Re: Haiti OLPCs - some news
Hello On Thu, Nov 26, 2009 at 04:55, Edward Cherlin echer...@gmail.com wrote: On the data entry, what medical software is in use? There's no data entry for the moment. The first part of the project is to provide access to html content, with a strong focus on drugs information, but anything can be put on the SD or USB key. The olpc are set to use a fullscreen browser with a minimal debian based distribution. Using medical entry software is an interesting idea, but I believe trying to adapt an existing solution instead of creating something based on the local needs is wasteful. Are people thinking about how to network from village to town to city for such purposes? The debian is set up so that when olpcs are near eachother, the homepage of the fullscreen browser shows the list of near olpcs, with an hyperlink. The content of the SD and USB keys are shared by default - therefore if there's no SD or USB key, it shows nothing. I purchased SD card with USB interface (sandisk) for each olpc. The idea is to let it inside the olpc, using the USB port to access its content on another computer if needed. SD are small enough to be easily exchanged The idea is to ease the propagation of any html content (the LCD of text information) between the users. Is there any thought about telemedicine using the camera, video chat, and data acquisition functions of the XO? Hopefully - I did some tests, aiming for something simple (saving jpg files on the USB/SD, to be shared wirelessly). But for now I want to get the devices in the hard of the users and let them tell me what they need. -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Haiti OLPCs - some news
Hello all, Long time no see. So I've got news to all y'all volunteers who helped me raise dead XO to the Vodoo of Haiti :-) The machine are safely stored in vice director locker of the University Hospital of Martinique. First tests and OS (non sugar) - for now, work as expected. 6 terrain units were deployed in the hospital. They performed as planned (but wifi mesh - last minute mistake I made on a script) A group of infectiologists and administrators flew to Haiti 2 weeks ago. There, demos were given to various health programs, mostly focused around AIDS. I was told that many people loved the hardware- more so that my software, but never mind :-) They were glad to see a cheap machine suited to their terrain. And I'm always glad to give a hand. My project was about giving drug information to primary practitioners. But apparently, there's a strong interests in simply using the devices as data entries for outpatient clinics. It's an evolving process and I'm open to anything that can help people, as long as the units are not diverted into some obscure research program with very little benefits to the sick people. Soon, we should receive a visit of their team, (I will do my best to make pictures - I'm not in PR :-) during which I'll offer to adapt for free their existing software, if they can also put some of the XO loaded with drugs prescriptions, interactions, etc. to primary care doctors. Then finally the XO will be at their new home in Haiti ! -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Bootloader question
Hello, On Tue, Jun 2, 2009 at 22:12, Mitch Bradley w...@laptop.org wrote: Don't believe everything you read on a wiki. OFW has included support for partitioned NAND since the first production shipments, dating back to January 2008. The idea is to have a small boot partition that can be in any format that OFW supports - JFFS2, ext2, FAT, or even a .zip archive. JFFS2 is just fine on a small partition; the scan time for a few MB is negligeable. I have been lobbying for such a structure for about 2 years now, but never managed to get enough traction among the OS people to actually implement it in the XO software distribution. The OFW support for this is known to work, as debxo uses it. I'm sorry, I didn't know that and I didn't want to imply OFW was not the right tool for the job. I'm just very concerned by the current time it takes for a vanilla olpc to be ready, especially when compared to any netbook running moblin, so I'm exploring various ways to fix the problem (actually rereading every documentation that was send to me explaining various aspect of the OFW before starting the UBIFS tests, but I have limited time and I'd like to spend in on the UI rather than on the boot process) What would you suggest to have the kernel loaded in ram as quickly as possible? (I'd guess execute in place, but I think that's not possible) A fat partition with the zimage ? I'd also like to remove the initrd to try to shave some seconds. I don't need any antitheft protection, I just want to protect the nand against a reflash with a non approved software image, which IMHO is the most interesting feature of OFW. But if that's too complicated/requires the initrd or some weird other stuff, I'll scrap that too. BTW, could you point me to some documentation explaining how to have OFW immediately boot a kernel, without fancy sound/screen/counter? (only prioritising USB or MMC, so that it boots first if a media is inserted or if a struck esc key is detected it gives a command prompt) I'm open to any additional suggestions. (I'll consider the sysvinit optimisations later) Thanks -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Bootloader question
