Lessons To Learn From The OLPC Project 261
FixedSpelling writes "Whether you're impressed with it or not, the XO-1 could have a major impact on notebook design. The concept behind the OLPC's development brings outside-the-box thinking and cost-consciousness to a level that we rarely see in portable computing. There are a number of lessons that can be learned the from its unique design and we can already see that some of these concepts have been noticed by manufacturers. 'The biggest attraction to the OLPC project has always been the price of the system. You don't have to be a cynic to understand that the impact of a $100 notebook could be huge and the price has generated the majority of the interest in the project. Notebooks break, they get lost, and they are replaced frequently, so the cheaper, the better. The low price was originally important so that the XO-1 could be produced in large quantities without putting too much of a burden on the buyer but the low cost appeals to everyone.'"
Re:first post (Score:2, Informative)
Well if you don't want to RTFA, or search google, or wikipedia: Its One Laptop Per Child http://en.wikipedia.org/wiki/Olpc [wikipedia.org]
Re:How do you get one of these? (Score:5, Informative)
Re:How do you get one of these? (Score:2, Informative)
Re:Crank it up (Score:3, Informative)
here ya go (Score:5, Informative)
All of this could be ripped off... (Score:2, Informative)
OLPC espouses five core principles: (1) child ownership; (2) low ages; (3) saturation; (4) connection; and ***(5) free and open source.***
Someone else can run with option 5, to make an equivalent, for adults laptop. Depending on performance, we may finally see a machine mass-produced, showing acceptable speed and avertising that it's doing so despite "under-powered" hardware.
If this was mass-produced, people would finally have reason to question: why do I need this super-great/expensive machine for the latest OS? Sure we have plenty of tiny OSes out there, Puppy Linux, D*** Small Linux and various others from scratch. The problem is the same that kept Linux from the spotlight... it's not pre-installed on PCs sitting on store shelves.
(Sure the above efficiency question is asked frequently from one version of Windows to the next, but default installs of Linux flavors trying to be mainstream-ready are a bit slow on older hardware as well.)
I can't wait to see the results on the marketplace...
Re:How do you get one of these? (Score:5, Informative)
Without any other committing statements contradicting this, I take it to mean that if 3,000 people sign up, they'll send out 3,000 very overpriced XO's to those who order, and the poor kids get no machines.
And no mention about who is going to pay for the infrastructure needed for the machines either, if they reach the 5,000 goal. Not only do they need a support apparatus, but the machines themselves need electricity (the crank never came out, and the other battery charging implements are still not in production, if they ever will be) and Internet (the applications on the XO are leased and have to be renewed over Internet every so often). Just handing out machines like it was bread won't do people any good.
Re:my favorite lesson (Score:3, Informative)
Of course, the main problem is that to own one as a US citizen, you apparently need to pay more like $400 [olpcnews.com].
And for $400, you can get a nicer laptop online or even at your local walmart [walmart.com].
Wake me up again when I could actually buy them for a non-profit charter public charter school for $100, or even $150 each.
Going to a conference (Score:3, Informative)
My local LUG is having a large conference tomorrow, where one of the highlights is an introduction to programming on the OLPC.
At least in Argentina, where a deployment is being scheduled, the entire Free Software community has the hots for this. Whether it succeeds or not as en educational tool, it's pioneering a new paradigm of computing; the truly small, truly cheap, truly rugged laptop.
Re:Crank it up (Score:4, Informative)
And the pullstring is better than a crank because you can put your foot through it and just keep tapping while you work.
OLPC system images (Score:4, Informative)
You can emulate most of it with qemu or vmware. It's easy.
See: http://wiki.laptop.org/go/Emulating_the_XO/Quick_Start [laptop.org]
Seemed a pretty sluggish on my wimpy Core Duo 1.66, but lots of that may be due to a lack of hardware accelerated video in qemu.
Anyhow, check it out. Good times.
(It does seem odd to use Python as the primary language on a slow CPU with little memory, but it seems to work okay...)
