton's tree; some is trickling into Linus's tree as well. Andres also did the standard merge dance (nothing exciting there, though we should be approaching a final 2.6.22 soon). Marcelo Tosatti spent time merging his USB changes to master.
Marcelo and Chris Ball looked into some suspend/resume failures. We can reliably suspend and resume wireless so long as the wireless module remains connected to the USB; Marcelo and Cozybit continue to investigate issues suspending and resuming with the wireless disconnected entirely.
The “pop” that we receive from the amplifier is barely audible on the B4 machines, but we will continue to investigate it—if the pop is caused by a hardware bug, now is the time to fix it.
The microphone LED was turning on regardless of whether we were using the audio device for playback or recording. Chris came up with logic to keep it turned off when we aren't recording, implemented it, and will send his patch to Jaya Kumar for review.
Scott imported glibc into git and (after some fighting) got it to build, so he is setup to write the glibc RDNSS-over-RA patches.
14. School server: Mitch Bradley got a preliminary version of PowerPC Open Firmware running on the school-server development machine. Mitch also released the core PowerPC Open Firmware code to openbios.org under an MIT license.
Dan Margo completed the scripts that will let us build OLPC-specific RPM configuration packages. These packages follow a fairly complex install-and-update policy that: (A) organizes OLPC-specific configurations into a central directory, etc.olpc; and (B) respects user's configuration changes, while clobbering RPM defaults. These scripts are in want of more debugging on large, serious use cases, but are concept-complete.
Next week we plan to start building Fedora Core 7 livecd build scripts from a clean slate (and merge the work Holger has done based on Fedora 6) Once that is straightened out, we can start determining and then exporting the XS-specific configurations.
Scott also fought the QEMU/KVM (kernel-based virtual machine) fight and got XO and (sort of) XSX images built and working under emulation. The ground is laid for Dan and Scott to create XSX builds based on Fedora Core 7 images next week using Dan's rpm-configuration code.
15. Security: Mitch and Ivan Krstić hashed out firmware security issues around activation, firmware updates, and developer keys. Ivan, Mitch, and Scott defined three code-paths: (1) the chain from booting a signed kernel though invoking activation or pivoting to one of two base file-systems (upgrade or backup); (2) the anti-theft server, anti-theft client, and “invoke-upgrade-now” interactions; and (3) the fetch-upgrade hand-off to the security kernel, which validates the upgrade and tweaks some bits to signal the boot code (closing the circle). Implementation has begun: Ivan has built a skeleton antitheft server and client, and Scott has an initrd that takes control immediately after open firmware boots our (signed) kernel.
16. Embedded controller: Mitch helped Joel Stanley (Richard Smith's Summer of Code intern) get started on embedded controller (EC) development and in the process discovered a fix for the flakiness of the EC recovery process.
To help with the ongoing debugging efforts of Trac Bug #1752 Richard created a test version of the EC code that asserts the WLAN wakeup pin before issuing the SCI to the host. This allows the WLAN to have its USB bus up and running prior to the host. Experiments with WLAN firmware that does not detach from the USB bus works with suspend/resume. The EC test firmware was a possible workaround but Chris found that it still had issues.
David Woodhouse and Richard worked on making the PCB temp sensor work and verify that the resistor divider is set correctly. However, they found that the EC code is not working correctly. The voltage A/D readings do not appear to be updating. With a bit of Forth code under OFW to read the A/D directly and then they looked at the range of the sensor. The hardware is OK and there is enough resolution for our purposes
17. Content jams: The Commonwealth of Learning (COL) is working actively on open content for educators and students, especially those in regions newly introduced to computers. They are interested in starting a Summer of Content project along with OLPC, supporting existing open content efforts and focusing on short-term goals that will lead to literacy and more content creation. They are starting work on a Spanish Wikieducator, and planning content-creation workshops in each of the 53 Commonwealth states. Mel Chua met with Wayne Mackintosh of COL and worked out some initial details. Summer of Content would provide visibility to open content projects that need help and a unified way for interested students and young teachers to intern for these projects, focusing on potential interns in the Southern Hemisphere and running two summers a year, starting at the end of this calendar year.
18. Our Stories: UNICEF has presented “Our Stories” to their country offices in Uruguay, Brasil, Argentina, Nigeria, Rwanda, and Ethiopia, and are starting to collect stories that can be published and shared from their existing youth networks. Google's OurStories team now includes two engineering leads and website and interface leads.