Hello I'm still preparing my custom images for the Haïti project, and I am quite disturbed by the JFFS2 boottime. From what I've read on the wiki, JFFS2 is here only because OFW doesn't know how to use UBIFS. This brings a question - is it possible to replace OFW with something that could use UBIFS? Say coreboot , or even a bios with grub, anything will do! If there's no security, if there's little functionality, not field upgrades etc, it will just be fine as long as it can boot any quicker. I just can't keep the boot delays currently experienced with jffs2 -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: VGA questions
Hello On Sun, May 17, 2009 at 20:54, Bobby Powers bobbypow...@gmail.com wrote: on a B4 (which just needed the VGA connector, no other components), Same here. I have a spare B4 board for these tests (I would like to reproduce the identical picture of the XO on a projector for a formation - and I'm talking about openfirmare and all) I was able to drive full screen video on a large LCD from the XO, so its not useless That's just what I want to do. No internal LCD mirroring - plain VGA output only So far I've been stopped by : - the video routing question (CN18) - the plug. I would like to avoid soldering wires to a VGA plug. Soldering the plug should take less than 5 minutes once I get the correct type from digikey/mouser If you are still interested, you could start with the following references (although I didn't see any part numbers there). Already know them :-/ -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
VGA questions
Hello A boards had CN18 to select the output between VGA and LCD. It is no longer presnt on the boards. How can it be implemented? (strip wire, software, ... ?) For the VGA plug, I'd also be interested in a P/N to get one from mouser, mousekey etc. Thanks -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Custom firmware with a different bitfrost key
Hello For a project I'm working on, I need to sign images. However, these will be custom images, not olpc official ones - so I am wondering how I could sign them myself, creating a custom certificate for this. From what I understand, it means changing the keys in the firmware. What do I need to do then to sign my images? Any pointer is appreciated. I'll add my experience to the firmware wiki page. -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Xo 1.5 - some thoughs
Hello I am very happy to see that 1.5 will hopefully bring the price per unit down. We'll see if it can really break the $100 barrier. Yet if the plastic case moulds are to be redesigned I wonder if it could also be the chance to fix some of the most annoying issues we are all experiencing : - the SD reader should be placed *anywhere* but where it's now. I understand that it was a late addition and it couldn't go anywhere else because the plastic mould would then have had to be redesigned (a costly operation), but please take advantage of the current evolution to put it say near the USB ports on in the bottom case, below the trackpad. Developpers will bless such a change! - since the wifi chip is going to be replaced, hopefully it could come with a free wifi firmware this time, or at least a thin firmware could be available as an option for when 802.1s is neither possible nor desirable. Having the possibily to run one olpc in AP mode, and the others in client mode would IMHO be simpler and better because it just works fine. No need for fancy untested and non-free firmware. - the internal serial ports (J1 and CN24) pins should be exposed in the battery compartment. From what I've read, the wifi chip will be field replacable without having to discard the whole motherboard - this is a good idea. Apply the same logic to the serial port! It is simply a huge waste of time to have to remove everything to plug a cable to unbrick an olpc with RTC problems, or perform hardware diagnosis. I'm not sure how much it would cost, but even a hole in the plastic to solder wire leads would be better than the current setup. Additionally, some weak points could also be fixed: - if it is possible, the VGA pins should also be exposed in a similar way, so that an OLPC with a damaged screen could be turned into a fixed computer with only some soldering. No need for edid or other fancy stuff - just add the resistors and some holes to solder a VGA port or even just wires. People who will convert an OLPC to a fixed computer will know what they are doing. We all have plenty of screens around - good old VGA CRT which will not cost anything, compared to the costs of changing the display. - current speakers produce horrible sound. Would NXT speakers of the same size be too expansive to add? I tested that, it works just great - a bit low on the bass, but we aren't talking about a mp3 playback system. Just something decent enough. - could the membrane keys be closer to each-other ? Learning to type is important - the current keyboard is far too different from other keyboards. Do not believe children have never been exposed to a traditional keyboard - if only at a cybercafe to read/send email. I also think the keyboard should be more like a standard keyboard (ctrl-cap). I'm saying that while I personally enjoy very much the ctrl key where it is ! But that's because I'm a hacker. We shouldn't force our choices to children who first need to learn how to type. This redesign may not be desirable however, because it would introduce a serious difference in the OLPC user interfaces - yet I think it should be mentionned. -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Recompiling the firmware
Hello I couldn't find a lot of informations about firmware recompilation. From what I've read, I think I can get it from http://openbios.org/viewvc/?root=OpenFirmware But how do I recompile it, to get a new .rom file ready to by flashed ? (the same question appears on Firmware_hobby_hacking on the wiki, without any reply) Do I need specific tools ? I have the developper key for the machine I would like to experiment on. Thanks in advance for any information -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: running a motherboard without a keyboard