Simple hardware can do lots of stuff. (Score:3, Informative)
http://www.youtube.com/watch?v=0cxbstn2IZM [youtube.com]
Sure that one doesn't cover web usage, but without stupid Flash it would probably do decent in that area aswell.. Some more ram usage and so on is ok but the current rate and state are ridiculous.
I've used claris works on mac classics and LC IIs myself and I prefered it to works on win 3.11 on the 386s, and for most peoples use it would be sufficient today aswell..
Re:Serious question: (Score:4, Informative)
NO, YES, and NO respectively.
ARM and SH are both very low power, but that is entirely at the expense of performance. A good trade-off for embedded systems that don't need much processing power, but certainly not for multimedia applications. As soon as you start trying to do floating point calculations, watch your ARM/SH CPU grind to a halt. They certainly do less per clock than even older x86 CPUs, and are a long way behind fairly modern x86 CPUs like the Geode. That goes double for Intel's clock-inflated XScale CPUs, pushing 1GHz.
Re:Maybe they should have patented it all (Score:5, Informative)
Re:computers in education, smalltalk (Score:2, Informative)
Re:Serious question: (Score:3, Informative)
Wrong, wrong and wrong.
SH isn't much of a performer, but ARM certainly is - especially anything from ARM11 and beyond. The AMD Geode chip in the OLPC is 433MHz and single issue. Pretty much all modern low power ARM chips are at least that clock speed.
So, compare what they do with that clock speed? If you were to compare general/integer computation between even an old ARM9 and the Geode, an ARM9 will beat it clock-for-clock. Modern desktop x86 chips are only fast because they do a ton of translation from x86 code and shove it down multiple pipelines. Geode only has one execution pipeline.
We're also talking about a VERY large difference in power consumption per "MIP" here too. From the Geode datasheet, it will be chewing nearly 1W of core power at 433MHz (650mA, 1.45V core). A comparable performance ARM would be more like 250mW-500mW.
Floating point? We're not in the dark ages any more in the embedded community. For starters: WE DON'T NEED IT. Emulation suffices pretty damn well for every day usage. It's not like Excel requires 8GFLOPS to recalculate 1,000 cells. These machines aren't for games either (and hell, you can do games without an FPU anyway - we managed for decades perfectly well). If you still really, really must waste a big bunch of silicon area to hardware floating point, then there's plenty of solutions out there for ARM which are 1-2 MACs per clock. That'll match the Geode easily. Oh, and also vector stuff should you want that.
Sorry, but you really aren't giving ARM (and XScale) based stuff enough credit. For the OLPC platform, it would have been ideal, and would have performed equally well if not better. I suspect the real reason they chose AMD was fear of a non-x86 platform and not "fitness for purpose" at all.
BitFrost (Score:2, Informative)
BitFrost (see http://wiki.laptop.org/go/OLPC_Bitfrost [laptop.org]) is the set of security mechanisms present in the OLPC.
Though I certainly wouldn't care to summarize the entire thing, here's what it comes down to.
User programs don't automatically get the running user's full rights. A calculator has no reason to delete your documents, so why should it be able to? And without your knowledge to boot. On the OLPCs, documents are kept in a special storage area. It isn't a matter of owner read access. In general, for a program to get a user's file poofed in to its chroot sandbox, it has to ask the document service (which presents a consistent dialog). Further, a text editor doesn't need to access the network. The user can access the network, but his or her programs can only do so if explicitly allowed to (various such rights are set at install time, configurable later). Certain combinations of program rights are disallowed at install time (such as both network access and webcam access) but can be enabled later. Plus a lot more.
Sudo/UAC sound nice and all until you realize that programs and users are separate entities.
Yes, there's a lot to learn from the OLPC project. It's designed to be used (safely) by computer-illiterate children who can't (or can scarcely) read. If you think that sounds like a good description of computer users in general, then you're absolutely right. Security as seen in *nix and Windows makes perfect sense for protecting users from each other. That was the goal back in the day. The people with access to a server were supposed to have a general idea of what they were doing (entirely on them if they didn't), and in that case *nix security works well. But computers have gotten more personal, and that assumption is now blatantly false. Anyone thinkng that Windows security problems stop at buffer overflows, or that Linux on the desktop will change anything, is a fool.