19. Educational activity guidelines: Lauren Klien has been working on templates for learning activities and guidelines for people writing them (See Educational activity guidelines, Sample learning activities, and Learning activities). If you are working on activities of your own, please post them to the wiki and add/suggest guidelines of your own that you have found useful.
20. Distributed content: Thibaut Lamadon is working on the next iteration of his MeshBoard activity—a community bulletin board that runs serverlessly on the mesh network. Thibaut is testing out a new implementation that uses Tubes APIs.
21. Wiki: A weekly review of the OLPC wiki changes (now ~1000 a week) began last week. Xavier Alvarez’s latest round of categorization and organization of the wiki includes a proposal for a standard set of information boxes covering licensing and subject matter and audience, where to get more information, and what features or WikiProjects something is associated with. Comments and suggestions welcome (See OBX proposals).
MaMaMedia has started to move content into the OLPC wiki (See World Wide Workshop Foundation).
22. Reading: David Teller reports that the Lector book-reading project, which is javascript-based and runs inside the browser without its own UI, has a new version. They are just now in touch with Ian Bicking and Stephen Thorne about integrating their work on an in-browser interface for XO. Ian has a simple booksplitter working to split large files into small readable chunks. Josh Gay is working on a proper annotation spec to send around for how a generalized stet implementation will work; for the coming week.
23. Games: Julius Lucks and team continued to tweak their number munchers game, making sure it was properly localized and can share tilesets with the Memonumbers game (See Kuku).
More News
Laptop News is archived at Laptop News. Also on community-news.
You can subscribe to the OLPC community-news mailing list by visiting the laptop.org mailman site.
Press requests: please send email to press@racepointgroup.com
Milestones
Latest milestones:
Nov. 2007 | Mass Production has started. |
July. 2007 | One Laptop per Child Announces Final Beta Version of its Revolutionary XO Laptop. |
Apr. 2007 | First pre-B3 machines built. |
Mar. 2007 | First mesh network deployment. |
Feb. 2007 | B2-test machines become available and are shipped to developers and the launch countries. |
Jan. 2007 | Rwanda announced its participation in the project. |
All milestones can be found here.
Press
Template loop detected: Press More articles can be found here.
Video
Miscellaneous videos of the laptop can be found here.
- A collection of several videos can found at OLPC.TV
- IBM Podcast, Walter Bender on One Laptop per Child [1]
- Ivan Krstić delivers a technical presentation of OLPC at the Google TechTalk series
- 60 Minutes, What if Every Child had a Laptop [2]
- CNN, Should Intel Fear $100 Laptop? [3]
- Red Hat Magazine: Inside One Laptop per Child, Episode Three
- Red Hat Magazine: Inside One Laptop per Child, Episode Two
- Red Hat Magazine: Inside One Laptop per Child, Episode One
- OLPC Video from Switzerland, 26.01.2007
- Interview with Nicholas Negroponte on the &100 Laptop
- Presentation by Jim Gettys at FOSDEM 2007
- GLOBO- BRASIL: Crianças testam computador portátil/ Students test the laptop
- Mark Foster delivers presentation to Stanford University
- Technology Review Mini-Documentary
- A Brief Demo
More articles can be found here.
Video
Miscellaneous videos of the laptop can be found here.
- A collection of several videos can found at OLPC.TV
- IBM Podcast, Walter Bender on One Laptop per Child [4]
- Ivan Krstić delivers a technical presentation of OLPC at the Google TechTalk series
- 60 Minutes, What if Every Child had a Laptop [5]
- CNN, Should Intel Fear $100 Laptop? [6]
- Red Hat Magazine: Inside One Laptop per Child, Episode Three
- Red Hat Magazine: Inside One Laptop per Child, Episode Two
- Red Hat Magazine: Inside One Laptop per Child, Episode One
- OLPC Video from Switzerland, 26.01.2007
- Interview with Nicholas Negroponte on the &100 Laptop
- Presentation by Jim Gettys at FOSDEM 2007
- GLOBO- BRASIL: Crianças testam computador portátil/ Students test the laptop
- Mark Foster delivers presentation to Stanford University
- Technology Review Mini-Documentary
- A Brief Demo