A similar question about dcon. From dmesg, it looks just fine. $ dmesg|grep dcon [ 26.481778] olpc-dcon: No DCON found on SMBus [ 30.651493] olpc-dcon: Discovered DCON version 2 [ 30.708411] PM: Adding info for platform:dcon But apparently dcon sleep, output and freeze also refuse to change. Is the lack of screen causing the problem, and how? The brightness can be disabled however. $ cat /sys/devices/platform/dcon/sleep 0 $ echo 1 /sys/devices/platform/dcon/sleep $ cat /sys/devices/platform/dcon/sleep 0 $ cat /sys/class/backlight/dcon-bl/brightness 15 $ echo 0 /sys/class/backlight/dcon-bl/brightness $ cat /sys/class/backlight/dcon-bl/brightness 0 The only guess I have is i2c_smbus_write_word_data , called as dcon_write from dcon_sleep is not working as it should. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Turning off the whole video susbsystem while keeping the cpu on
Hello I am wondering if there is currently any possbility to completely turn off the whole video subsystem, while leaving the cpu running? [and is it also possible to reclaim the video RAM for the system BTW ?] That's for a screenless OLPC I'm experimenting with, and for which there's no need to power the non-existent screen :-) So I'm curious to try ; according to http://wiki.laptop.org/go/Hardware_Power_Domains, I guess it should also be possible to power down the DAC, video misc register and graphic processor. How can it be done? Is there an entry in /sys? So far even dcon refuses to go into sleep : echo 1 /sys/devices/platform/dcon # sleep $ cat /sys/devices/platform/dcon/sleep 0 The ec buttons are now working fine however. Another question : what is the status of ec-wakeup, to replace rtc-wakeup? (http://dev.laptop.org/ticket/4606) Thanks Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: running a motherboard without a keyboard
Hello It works [ 4300.811798] drivers/input/serio/i8042.c: 69 - i8042 (interrupt, 0, 1) [427524] [ 4300.812015] drivers/input/serio/i8042.c: f2 - i8042 (kbd-data) [427524] [ 4300.937465] drivers/input/serio/i8042.c: fc - i8042 (interrupt, 0, 1) [427537] [ 4301.011188] drivers/input/serio/i8042.c: ed - i8042 (kbd-data) [427544] [ 4301.107998] drivers/input/serio/i8042.c: e9 - i8042 (interrupt, 0, 1) [427554] [ 4301.134320] drivers/input/serio/i8042.c: fc - i8042 (interrupt, 0, 1) [42755 The IRQ1 counts are also progressing. Only /dev/input/event3 is not bound to anything, and setkeycode logically fails On Tue, Nov 25, 2008 at 8:23 PM, Richard A. Smith [EMAIL PROTECTED] wrote: Offhand I don't know what the kernel does when you boot and nothing responds to the keyboard reset. Maybe it requires a tweak in i8042 module to give key even even if keyboard resent doesn't respond? Something like a module parameter ? Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
running a motherboard without a keyboard
Hello After my (mostly) successfull B4 to C2 motherboard surgery, I am left with a working B4 motherboard that I would like to use for various tests purposes, including kernel stability tests and perfs optimisation. It is a bare motherboard- no display, no keyboard, no battery. Just a power adapter to give it 12V, and a 256Mb SD with a minimal debian to give me a network connection I can use to access the nand when I do some mistake, to investigate what happened. I'm thinking about getting a USB-serial cable to connect to J1 ; meanwhile, I figured out I could use the gamepad, rotation button and the 4 game keys, along with blinking the leds, but looks like they are not working, while the power button is working fine. Is it due to the missing keyboard part? Is there a quick way to fix it? I would simply need to map a couple of buttons for basic functions like wifi up/down (access point connection is unstable), and to start/stop a test script. Thanks Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Audio routing question
Hello Whenever a jack is plugged in, the audio is routed to the jack instead of the main speakers. It may not always be desirable (ex: wakeup clock) Did try to tweak alsa settings, but I couldn't change that. Is there a way to override it? Thanks Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Any cure for a washed out keyboard
I haven't had a soaked olpc (yet), but most of the other devices that hand these kind of problems where much simply cured by : - being taken apart - carefull cleaning with a cloath, especially for the tip of flat cable going to a FPC connector which some dirt (oxidation? short circuit? isolant?) usually accumulate. Saved a wirless phone (display FPC problem), and a smartphone (keyboard FPC problem) that way. If that doesn't work, yes you should certainly try washing the keyboard board and plastic membrane. On Thu, Jul 31, 2008 at 7:40 AM, [EMAIL PROTECTED] wrote: Hi, the best thing you could do is to put the device (keyboard) into a bathtub with distilled water. No joke! after one or two days the electrolyte ingredients will be washed out. After that drip of the water and be patient one or two days. The device has to be very dry before you should activate the device. Maybe the display do not like it. I am always treating sunken electronic devices that way, including still cameras. A good alternative is ethanol (but not denatured alcohol!). Best regards, yokoy On Thu, 31 Jul 2008 12:44:08 +0545 Bryan Berry [EMAIL PROTECTED] wrote: The first XO casualty at Nepal's pilot schools a few days ago. A second grader washed his XO because it had gotten too dirty. Thankfully, the display, cpu and motherboard seem to be working fine. The keyboard is non-functional and the mouse is nominally functional. Anyone know a fix for a washed out keyboard besides complete replacement? -- Bryan W. Berry Systems Engineer OLE Nepal, http://www.olenepal.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 -- Dr. Guylhem Aznar, MD PhD Unité d'Analyse Médico-Économique Service de Santé Publique et d'Économie de la Santé Pôle SPSSR CHU de Fort de France BP 632 97261 Fort De France Cedex Martinique, France Tel : 05 96 55 23 47 Fax : 05 96 75 84 57 ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: [RFC] Four solutions to NAND fillup
