XO motherboard close-up
Bunnie Studios (e.g. Andrew Shane Huang), who just finished his Ph.D at MIT's Project Ares group, has published an insightful review of the hardware design of the XO
The review skips over the basic disassembly and looks more at the components and hardware design itself; concluding overall that
"its mechanical design is brilliant. It’s a fairly clean-sheet redesign of traditional notebook PC mechanics around the goal of survivability, serviceability, and robustness [...] When closed up for “travel”, all the ports are covered, and the cooling system is extremely simple so it should survive in dusty and dirty environments. [...] That’s thoughtful design."Not all was happy hacker roses; though to be fair, on the hardware side of things, it basically was.The innovativeness of the design includes the oft-lauded power and heat management, where the XO really shines, such as the low-heat Geode paired with a heat spreader letting the cpu be more flexible in where it was placed in the design.
Beyond just the ruggedizing of the laptop, it's also designed to be field serviceable to a large extent, focusing on making the parts most likely to fail easy to replace, including the shock-mounted LCD (and its backlight).
Andrew found that despite the high sensitivity of the wifi antennae, he had trouble getting Sugar to find his access point and stay there. Besides, everyone knows that the only real outstanding question is, in comparison to the iPhone, will it blend?
He further was under-impressed by the software and ease of adding new tools, seeing it as "appliance-like." To an extent, however, it is an appliance, and designed to be less of the super-over-flexible Linux desktop style. Not unlike the Eee, you want to drive the new user to very safe and stable actions.
The XO does allow advanced users to install their own software (or even their own non-RedHat/Sugar OS). If anything, there's a lacking middle layer enabling easier addition of known-safe programs that aren't packaged into actions as yet:
Extreme XO screen closeup
Overall, the software on the OLPC is clever but very “appliance-like”: there are some pre-loaded applications and it’s not immediately obvious how to add new applications using the native UI (it’s hackable from the command line but that’s not very beginner-friendly).In the comments section, he returns to this topic, adding an interesting angle on the need for novelty to sustain interest:Then again, it does include some education-oriented scripting languages that kids can use to write programs, even if it does lack a local gcc installation, and it includes the basic infrastructure for chat, video, audio, and photo sharing functionalities.
I guess my strongest gut reaction is that once the novelty of the few preloaded apps wears off, what’s there to keep you coming back? [...] It feels like the pre-loaded apps on the XO are geared toward more advanced curriculum–not quite something I’d see being the focus of entry-level curricula, and I don’t see entry-level teaching staff necessarily generating new applications on the fly.Ideally, and as he discusses in his post, Squeak/eToys will provide the base for lots of mostly-easy-to-create (and share) software and tools, but it would be handy to have some more things ready out of the box that tie into curriculum needs as well as an easy, straightforward way to discover and install new tools.
I am 71 and have excellent eye sight. But for the visually impaired child that may receove one of these units and all others. I would suggest a keyboard that has greater contrast. Not every user will be touch typing.
What's new under the sky?
From day one - circa 2005, the XO has been a bunch of hardware wrapped around the shiny new screen, trying to pass off as some sort of educational platform that will revolutionize the learning process in the poorest corners of the world.
Three years later, we have a very crude prototype that costs twice the original price and doesn't deliver on most of the promised features:
1. Crank (or other) power generator. Not Delivered.
2. Battery life measured in days, not hours. Not Delivered (battery life as of today, is in the 4-5 hour range. Really poor for such s,all, limited device)
3. $100 price tag - Nt Delivered. Current price is about $200
4. Google storage for kids' work - Not Delivered. Users need to invest additional money on peripherals in order to get some storage space. The bad side to this is that it increases the price while making the laptop less portable and dimishing battery life.
5. "Contructionism" - Not Delivered - unless "constructionism" (or "constructivism", whatever Prof Negroponte calls it these days) amounts to a few pre-installed toy applications, like Tam-tam
6. Mesh Networking - Partially delivered, with lots of issues yet to be resolved.
7. Implementation plan. Not Delivered - the latest "theory' is that countries themselves are supposed to take care of it, thus rendering this into a "laptop" project, having nothing to do with education. If we accept the premise that OLPC is not responsible for curriculum or implementation, we are in essence saying to potential buyer countries: "Here, buy this nice LAPTOP and find a way to use it to improve your kids' education".
There is a lot to work to be done in order to fulfill prof. Negroponte's vision. Only time will tell, but the window of opportunity is closing rapidly, with Mary Lou Jepsen's wonderful new $75 Laptop around the corner...
1: There are pull-cord chargers and petal/cranks deployed. Just not to the american market yet.
2: Current joyride builds do provide much better battery life, and many of the improvements will make it into Update.1 This is a software issue, not a hardware issue.
3: Agreed, they cost $188, not $100. But look at american inflation over the same period.
4: Actually they are deployed with school servers with extra space, a much better solution than piping data over limited internet.
5: ? have you played with Pippy? the teachers in the deployment schools have
6: The mesh works great, it's support for WPA and (to a lesser degree) WEP that's lacking, but that's not was is going to be used in deployment countries.
7: ? The ministries of education are handling this, that's their job.
Irvan wrote:
"There is a lot to work to be done in order to fulfill prof. Negroponte's vision. Only time will tell, but the window of opportunity is closing rapidly, with Mary Lou Jepsen's wonderful new $75 Laptop around the corner..."
Irvin, we could only really appreciate your insights if you give us a full post.
But now, I can't say that your list is impressing me.
The price now is around 150 Euros for single items. That does not sound too bad.
Mesh, and other hardware, do work, although many (you?) predicted it wouldn't.
If Mary-Lou succeeds, that will be even more income from licenses for the OLPC == more laptops for children.
And all the other things were indeed well answered by the previous poster.