Hello A suggestion for similar problems, which I experienced in the past for other hardware. The /var tree is mostly used for logs and caches - stuff that could be discarded at reboot. And usually, there's a lof ot them (see with du -ksh) There are some important subdirs that however should be kept. What I did : /var is a link to some directory mounted as shmfs (there are various ones, take the one you prefer) when no shm is mounted, this directory contain a skeleton /var to keep some apps happy during a reboot or in case something bad happen, like removing kernel shmfs support as soon as the first script is run and a shmfs is mounted, the skeleton /var is untarred. takes 1 second important subdirs that should be kept (you decide which) are links to normal locations. I use /srv. The symlinks are in the var skeleton tarball Of course the solution can be simplified and improved, like by keeping /var intact and instead using symlinks and a skeleton tarball for stuff you know you want to discard at reboot (/var/log...) but this approach forces me to consider each situation individually. By default whatever is not a symlink to a stable location (/srv) will automatically be discarded on next reboot. This may not be very pretty, but it is quite usefull. This provides room for fixing the situation because a simple reboot clears the log, giving enough space to at least run some delete commands. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: B4 motherboard question
Hello I wouldn't necessarily want another machine - just a motherboard from say a machine that has been returned with a broken screen, or whatever, would be fine A fully working XO would be best used for someone who hasn't one yet IMHO. Anyway, I'll drop a message Thanks Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Reactos.zip and motherboard question
Hello Today I had a little problem with my OLPC B4. Long story short - windows XP support announcement, although controversial, was good news for my project acceptance (and more importantly, financing)- which really did surprise me, but whatever. So I wanted to resume my development efforts, and decided the first step was to upgrade the firmware and image to current. I was running q2c25, which had no security enabled, and which I feared could cause problems (it did!) So I dug into the source and found q2c27 was the first version with security support - I therefore upgraded to q2c27, and tried disable-security to avoid future problems. It said no wp tag, which was consistent since there wasn't security before, but I thought it may need at least the ww tag for the current firmware (q2dxx), or it may refuse to run. I checked the source in http://www.openbios.org/viewvc/cpu/x86/pc/olpc/setwp.fth?view=markuproot=OpenFirmwarepathrev=622 and decided to run enable-security followed by disable-security. Bad move- after enable-security, it did power off, which wasn't apparent in the sourcecode I read. So I was unable to type disable-security and follow up with the upgrade because 1) it was refusing to boot the old image I was using and 2) after pressing the X game key, the esc key was ignored during the countdown. However, by pure luck I guess, I was able to issue a disable-security command while trying to run Actos.zip/Runos.zip to get the UUID and apply for a dev key to fix the mess I made : after booting on a usb key containing both files and pressing the X game key, I got a working ok prompt. I don't know if that is a bug or a feature, so I thought I should report it to the list before reporting it as a bug. ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
B4 motherboard question
Hello I was wondering if I could return my motherboard to get a current mass-production one without the documented hardware bugs : I fear I may not be able to perform all the fixes required (http://wiki.laptop.org/go/B4_Suspend_ECR), while I would need suspend support. I guess the current motherboard will also support Windows XP, which would be a good thing. Alternatively, I am now located in Martinique (the island next to Dominica) : if there's anyone from the project in the Caribbean who could give a hand with the hardware fix, I could certainly carry my B4 over there. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
sign language update
Hello Some time ago I did post a mail about sign language work. I just wanted to let interested parties know that I have finished redacting my thesis, and that I am trying to have someone work on an implementation for at least proper fingerspelling support. If you have some free time, know how to edit fonts and an interest in experimenting with sign languages, please let me know by a private reply. I will be happy to provide guidance to offer a proper sign language support following a Unicode proposal. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Anybody do BGA rework?
Hello, Been there, done that, for different devices - mostly PDAs. First you *must* check the specs : optimism won't get you far if the pins have different fonctions (like they have been swapped) or require some minor adaptations. Get the reference sheets for each device, read them and understand them. If you still believe it should work, then comes the actual physical work of reballing, etc: normally you would use a BGA rework station, but that ain't cheap. However, you can do that with easy to find and unexpansive tools. There is a nice visual guide with pictures and explainations on : http://digit.que.ne.jp/visit/index.cgi?Linux%a5%b6%a5%a6%a5%eb%a5%b9%b3%ab%c8%af%a5%e1%a5%e2%2f%a5%cf%a1%bc%a5%c9%a5%a6%a5%a7%a5%a2%2fC700%a5%e1%a5%e2%a5%ea%c1%fd%c0%df You can use excite.co.jp if you need translations, but the pictures alone are quite evident. Train yourself with dead hardware from ebay - you make many mistakes the first time :-) Be sure to have a soldering iron with the thinnest tip you can find, stripping wire, and a hotglue gun. It will come handy should you damage the motherboard by accident (it can happen; cutting a track, etc). If you need more information/help, tell me. I'm not sure I can help you (it's been a while since my last serious hardware hack) but I'd be happy to point you to people who can. Guylhem On 9/14/07, Albert Cahalan [EMAIL PROTECTED] wrote: I'm wondering if one could replace the 88W8385 on a gumstix board with an 88W8388 swiped from an Xbox 360 wireless device. They appear to be the same chip package. This of course does not ensure that enough of the connections are compatible, but I think there is reason to be optimistic. It's this horrifying package: TBGA 132pin 8x8x1 (any better ideas for getting 88W8388 hardware?) ___ 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: accessibilities first tests - many questions
Hello On 8/20/07, Jordan Crouse [EMAIL PROTECTED] wrote: It'd be great if this could be included. Better yet would be to allow specifying the raw register value, of course with an -EINVAL if bits unrelated to swizzle and backlight are set. Again - can I ask why? The sysfs/ interface exists to provide the right interface to the applications and the user to accomplish what they want to do. If you have a good reason for exposing this functionality, then I'm all ears, but I think that just for giggles doesn't quite cut it. What about because it has not been tested? In the end, I guess it depends on how you design your software. Personally, I write quick and dirty test code where I do my best to keep every possibility open. Then I work around some specific parts. Sometimes, I even avoid includes to have some nearly identical version and play with tiny parts of the code (which way is faster, how does it react to tests with real data, etc). If I can't get any measurable difference, I consider running benchmarks ten thousands of loops, etc. But unless some feature has been *tested* (not considered) to be useless, I keep it at that step. When I remove something, that's for a serious, documented reason. Then comes optimisation time, and I decide what makes it in, and what doesnt, based on the data gained on the previous tests. Lot of stuff get removed then, and some parts are even fully rewritten and fully tested again. Removing a feature that people want to test, doing someway instead of the other way, just because you are guessing it won't be helpful, is just wrong to me. That's Premature optimisation, design by commitee - name it how you want. I only trust tests, data and conclusions. Likewise, I have a friend who is a damn good coder - far better than I am. However, he has this tendancy to ignore tests and error checks, write evrything in one shot, and then spend some serious time wondering why it isn't working as it should. Same mistake IMHO. If you want to write directly on the device for testing purposes, then the i2c-tools work great - you can bang on the registers all day. But you are making that unreasonably complex. What about other features? Will everyone will have to do i2c? What about switching GPIOs? (I haven't checked that yet) An echo 0/echo 1 in /sys really saves testing time. A real life example with the XO - did you try the script I posted with black and white + backlight mode? Some people I showed it to actually enjoyed it, not for the high contrast mode as I initially supposed (bw with full backlight), but for reading ebooks at night (!!!) - they said it was easier on the eyes, at least according to them, than the color mode. The pale backlight made it suitable to be used with the lights off. Do you want to actually prevent creative uses? Something even more funny - someone turned the screen to pda mode, but did not close the keyboard all the way, so that it made the screen stand at a strange angle suited for standing on a desktop, like a picture holder. How clever users are - they think in ways we don't. So we shouldn't be arrogant and try to dictate them what's best for them, but see what they do with the possibilites we provide. Therefore, I think we should not currently be removing possibilites but adding them instead, test them, and remove what has been *proved* to be useless. Wild guessing is not a good strategy. And anyway, what's the cost ? That's won't make your kernel bigger. That won't make it run slower or eat more power. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: some first impressions