In total, those involved in education, computers, or software, and the children, all love the XO.
Which does make me wonder were you fit in? Did you have any involvement with these groups lately, ever?
Winter
I wrote:
"Which does make me wonder were you fit in? Did you have any involvement with these groups lately, ever?"
Rethinking my post, I fear it might be seen as too antagonistic.
My point was that most (almost all?) people "on the ground" working with the XO seem to love it, one way or another.
I am really curious what angle Irvin is approaching the OLPC to be so unconditionally negative about it?
From the children, their education, the technology, or software?
I would need some more direct involvement and "facts" than just reitterating old lists of "requirements". If those well experienced in technology and software are saying the XO is a breakthrough, what does Irvin know that we don't to contradict that? Same about education and development. Why is the XO not useful for the purpose stated?
Absence of evidence is not evidence of absence.
Winter
So will it blend? I am not sure that the XO will fit down the jug to the blade, it might need some deconstruction and breaking up first. Certainly not one to try at home. Or school for that matter!
"I guess my strongest gut reaction is that once the novelty of the few preloaded apps wears off, what’s there to keep you coming back?"
This is "to be or not to be" question for OLPC project.
Sorry Wayan, but I have a big doubt that "Squeak/eToys will provide the base for lots of mostly-easy-to-create (and share) software and tools". Why? Because it's easy to create "$100 laptop", and it is hard to create educational software.
"it would be handy to have some more things ready out of the box that tie into curriculum needs as well as an easy, straightforward way to discover and install new tools." It means - we have pretty much nothing.
Irvin is right - without proper Implementation Plan this project is doomed.
Does anyone have a plan "how to tie XO into curriculum needs"???
Irvin,
Let's wait and see how the children use the laptop as a learning tool. With internet access there are many applications that work right in the browser. Moovl, Soda Play, Flux animation, scrapblog, timeline makers, interactive activities that allow children to practice computation skills, to name a few. Servers based in the schools with internet access can open the world up to these students. Most of the applications I use in my classroom are browser based.
> once the novelty of the few preloaded apps wears off, what’s there to keep you coming back
Schoolwork?
It would be a very good thing aiming to provide universal access to a tool that just provided
* the ability to access written materials that would otherwise be more expensive to have (Each of my kids in primary school has to spend every year in books about the same that a XO costs)
* the ability to deliver spoken messages between teachers and possibly illiterate families
That alone wold be good, but the XO is capable of much more.
Grishel and Kelley are right.
The hardware is brilliant. Everybody knows it is. Bunnie tells us further exactly HOW brilliant it is. Other computer makers should be taking notes, etc. This is only half the story.
As Apple has learned and relearned over the years, great software completes great hardware. Here's hoping that OLPC's renewed focus on America will yield many more volunteers to create Activities and books for the XO. I really hope G1G1 and purchases can fund OLPC through to at least delivering a refined and optimized Linux distribution for the XO-1, XO-2, XO-3...and the EEE and Classmate.
As I've said earlier, the XO-1 platform is very much a 1.0 release. Promised features got cut, some things are only partially implemented and not everything came together. This happens with most software that has a deadline. Still, it's pretty stable and kids can learn, collaborate and play with it NOW.
What OLPC LACKS is enough structured, traditional, educational content for countries that aren't sold on constructivism. I say let it be introduced virally. Promote it, encourage it, but know that kids (and many adults) will figure out all kinds of craziness on their own anyway.
Now, if only I had mine :(
Most people consider the XO as 'a laptop' when in fact its an education tool. The XO is intended to be used in schools by hundreds of students all having access to a school server (or servers). It therefore should provide education tools in conjunction with that high capacity server.
School servers can provide locally served versions of Wikipedia, they can provide e-Library access to thousands of books, they also provide curriculum access using products like Moodle and providing a Jabber server for inter-laptop communication.
I've always thought the web browser to be the 'killer app' that provides the access to education tools. Thats really all that is needed.
"Constructivism" learning theory tend to ignore socio-cultural aspect of education and idealize children's ability to learn by themselves.
These flaws affected the way how OLPC do its business.
"OLPC News" has one of the best communities on Internet. More than 5,000 people... Any open source start up would kill to have such a community. And yet we are not able to reach people from OLPC. They just ignore us, don't answer our questions and concerns.
I begin to worry, may be XO is not an education, may be it's just a cheap PC. And they use "Education" as marketing trick to sell a cheap PC in huge volumes? Please prove me wrong.
"once the novelty of the few preloaded apps wears off, what’s there to keep you coming back"
A joke, isn't it? You are not serious?
What do children do most? Chat. I can assure you that mobile phones, SMS, and IM are basic necessities for "modern" children.
Children in the West are nagging their parents endlessly when they are forced to do without IM just because they have to visit some holiday resort or beach. A connection black-out is unearable, making a child feel like Robinson Crusoe. So we can expect children to use the chat activities whenever there is *some* connection.
Exchanging movies and music will be big too. For most children, the XO will be their music, radio, film, and TV box. Must I say more? Next are all the games they will load onto their computers. And some might even get a hang for reading, as the XO is a mobile library.
If I think what I would have been prepared to do to get an XO with internet access given the opportunity when I was 8, I can only shudder.
As usual, productive educational work will be just what does pay for the XOs.
Winter
Great hardware review from a hardware guy.
"I guess my strongest gut reaction is that once the novelty of the few preloaded apps wears off, what’s there to keep you coming back?"
I guess some people could use the internet browser for almost all their needs (eg Gmail Docs) but if you need more software, go to a rapidly growing list of apps for download:
( http://wiki.laptop.org/go/Activities )
"would be handy to have [...] an easy, straightforward way to discover and install new tools."
Good point - it's already available:
( http://wiki.laptop.org/go/Image:Xogetgtk_main.png )