Hello, On 8/18/07, Yuan Chao [EMAIL PROTECTED] wrote: I've tried one SD card made by a local company. It's a bit thinker than the genuine panasonic one. So this results in that it can only be ejected out a bit and I have to use my finger nail to grab it out. (probably same situation as you) Usually there's a groove at the tail of an SD card top side. However, the access on XO SD slot is only on the bottom side... :( Currently I have a 4 Gb SD (plain SD, no SDHC) and after comparing it to another SD, yes it's thicker Jim said the SD port was not for removable media use - I fully understand since I put the SD in to have storage space to try to set up stuff, and anyway USB keys are more practical But then the current slot is not fulfilling either goals. It's hard to use, and still open for the water to get it. Maybe the plastic could be shut tight? Tried FBReader on XO. It needs some fine tuning for XO's special resolution and works well. It may not be a bad idea that FBReader can be merged into XO as the current Read activity is mainly for PDF. The browser can be used as offline reader but still too heavy and not optimized for book reading. I am preparing an opie-reader for people to try, along with some sample ebooks of mine. IMHO it's far better than fbreader. Though they are currently working on the search result display. They tried to parsed the xml in their engine for better performance in stead of the whole LAMP but not complete yet. You can try the above link. There was an interesting article on slashdot this week about offline wikipedia. It currently is 2.9 Gb, I am sure with better compression it could be made to fit on a 2 Gb USB stick. But that would only be usefull for englsh speaking countries. It's best to work on ebook reading than on the wikipedia IMHO. The picture book looks best to me when rotated in 90 deg. It's just need some optimization on the page layout (remove menu frame) and some javascript for hotkey to flip pages. Or maybe FBReader serves better here? I am currently evaluating different tools (back at home with full net access and wifi !) As soon as I will have found a good setup, I will let you know Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: accessibilities first tests - many questions
Hello, On 8/18/07, Jordan Crouse [EMAIL PROTECTED] wrote: We didn't enable this ability in the sysfs/ interface. I have never been too clear on what the actual practical uses are for something like this, so the control never got added. In a pinch, you can use the i2c-tools utilities to write to the device directly (use at your own risk!) I'd prefer not touching the kernel yet and concentrating on the software side. If you have a chance to add this to sysfs, I will test it. Basically, I would like to compare the screen readability in as many different modes as possible, letting people freely switch between display modes in different situations so they can chose the one they best like. It's just an experiment - I would like to have data proving users actually prefer using the display when the algorithm is enabled. Sometimes you have surprising results. This is difficult to do - since it would involve synchronizing with the video driver which with X and the framebuffer driver will invariably result in a screen glitch (note that just switching the rate on the DCON itself doesn't cause a glitch - its the software that is braindead here). But we don't have any support for this in the kernel. Hmm - that I won't touch at all. If you plan to add support sometime in the future, I think it could be quite handy to let applications lower the refresh rate at will (ex: when switching pages in ebook mode, a lower refresh rate would save power, while during a page read, the dcon display mode could be used) I have done some shell scripts to test my stuff (ugly but handy, esp Here it is BTW. Thats because the DCON driver does the freeze on its own while the system is suspending, and it restores it long before userspace gets unfrozen, so from your perspective, it will always be 0. I thought about it but I wasn't 100% certain. Ok, at least that is sure :-) Regarding the zoom/overlay, I will see what can be done in software. Guylhem contrast.sh Description: Bourne shell script ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Understanding the power management
Hello I am trying to figure out more about how power management work on the OLPC. It's quite different from what I am used (acpi, etc) and the documentation/scripts doesn't seem up to date For example, I saw references to respeclaration in the xinitrc while it's supposed to be unused and replaced Here is my current understanding, based on the documentation I've read and some tests. Please correct me if I'm wrong. - ohmd performs power management tasks when ordered to do so by the hal on dbus but killing it does nothing bad (??) so which purpose does it serves? - respeclaration has been replaced by the following tools to ask for pm tasks execution : - hald-addon-inpt - olpc-hardware-manager - how are buttons mapped to pm tasks ? hald-addon-input apparently does rotation/suspend. I am wondering about the others functions. - what is the role of olpc-hardware-manager? Execure ohmd dcon requests ? Or manage ohmd? killing it stops screen brightness, does not stop screen rotation dbus-daemon --system : is where they are talking console-kit-daemon : serves which purpose ?? I'm downloading a FC7 iso ATM to be able to fetch the git tree, (not git on macports :-/) because the documentation doesn't help a lot. Reading the code might be much more useful. Meanwhile, I am still missing 2 things on power management - how to turn off display completely on standby (I know how to do a freeze, now I would like to turn it off) - how to turn off wifi (flight mode) so that I can work on the olpc next week. My day job is in an hospital where wifi is *not* allowed. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
accessibilities first tests - many questions
Hello, As explained before, I started playing with the DCON (displaying test images in X, changing various parameters, etc) and other stuff. The results are interesting. For example, a black and white mode will backlight on, even if it reduces the percieved resolution, could be usefull to read at night, or for people with sigh problems who need high contrast. I am exploring other similar options. The only things I don't know yet how to do with the DCON : - how to disable the smoothing algorithm applied in color mode - how to reduce the framerate (for ex for ebook reading, but it could also be handy in text mode) I have done some shell scripts to test my stuff (ugly but handy, esp when you are only testing ideas) ; I will post them here when I will get back home for feedback. If someone can help with the smoothing/framerate, some quickdirty bash code to do it would help. Regarding power management, I have a problem with the DCON freeze before suspend to ram: the display looks like frozen, but when I query the freeze file just before and right after the suspend, I only get 0 while I should get 1. Can I also ask for some help there? Regarding the X being used, I am curious to know if there is a way to do live screen scaling (zoom function, where the whole screen is magnified) ? Ideally, it would be hardware managed, but that could also be done by software. See this if you don't know this function: http://www.youtube.com/watch?v=LFQAqxoKyAE The idea would be to do as fast as possible for moving around, then when stopped on something, try to be as detailed as possible: http://lists.freedesktop.org/pipermail/xorg/2005-April/007640.html I am also wondering where/how the cursors are set. Matchbox theme is only changing some cursors. I found a sugar gtk theme fixing additional cursors as well, but it is not clear how everything interacts. That's because I am looking for a way to magnify the mouse cursor when a special keyboard sequence is pressed, to reveal where it is on the screen, along with a way of magnifying the whole screen to ease reading of content (ex: text from the internet, a small picture...) Finally, how can I interact in pen mode with the whole pad? Jim said it was possible, so I tried to run a cat on various /dev input devices, while moving my fingers on the left and right part around the zone used for the pointer. However, I can't generate any data. What's the problem ? Side question - is is possible to get data when two objects (ex: 2 fingers) are moving ? (either in the same surface, or one in the pen area mode and the other one in the mouse mode) Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: Power manager specification... (request for comments).
Hello, On 8/17/07, Walter Bender [EMAIL PROTECTED] wrote: Lets please be careful not to over-engineer. While Mike makes good points, we have this wonderful human social network we can depend upon as well. E.g., If I am downloading something from your machine, I can ask you to hold on a second until I finish. Let's take advantage of the fact that the kids are in the same community/school most of the time and not worry so much about corner cases until we have some more breathing room. Yet thinking before implementing can easily overcome future problems. I believethe idea of inibitors for the various power schemes should not be overlooked since their benefits can be important. In your example, a download activity could make the suspend wait an additional minute or two, explaining the user than its request was noticed, won't happen until the download/upload is over, unless it is overriden. If people are in the same class, of course, but what if the person is several hops away on the mesh network? Moreover, this interesting idea could also be applied to video playback/screen rotation requests, explaining that the screen can't be rotated or the playback will stop, etc. There's a great potential in such examples to go beyond the traditionnal power management done in GNU/linux. But anyway, if you think these cases are so special and supporting them will take too much time, write a quick shell script to test the concepts, play with it, and see if it helps you or if it's just a waste of cpu cycles. PS I have some more suggestions (ex: a maximal suspend mode to carry the machine without using it) but on a computer I don't have here - I will post a message a little bit later. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Re: some first impressions
On 8/11/07, Jameson Chema Quinn [EMAIL PROTECTED] wrote: I've archived this discussion on robotics/LED output, with some points of my own, on the wiki at http://wiki.laptop.org/go/Electrical_output. Thanks a lot for the summary on the wiki. A quick suggestion : if there is a serial port in the XO (according to the boot message, there is one) maybe it could also be used. rx/tx/gnd/vcc already allow a lot of fun stuff and experimentation. By the way, if there are other beginners out there, who are a bit lost in the wiki, http://wiki.laptop.org/go/Getting_started_programming is a great page to get started. There is too much information to read everything in one day, but so far I have found the following documents very interesting: (I'm new to python, but I know a little javascript) http://wiki.laptop.org/go/Understanding_sugar_code http://www.pygtk.org/dist/pygtk2-tut.pdf http://www.ibm.com/developerworks/linux/library/l-sugar-olpc/index.html http://downloads.egenix.com/python/LSM2005-Developing-Unicode-aware-applications-in-Python.pdf http://wiki.laptop.org/go/DCON This page however looks incomplete : turning the backlight back is impossible with the current instructions: -bash-3.2# cat /sys/class/backlight/dcon-bl/power 0 -bash-3.2# echo 1 /sys/class/backlight/dcon-bl/power -bash-3.2# cat /sys/class/backlight/dcon-bl/power 0 I guess I will have to look for more information since the sugar gui does it just fine. Regarding the underlying gnu/linux system, to explore how it all works together, my preferred method actually is ssh and console mode. To do that, type alt+m to get a terminal, then passwd root and passwd olpc to setup passwords for remote access. This also let you use the console on ctrl-alt-F1 (magnifier) or with chvt 1 Currently I am exploring a little more the fedora distribution, the hardware support, and I am doing some benchmarking, especially for power management stuff. Last night I found that closing the screen turns off networking, while keeping the mesh network (as specified), but does that until the battery is fully depleted. Wow. I wonder if a graceful suspend when there is say only 50% of battery power left wouldn't be nicer to the users. Else, kids who come to the same conclusion may prefer to fully turn off the laptop to have some power left to read books/play games/whatever in the bus after school etc. There should be a good equilibrium between the user own interest (having some power left) and the community interest (keeping the mesh network up) Also I noticed SHM support has been compiled in, but /dev/shm in not used. Is it by design? I couldn't find references on the wiki. /var /tmp and similar files could certainly live there to save some nand write cycles. Did that on the zaurus : you simply keep a tarball of a clean /var in / (say /.var.tar) which you untar as soon as the shm has been mounted. Before the shm is mounted, you keep a skeleton /dev/shm/var so that /var symlink is not broken and application which may depend on the existance of the directories won't be confused. Guylhem ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
Presentation questions
Hello I have many many questions but starting with them would be rude - so I will first introduce myself :-) My name is Guylhem Aznar, I am a medical resident doing a PhD in computer science, working on computer support for a written form of sign languages. (for a good introduction on the topic, see signwriting.org). I have been using GNU/Linux since 1993, wrote various application (xiterm, etc), documentation (en.tldp.org) played with Linux on ARM devices (first with the zaurus, then letux.org), etc. For my CS PhD thesis, after having worked on the other aspects of the problem of global computer support, mostly revolving on the encoding/Unicode issues, I am now working on the display and the entry process of signwriting support, Following a real-life test of an XO at the RMLL in Amiens, France, I reoriented my work from the basic Gnome support (which isn't research work by itself) to the more interesting issues an OLPC implementation would bring (which is full of unknowns - like will the faster algorithms I proposed will be fast enough on an OLPC?) Anyway, I currently have 2 research topics : 1. how can the OLPC display signwriting text. I read some good guidelines on http://lists.laptop.org/pipermail/devel/2007-January/003817.html but I wonder how children will perform with signwriting, which is quite different from written text 2. how can the stylus area be used to help writing in signwriting. This is still an open problem, for which current methods relies on either on-screen pickup of signs or keyboard based shortcuts After this quick introduction, here are the questions I have. Spent a lot of time with google, the wiki, etc. but I still have no definitive answer, so I'm asking. Tell me if there are dumb questions, I could understand :-) 1. Is the keyboard on a B4 identical to the pictures posted on the wiki ? http://wiki.laptop.org/go/BTest-4_Release_Notes doesn't say anything. I need to know because I may have to use shortcuts until other entry methods have been completed 2. Is the touchpad working in pen mode? I'm asking because I found conflicting information : I read on page 30 of http://wiki.laptop.org/go/Image:KGDMFA001-non-confidential.pdf 10.3.2 Because GS is designed to be operated by a finger, a finger wearing a glove, a pen, a ball-point pen or pencil will not make it work. GS will not behave normally with two or more fingers on the surface or with something laying on the surface. I thought pen mode was supported? : on http://lists.laptop.org/pipermail/devel/2006-November/003069.html Jim said it wasn't possible *yet* - at least that brings hope. http://dev.laptop.org/ticket/2341 let me think it is possible now. In the end, is it? 3. Can you disable anti aliasing for the color mode ? The anti aliasing in the dcon specs is interesting; I will have to test it to check how much it may impact on the display process sice the wiki says it currently can't be turned off, while the DCON specs let me think it could be. Another open question for me. 4. Who is working on the display issues ? While studying the wiki, I noticed on the display page a mention to perception tests which had been performed, and whose results would soon be published. I wanted to run various tests before doing on-field tests with deaf children, but such information could save me some time and effort, to concentrate on the sign writing specificities. I was initially asking my dumb questions to Jim Gettys, but I figured they would fit better on the mailing list and save his time. It's a bit late here (3:30 am here in France) and I'm tired, so I may forget something - feel free to ask for more details if I haven't been clear enough I'm eager to join all y'all in this interesting project, and hopefully help making the OLPC more usable to deaf children. Thanks for any help, Guylhem PS: if any of you is in Porto Alegre, please contact me offlist ___ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel
