> With the hardware now long past its life expectancy, spare parts hard to find, and zero support from the One Laptop Per Child organization, its time to face reality. The XO-1 laptop is history.
The harsh truth about any development project: building a building, agricultural equipment, that is easy; guaranteeing its upkeep, even the gas to run it… that’s rare, if not unheard of.
I've often wondered about that, myself, especially after reading stories about missionaries building schools in war-torn areas of Africa. "That's great and all, but what happens to that school after all the white people with the international press watching them leave?"
This is a terrible analogy, but it reminds me of playing PC real-time strategy games. I've never really gotten into them that much, but I always thought it would be a genre I would like. So every two years or so I pick up the latest and greatest game and play it for a little while. It almost invariably goes that I'll have a lot of fun in the first two levels before getting completely crushed in the third.
After 15 years of this, it's only been in the last couple of years that I've learned that I've been playing the games "wrong". I used to take a strategy of incremental building, of amassing an offensive for a massive push against the enemy. This works in the early levels because the early levels are easy and the enemies are usually programmed not to attack you head on. But it doesn't work in the later levels because they are constantly harassing you while they are also discovering key control points on the board and building at a much faster pace.
I've never learned how to play at that pace. It requires a certain macro-level awareness of the play to be able to juggle all of the resource gathering and building queues so that nothing is sitting idle. I've tried, and I usually end up getting trounced even worse, because I've left my rear unguarded and my resource gatherers get wiped out, cutting off my supplies.
Point being, I get the feeling that international aid works a lot like how I play strategy games. They over extend into enemy territory and don't protect their supply lines. And their efforts end up getting pretty well and good destroyed because of it. There doesn't seem to be any macro-level awareness of the whole scene. It's just a lot of little efforts, all scattered around, and none of them are watching each others backs.
Sorry, that was a completely inappropriate analogy, but it's all I've got in me after only one cup of coffee today #firstworldproblems.
Great comment. I don't think anyone ever thought the XO itself would be a panacea for any problem. It was meant to act as a starting point to help bridge the digital divide and act as a gateway to content and educational material (and maybe eventually other services). In fact one of the first projects that OLPC was working on for the XOs was a huge built in library.
The project was never about the laptop as a panacea, which is why the project wasn't limited to the hardware, but has always been about educational content, instructional methods, and support infrastructure, as well.
This is just as true in the developed world as well. I always cringe at laptop placement programs (or even worse, iPad placement programs) in k-12 schools. They always turn out to be a waste of money with nothing to show. The most needy students don't even have internet access at home anyway.
I contributed to OLPC during my time at Red Hat and beyond. For a while I was the only person in NYC with an XO and everyone used to swoon whenever I pulled it out. Sad to see it go, but happy that lots of the thinking and work behind it helped push the envelope in two ways; it led us to tablets and smartphones and got people thinking about using technology to seriously help improve and deliver education worldwide, especially in the places that could benefit the most, not just as an afterthought.
While it definitely didn't the dent in the universe that we were looking for, at least not our universe, we can't say it didn't at least make a few scratches.
I think the gist here is correct. In some ways the OLPC reminds me of the Crunchpad, though at least the crunched failed fast. The OLPC was probably flawed in concept. Creating the hardware, software at that price & at that quality. Figuring out a way for very underprivileged (they might have one under-qualified teacher for a mixed age class of 50) students to effectively use it. Getting deals done with developing country ministries. New user interface ideas. etc. Every one of those was a big hurdle and the whole amounted to an enormously ambitious project..
I don't think the idea should die. The goal of getting connected devices to underprivileged students is more relevant and achievable than ever.
The OLPCs mission can be superseded. It can be done as dozens of smaller projects:
- connectivity projects; cheap wifi in hubs, schools and centers
- Free education apps, maybe even a dedicated app store
- Suitable devices
A $100 rugged tablet/laptop seems more doable than ever.
- donations of devices, hubs, etc
These tasks can be done individually by people, charities, governments or companies competing, collaborating or completely separated. I bet most people on this site could think of several they could contribute time or money to. Some might even take one as a startup idea. The whole thing doesn't stalemate if some projects fail. Wifi in somas rural schools can be on project. Wifi hubs in urban poverty locations can be another. Free educational apps are a completely unrelated category of potentially excellent contributions.
Good riddance. This project was a failure before it ever got off the ground. Not because they couldn't build the hardware or software (they did) or because they lacked smart, talented people (the didn't), but because handing laptops to poor kids was just never that great an idea to begin with.
Technology isn't magical, it's a force-multiplier, and anything multiplied by zero (or something very small) is still zero (or very small). To make matters worse, a great deal of resources went into the project, resources that could have been used providing more basic types of development aid that would have created something to multiply in the future.
As an aside, has there been any serious evaluation into whether the laptops "worked"? Were there actually even any metrics laid out in advance by which success can be judged now? I'd be willing to admit that I'm wrong about the efficacy of the project (though the cost-effectiveness could still be debated) if there are positive empirical results.
A person, living in poverty, solved major problem. In this case, he did his work by hand and by phone. Absolutely something where a communications tool like an internet-enabled laptop would have helped.
I'll presume you're not intentionally misunderstanding...
glesica and I believe that the OLPC didn't do near enough to actually try to TEACH. Multiplying 0 teaching times a force-multiplying computer yields 0 teaching.
>Assuming that there is avaliable internet for your internet connected laptop
>Assuming there is available electricity for said laptop
>Assuming there is clean water, fresh food, good shelter/lighting available so the proprieties lie in developing with said laptop.
Technology is an amazing force for positive change. However, I think, as I believe so does the parent, you get more for your money if you first develop the surrounding infrastructure, then help provide the technology. No one is implying that the people living in these areas of the world are worth nothing. The implication is that the foundation around them is minimizing the potential gains via technology.
> Assuming that there is avaliable internet for your internet connected laptop
You don't need internet to have a communication tool. Mesh networks can actually work pretty well. OLPC had mesh networks. And they don't need to eventually reach the internet, either. Being able to send a text message to your neighbor, even 100 yards away, could have been pretty powerful. Connect a few of those together in a mesh, and you might be able to communicate MILES away. I don't actually know if the OLPC was any good at this, but it was built out of pieces that could have enabled it.
> Assuming there is available electricity for said laptop
You don't need electricity for a hand-crank laptop. OLPC had hand-crank power.
The fact that you bring up these factually incorrect arguments makes me think you don't actually know anything about OLPC. It's worth educating yourself, they made some interesting and good choices. They also made, in my opinion, some TERRIBLE choices.
Oh, and you said "lighting." Since the OLPC was hand-crank powered, it was often the brightest light source available to the families that got them.
...and on to your next point...
I think that OLPC computers COULD HAVE HAD instructional videos, like Khan Academy, in how to improve water quality, preserve food, deal with pests and diseases and infections and wounds, reproductive health, building good shelters, identifying cloud patterns. Kind of like a US Army Field Survival Guide. All tailored to the local language and culture.
In that respect, it could have been like air-dropping a Peace Corps volunteer with engineering know-how, into a community.
Sadly, the OLPC project didn't address those problems.
Hi, I worked at OLPC. You might not be aware that OLPC shipped offline Wikipedia snapshots (in an appropriate language) on each laptop, or that people made Khan Academy video packs that were downloadable to a "school server" with enough disk space, or that children who received OLPC laptops helped to translate Khan Academy videos to their own language:
So we did work on some of these issues, often successfully e.g. in the case of Wikipedia. We didn't have our small US team make instructional videos, though -- that doesn't seem wise (smacks of colonialism) or sustainable (OLPC laptops went to something like 30 countries).
There are plenty of things we should have done better, though. The mesh network was a failure in most locations, and the "hand crank" power didn't work, though we were able to run laptops off single solar panels:
A Wikipedia snapshot is great, but it's not a textbook. It's a reference.
If a community was going to get 100 laptops, I was hoping each laptop would get something like 1/100th (or 1/50th, for replication) of the videos. No "school server" necessary. But I guess that idea is dependent on the mesh network...
It sucks that the mesh network and the hand crank didn't work. Did the rip-cord work?
I'm glad the Khan Academy videos were getting translated, that's great.
> our small US team
I wish you'd figured out a way to go big. Something like enlisting the top 10 US universities for Education degrees to get all of their graduating students to each make a 30-minute educational video for OLPC as a senior project. Make it trivial to record videos on the OLPC itself, and trivial to share them with each other...
Never mind me. I'm just an armchair quarterback. Thank you for working on it.
I don't think he's equating the people themselves with zero, rather the benefit to them. A good example where technology has really helped in Africa are cell phones. So many stories of fishermen being able to call and ask "who needs fish?" and they're able to get to the right markets with better fish.
I didn't see these as having that 'killer app' for the populace where it noticeably makes life better, immediately, for the people they're given to.
Even in the sanitary-pad case above, the man said he was illiterate, IIRC. The computer wouldn't have done much there...
Everything people are saying here against OLPC, such as "give them food and build infrastructure first" equally applies to cell phones. But cell phones clearly are win in Africa and have really helped.
Would we all criticize a one-cell-phone-per-person charitable organization? Would people like glesica rant about how a one-cell-phone-per-person charitable organization is wasting, as he said, "resources that could have been used providing more basic types of development aid that would have created something to multiply in the future"
Technology is not magic, but it does help, even poor people.
I think you are misunderstanding. If you give a laptop to a kid that is getting a good standard of education (and nutrition) at home and at school, it will probably have more of an effect than that same laptop would have on a kid with limited access to good education which is the norm among potential recipients.
If a school has 50 pupils in one class, the cost of 50 laptops could instead be used to split the class in two (1 extra teacher) and provide students free breakfast for a year, even at a cost of $100 per laptop. Both of these are proven to have big effects on attendance and achievement. 25 well fed 10 year olds in a class vs 50 pekish 10 year olds is a big difference.
I would never say give a laptop to someone that doesn't have food. Or spend money on technology over providing enough calories for someone to live and work. I don't think anyone at OLPC says that either. We aren't talking about failed states like the Central African Republic here. Argentina and Brazil were huge proponents and adopters of OLPC. This is not a "technology" or "food" zero sum game.
But the parent comment didn't say "handing a laptop to someone who has no food and shelter is never a great idea". glesica said "handing a laptop to a poor child."
Technology is a force multiplier for curiosity and ingenuity and for problem solving. "poor" people have curiosity, ingenuity, and have problems that need solving. Yes, technology isn't magic that just solves problems, but to say that poor people can't use or benefit from personal technology is just patently wrong.
Then enlighten us, what were the goals? Because based on my reading over the years (I've followed the project with some interest, having worked in the academic technology field) the goal, from a very high level, was to provide kids with laptops, which were then supposed to catalyze various other improvements in their lives.
OLPC (org and community) works on the scale of deployments. Villages and sometimes entire countries are organized around educational technology access and infrastructure is found or built. Volunteers orgs stand up to help translate content, support and repair laptops in situ, and help the local teachers, students, and parents.
Also, they make/design the laptops. And Sugar Labs makes the Sugar software. Check out the blogs on http://planet.laptop.org/ for more on all of this, including the OLPC news article tat kicked off this HN thread.
I don't have time to go link-hunting at the moment, but I know I have read they had big problems with the infrastructure and community aspect of the plan. This is almost always the case in technology applied to education.
Handing out laptops (or iPads, or whatever) is easy, developing a curriculum that uses them effectively, achieving buy-in from relevant stakeholders, and keeping the hardware updated and in working order are challenges that are orders of magnitude greater, yet they get less attention and funding.
This is why I would be interested in genuine, academically rigorous evaluations of the success of the project using pre-established metrics (not just press releases trumpeting their achievements).
A quick search shows the expected results: there have been studies, most pronounced the project ineffective (or a failure), and there is controversy and politics involved in it all: https://duckduckgo.com/?q=OLPC+study
Real work is hard and I agree with your point that the deployments have struggled with reaching their goals around content and infrastructure. I've tried to re-frame the discussion in terms of the goals, rather than the laptops, but I don't suggest it's all been unicorns and rainbows in the field.
And I didn't go out there, so I'm cautious of criticizing their efforts.
I do believe that success has already been achieved in several countries measured by the goals of donors and supporters of the project (and community), but the studies, and detractors, are looking for different goals.
After all, not every child has a (hackable!?) laptop yet, but millions do that didn't before. :D
My friend's wife works in Compton, which is one of the most distraught and lowest income areas in the country. I asked as a technologist what can we do to help her. She said she needs laptops for every child.
Standardized tests are being taken on computers now and many of her kids do not even know the simplest commands -- drag and drop, right click for more options, drag to expand, etc. This is an obvious incremental hinderance on their ability to catch up to the rest of society.
Additionally, they actually have funding for computers, but the district pays the lowest vendor for the most hardware. Makes sense from a non-technical bureaucrat's point of view. More computers means more children can have access. However, these computers break down before the end of the year and they mysteriously disappear after being loaned to the children.
My friend's wife now has to bring in her own computer and use that to teach her children how to use the basics of computers because the system is so broken and because people have a generally naive and ignorant view of what is needed and not needed by the education system, even at home.
I worked at an NGO in Cambodia at about the time they were considering a country-wide roll-out of OLPC. I believe the OLPC organization initially demanded an all-or-nothing implementation, but later agreed to a pilot program run in just Siem Reap. The pilot went poorly and the implementations wasn't extended. Because it was a failure, there weren't PR releases done by any of the parties involved.
From what I've heard, the problems they saw were with things like the cost of electricity consumed when doing multi-unit charging at schools and difficulty in integrating OLPC devices into teacher-driven curriculums.
In Cambodia at the moment, it seems like inexpensive Android devices are starting to deliver on some of the promises of the OLPC program. They're rugged, energy efficient, simple for children to use, and have a large variety of educational apps and material available (though fewer are available in Khmer). A family owns it, so issues like theft and damage are their responsibility. I vacationed in Cambodia last year, and was amazed to see how many smart phones were in common use, even in poorer areas.
Tommy Jordan of Youtube fame started his own approach to helping children in other less developed countries take advantage of today's technology. I think his approach was more practical than putting purpose built technology in the hands of everyone.
It's completely irresponsible for them not to have do some kind of post-mortem analysis of whether and how it was succesful, and why not -- you can't just take everyone's money and excitement, and then abandon it.
It doesn't have to be an expensive research project, but at the very least the key movers and shakers involved need to do some thinking and report back.
Instead, Negroponte just abandons it? Totally irresponsible.
There is no post-mortem because OLPC isn't, contrary to this always-hostile "news" site's headline, dead.
Yes, the founder has, after several years, moved on to start other things as those skilled at getting things moving often do once things are well underway. Yes, the HQ isn't in Boston any more. Yes, the original XO-1 series has been replaced as the flagship product by the XO-4 series. None of these are even loosely equivalent to the project being dead, especially the last.
Yeah, I don't think I'm going to listen to anything from a guy who habitually calls FOSS advocates "freetards." In fact, it might be safer just to believe the opposite of everything he says.
The comments there are enlightening. It seems the website (the very website linked here) was established in order to spread mostly-negative news about the OLPC. The main author was in a partnership with Intel. Intel had (has?) a competing product aimed at education: The Classmate PC.
So it's no wonder this is the first site to report "OLPC is dead." Maybe it is, but it was a worthy humanitarian project and some paid shills were strafing it from this blind for its entire lifecycle.
Edit: Seems like about half the articles on the site are laughably negative. Pretty great wolf-in-sheep infowar here.
I won't mourn it. I participated in the original G1G1 (Get One, Give One) program and have an OLPC XO-1. I developed some software for it, and a hardware temperature sensor that worked through the microphone socket (http://blog.jgc.org/2008/03/building-temperature-probe-for-o...)
It was truly a horrible machine, and the software was worse.
I hate One Laptop per Child. Which is to say, I'm so passionate about the idea that watching other people make choices that I disagreed with pained me greatly. I desperately wanted them to succeed, but I thought they were unintentionally poisoning the well for all future ideas in the same space.
It's a nice enough idea, to make a computing device that is rugged enough, powered by humans, and mesh networked. That's a nice idea.
But the vastly more important idea was to TEACH with computers. And I feel OLPC completely under-served that need. Putting all of their eggs in the Constructionism basket enraged me.
I'd much rather see a team take existing hardware, existing operating system, and develop an app for it that acts as much as possible like the "Young Lady's Illustrated Primer" from Neal Stephenson's "The Diamond Age."
It's not the total solution, but I think it would have been far better to take Khan Academy, and cram it in a computer. A little bit like "Khan Academy on a Stick." http://mujica.org/khan/ (Translated to each language, and culturally adapted where necessary.)
I love the idea of a laptop being the cheapest possible way to give a kid a ton of books. ...but then they didn't cram a ton of books in each OLPC! Hand out enough books, and you'll get more Srinivasa Ramanujan's.
And for full disclosure, I have a Masters degree in Computer Science, but no background in education, so my opinion is just my opinion. But as a CompSci geek, the OLPC always felt to me like "Computer Nerds Save the World." Rather than "Educators cram what they know about teaching into a computer, so every child can have their own, personal, education box."
"but then they didn't cram a ton of books in each OLPC!"
Why do you say that? The content packs used in the deployments are customized to local language and culture. Translating and sourcing the included materials was and continues to be a huge undertaking and the burden of that fall mostly to the local team and government liasons.
Yes, it helps, but I'm afraid it proves my point. (I'm happy to just look at the English Sugar Labs.)
Maybe I'm just missing it, but I see learning activities, not books.
I'd like to see lectures and/or text books. For instance, Pre-Algebra, Algebra, Geometry, Algebra 2, Trigonometry, Pre-Calculus, Calculus, Differential Equations, Linear Algebra.
I think Khan Academy has a nice map of their materials:
I agree. CJB commented elsewhere in this thread (
cjbprime) and he knows better than I do what went on back then.
I think that creating educational content is an incredibly important thing, and that it should be freely distributable to all and translated and localized broadly. It's a huge undertaking.
There are a few projects around trying to do things like this (against the impressive might of the textbook publishing industry) and I wish them the best of luck. OLPC will be only one benefactor of such efforts, and KA is certainly a great project.
OK, I've noticed the same constructivist trend you did, but am on the other side of the fence in terms of approving of it. OLPC was clearly of that school and did fail to provide alternatives, but I'm far from convinced that a viable alternative to constructivism exists. It's one of those things that's very easy to criticise, yet hard to suggest replacing with some other actionable idea, which is why it's destined to keep reappearing as the technologist's solution for education ad infinitum.
It's worth saying that things like the Khan academy just didn't exist at the inception of OLPC, nor did netbooks at all. I'm very critical of the project (I think Negroponte is a smooth talking arse) but things by technologists rarely are what they say they are, and more often than not the purpose is merely the development of more technology not the actual stated aim of whatever they're involved in. OLPC was never an education project at all, and anyone persuaded otherwise was merely deluding themselves.
> I'd much rather see a team take existing hardware, existing operating system, and develop an app for it that acts as much as possible like the "Young Lady's Illustrated Primer" from Neal Stephenson's "The Diamond Age."
I guess that's any compsci person's dream when they read that book, including mine.
One thing that I've noticed in recent days is something that is related to Popcorn Time (another top story on HN as of this time) -- young people love streamlined workflows and tend to avoid complex ones.
I think this observation is reflected in many popular apps/services of today (Facebook, WhatsApp, Spotify) but also in many startups -- most ones are like Uber and deliver something that we already know and use (taxis) but with a streamlined process.
I am getting a bit worried that this migration to the simple is killing a lot of projects that are noble, but "clunky". OLPC might have been one of those -- the interface was never as polished as Uber/Spotify, even though they certainly aimed at it being rather simple.
And here's my main point: I am worried Project Gutenberg is exactly one of those projects that will disappear because they're clunky in the eyes of the young.
The website is old-looking (a Wikimedia engine, likely) and it is not that easy (compared to shopping for apps on a phone) to search for the right book -- look for Emile Zola and you will have to look for many, many results, some of them being the same book but in a different language (and there is no sorting based on say your location).
Worst of all -- most consumer ebook devices are pushing their own bookstores, making it more difficult to load and read books from outside sources such as Gutenberg.
Loading up tons of free books to any kid's "device" sounds like the first baby step to a Primer, and yet even the basic foundation of that -- Project Gutenberg -- is as clunky as OLPC (and lacks a lot of old books, too).
In its place, is the reality that technology is a force in education, and we all need to be vigilant about when, where, and how it's used.
Is there any empirical proof that (computer) technology actually improves learning? That kids who used laptops/tablets achieve higher levels of proficiency in maths, reading, writing... than pupils who went to school with blackboard, pen and paper? Or show bigger improvement? By any measure.
They learn more about computers by using them. That was, I thought, the point. This whole "teach math, science, history, literature, etc. -- but better because COMPUTERS" crap was always overblown, in my opinion. I supported OLPC because I never thought it was about that. It was about bringing technology to children around the world and seeing what they could do with it. Maybe they had to shift their focus to get governments and teachers on board?
Wow, so much hostility in this thread. Why is there so much negativity about OLPC and XO?
It's been hard to find out from outside what's going on at the various OLPC orgs and companies in recent year, but they are doing quite a bit. The Vivitar devices sold quite well and have some of the same potential in the rich world that XO-1 + have had and continue to have in the world at large.
There are millions of laptops in children's hands. They are learning to use computers and networks and how to hack, not to mention to read , write, and cipher. How is that failure?
http://one.laptop.org/stories
Sugar is certainly alive. Sugar on a Stick 10 is quite slick and ships with dozens of educational activities. As noted below their development targets are set by the demands of the their users, educators and technologists who are using the system:
http://wiki.sugarlabs.org/go/Sugar_on_a_Stick/10
Even if all they'd managed was to design and produce a few thousand XO laptops: open architecture, open hardware , ruggedised laptops running Linux (oh so many years ago), that would be an amazing accomplishment.
Since we're arguing: what have you, or your code, done to make the world a better place? I'm proud of my tiny forgettable contributions to OLPC and wish I could do more.
The single biggest failing in my mind was the inability of low income folks to purchase one of these laptops. My Atari 400 and the magazines of the era (e.g. Byte, Antic, Creative Computing, Dr. Dobb's Journal) provided a lifeline to the world of technology. With the 8-bit era pricing long gone, I had hoped OLPC would be this generation's Sinclair.
It wasn't to be. OLPC operated "programs" and the times they let normal people buy were at twice the price (one for you, one for someone else).
Logistics and support remain a huge resource intensive problem for any organization.
To summarize a lot of arguments had internally in the project at that time: it was better to outsource as much of that as possible to keep the organization's focus and few resources on deployments and technology development, and the partners chosen for that role disappointed the staff, volunteers, and donors immensely (twice). Technology development went quite well, the deployments were wildly successful, and the "get XO laptops to rich world children" part didn't really get as far.
The Vivitar tablets seem like a much better plan for the North American market and seem to be pretty successful so far. they are available (or were) at Walmart and can be ordered online from Amazon, et al. Of course the technology (and cost of parts) to do that weren't available then, and can many believe be credited as a success of OLPC.
I must admit the phrasing of this makes me a tad bit mad. I'm actually at a loss for words to explain why and I believe it fits with why I see OLPC as a failure.
Looking at it, what is the difference between the Vivitar tablets and something like the LeapPad Ultra? They both look like locked down tablets with pre-packaged lessons and lots of parent reporting and control. The LeapPad looks better supported.
It's provocative phrasing, in no small smart because I agree with your upset. And it is deliberate to motivate you (and I) to seek change.
OLPC's efforts to link up with US school systems and do work here were hampered by the same local and national politics, entrenched interests, bad laws, and other ugly realities that hamper all such efforts.
The difference between the xotablet and the LeapPad are myriad but not nearly as significant as the contrast between either and an XO. The very openness of the XO is intrinsic to its educational potential and that openness is a core objective of the project.
That openness closed some doors and made them some powerful enemies but in exchange they created some things that may last a long time and be of use to a great many people.
In a very real way OLPC's successes in other countries benefit all of humanity, and thereby directly helps children in US schools. Non-zero sum. Just not as fast as I (and I believe you) would like.
It includes the article above, but also includes posts by people in the field deploying laptops -- still rolling along. As 'themodelplumber' pointed out elsewhere, this is just some blogger (who has invested a lot in the community) claiming the project is dead for unknown reasons.
I've gotta believe, give a valuable 1st-world electronic device to a child in a 3rd-world country: how long are they going to keep it? "Hey child, gimme that!". 10 seconds I would guess. Then its on the black market, which arguably might do the child more good (more discretionary money in a poor family)
To stonogo's point, the security systems for XO actually model that threat, and do a fair bit to reduce the risk.
Foremost protection against theft is that the laptops are brightly colored and made of shiny plastic, like toys. In a successful deployment the sight of children with their laptops is normal but an adult with one is quite unusual.
The technical security in the OS (wiki link above) is remarkable. I think it inspired many aspects of other security systems we use today (Gatekeeper/Sandbox, for one).
There's no actual news here -- just an opinion site from the always-OLPC-hostile olpcnews.com claiming that OLPC is "dead" in some metaphoric sense because the original device (XO-1) is no longer the flagship device (the new flagship being the XO-4/XO-4 Touch), and because the headquarters is now in Miami instead of Boston.
There's a lot of comments here that OLPC was a failure, and while true, I think people forget what an audacious and pioneering idea it was and where the OLPC "experiment" was a success.
Back in 2005, when the OLPC project was first announced, a $100 for a laptop price point was virtually inconceivable. I don't think they ever hit that price, but they introduced a lot of interesting funding vehicles that are pretty normal now, like buy one to give one. It also opened up lots of discussion about low power devices and the suitability of traditional screen tech for low-power use cases like e-book reading. The ideas were so good and so ground breaking that one of the partners, Intel started a class of consumer computers around them and the conversation for portables switched, for a time, from "expensive and high end" to "low power and inexpensive".
Moore's law, and the global shift to smartphones eventually overcame lots of this focus, but I think it definitely set a vector for a while that still informs the industry.
Didn't the XO basically kick off the netbook craze?
While not the intended effect, it was certainly a catalyst for a not insignificant market shift. I'd also argue that netbooks helped pave the way for tablet popularization to a small extent (much less than, say, smartphones, but still not unappreciable).
I actually feel that many people, especially poor people, benefited from the Netbook craze. At the time, I worked at a computer store in Sunnyvale, CA selling many different types of used PCs. Netbooks allowed many people to purchase a fully working small laptop for their kids and for themselves that they treated as almost fully disposable. The OLPC did challenge the industry to create a cheaper option.
Netbooks were a truly portable laptop. You could stick in your glove compartment, you could stick it in a small backpack. It just didn't get the respect that a full blown $1000+ laptop would get. Today, the computers that I stick in my glove compartment and backpacks are tablets, and this because the Netbook form-factor instilled certain computing habits that the Tablets later improved.
I worked on the OLPC during my undergrad at MIT (worked on mesh networking their DSP engine and ported cocos2D to Sugar/PyGTK/PyCairo).
Things that impressed me: the ambitiousness of the project, the fact that they started with some definite opinions and made something that matched those opinions, the group of talented people that flocked to be part of it and the energy in the building.
Things that did NOT impress me: Negroponte's aloofness with the team, his refusal to see the corpo-political implications of the project. I've always wondered what would have happened if had been more aggressive in defending his project by generating bad press about Intel's and Microsoft's attempts to tank the project.
This was the first netbook. Asus followed with eeePC a year later.
Hardware was ok'ish for the time. Nokia was already shipping ARM tablets running full linux, but OLPC went with x86 because Nagroponte loved playing golf with AMD and M$ sales people (M$ promised free XP licenses).
One Laptop Per Child is very much alive as an idea or even a right in developing world. They may not be using OLPC or Sugar but OLPC showed the way how it could be done and it was doable. Every poor nation in the world knows this by now that education is the ticket to prosperity and computers help deal with teacher shortage . Android, ARM and the Chinese manufacturers will ensure that one laptop per child is indeed a reality one day.
I always thought that OLPC was a great idea poorly implemented. Sugar was interesting but a bit too out there for most people to want hack on. The mesh networking was the best part.
I wish they had used standard PC hardware so that the communities that got the hardware could learn to service the devices and make changes to the software themselves thus building local technology savvy communities that could use technology to solve their problems with in community knowledge thus enable a ground up technology movement.
Uh.. if you are not trolling you are badly mis-informed.
XO-1 was, "standard PC" x86, in part to get cheaper available components. They didn't switch to ARM until after the price per processor came down enough to make it viable, though it would have been more efficient to start there (cf Intel-Microsoft-Negroponte politics).
Parts were and are shipped in to the locations. The laptops were designed to be field repairable to an extent well beyond any commercial laptop (then or now): resources start here http://wiki.laptop.org/go/Repair and include teardown and repair instructions.
XO-1 series are,were much more repairable than any new or 2nd-hand consumer electronics gear you can get. I only hope they manage it with XO-4. Touchscreens are pretty fragile.
79 comments
[ 3.1 ms ] story [ 144 ms ] threadThe harsh truth about any development project: building a building, agricultural equipment, that is easy; guaranteeing its upkeep, even the gas to run it… that’s rare, if not unheard of.
This is a terrible analogy, but it reminds me of playing PC real-time strategy games. I've never really gotten into them that much, but I always thought it would be a genre I would like. So every two years or so I pick up the latest and greatest game and play it for a little while. It almost invariably goes that I'll have a lot of fun in the first two levels before getting completely crushed in the third.
After 15 years of this, it's only been in the last couple of years that I've learned that I've been playing the games "wrong". I used to take a strategy of incremental building, of amassing an offensive for a massive push against the enemy. This works in the early levels because the early levels are easy and the enemies are usually programmed not to attack you head on. But it doesn't work in the later levels because they are constantly harassing you while they are also discovering key control points on the board and building at a much faster pace.
I've never learned how to play at that pace. It requires a certain macro-level awareness of the play to be able to juggle all of the resource gathering and building queues so that nothing is sitting idle. I've tried, and I usually end up getting trounced even worse, because I've left my rear unguarded and my resource gatherers get wiped out, cutting off my supplies.
Point being, I get the feeling that international aid works a lot like how I play strategy games. They over extend into enemy territory and don't protect their supply lines. And their efforts end up getting pretty well and good destroyed because of it. There doesn't seem to be any macro-level awareness of the whole scene. It's just a lot of little efforts, all scattered around, and none of them are watching each others backs.
Sorry, that was a completely inappropriate analogy, but it's all I've got in me after only one cup of coffee today #firstworldproblems.
While it definitely didn't the dent in the universe that we were looking for, at least not our universe, we can't say it didn't at least make a few scratches.
I don't think the idea should die. The goal of getting connected devices to underprivileged students is more relevant and achievable than ever.
The OLPCs mission can be superseded. It can be done as dozens of smaller projects:
These tasks can be done individually by people, charities, governments or companies competing, collaborating or completely separated. I bet most people on this site could think of several they could contribute time or money to. Some might even take one as a startup idea. The whole thing doesn't stalemate if some projects fail. Wifi in somas rural schools can be on project. Wifi hubs in urban poverty locations can be another. Free educational apps are a completely unrelated category of potentially excellent contributions.Technology isn't magical, it's a force-multiplier, and anything multiplied by zero (or something very small) is still zero (or very small). To make matters worse, a great deal of resources went into the project, resources that could have been used providing more basic types of development aid that would have created something to multiply in the future.
As an aside, has there been any serious evaluation into whether the laptops "worked"? Were there actually even any metrics laid out in advance by which success can be judged now? I'd be willing to admit that I'm wrong about the efficacy of the project (though the cost-effectiveness could still be debated) if there are positive empirical results.
Those are people you are so callously and casually equating with zero, and a person's economic status has nothing to do with their value.
I can't help but think of this earlier hacker news story:
http://www.independent.co.uk/news/world/asia/the-tampon-king...
A person, living in poverty, solved major problem. In this case, he did his work by hand and by phone. Absolutely something where a communications tool like an internet-enabled laptop would have helped.
glesica and I believe that the OLPC didn't do near enough to actually try to TEACH. Multiplying 0 teaching times a force-multiplying computer yields 0 teaching.
>Assuming there is available electricity for said laptop
>Assuming there is clean water, fresh food, good shelter/lighting available so the proprieties lie in developing with said laptop.
Technology is an amazing force for positive change. However, I think, as I believe so does the parent, you get more for your money if you first develop the surrounding infrastructure, then help provide the technology. No one is implying that the people living in these areas of the world are worth nothing. The implication is that the foundation around them is minimizing the potential gains via technology.
> Assuming that there is avaliable internet for your internet connected laptop
You don't need internet to have a communication tool. Mesh networks can actually work pretty well. OLPC had mesh networks. And they don't need to eventually reach the internet, either. Being able to send a text message to your neighbor, even 100 yards away, could have been pretty powerful. Connect a few of those together in a mesh, and you might be able to communicate MILES away. I don't actually know if the OLPC was any good at this, but it was built out of pieces that could have enabled it.
> Assuming there is available electricity for said laptop
You don't need electricity for a hand-crank laptop. OLPC had hand-crank power.
The fact that you bring up these factually incorrect arguments makes me think you don't actually know anything about OLPC. It's worth educating yourself, they made some interesting and good choices. They also made, in my opinion, some TERRIBLE choices.
Oh, and you said "lighting." Since the OLPC was hand-crank powered, it was often the brightest light source available to the families that got them.
...and on to your next point...
I think that OLPC computers COULD HAVE HAD instructional videos, like Khan Academy, in how to improve water quality, preserve food, deal with pests and diseases and infections and wounds, reproductive health, building good shelters, identifying cloud patterns. Kind of like a US Army Field Survival Guide. All tailored to the local language and culture.
In that respect, it could have been like air-dropping a Peace Corps volunteer with engineering know-how, into a community.
Sadly, the OLPC project didn't address those problems.
http://blog.printf.net/articles/2013/02/01/children-in-peru-...
http://blog.laptop.org/2011/06/06/khan-academy-videos-and-ol...
http://wiki.laptop.org/go/Khan_Academy
So we did work on some of these issues, often successfully e.g. in the case of Wikipedia. We didn't have our small US team make instructional videos, though -- that doesn't seem wise (smacks of colonialism) or sustainable (OLPC laptops went to something like 30 countries).
There are plenty of things we should have done better, though. The mesh network was a failure in most locations, and the "hand crank" power didn't work, though we were able to run laptops off single solar panels:
https://www.youtube.com/watch?v=ITHNbOrPQyM
If a community was going to get 100 laptops, I was hoping each laptop would get something like 1/100th (or 1/50th, for replication) of the videos. No "school server" necessary. But I guess that idea is dependent on the mesh network...
It sucks that the mesh network and the hand crank didn't work. Did the rip-cord work?
I'm glad the Khan Academy videos were getting translated, that's great.
> our small US team
I wish you'd figured out a way to go big. Something like enlisting the top 10 US universities for Education degrees to get all of their graduating students to each make a 30-minute educational video for OLPC as a senior project. Make it trivial to record videos on the OLPC itself, and trivial to share them with each other...
Never mind me. I'm just an armchair quarterback. Thank you for working on it.
Thanks for wading into this olpcnews FUD inspired mess ;) Glad to see you, even to rehash old arguments.
Cheers, adric
I didn't see these as having that 'killer app' for the populace where it noticeably makes life better, immediately, for the people they're given to.
Even in the sanitary-pad case above, the man said he was illiterate, IIRC. The computer wouldn't have done much there...
Would we all criticize a one-cell-phone-per-person charitable organization? Would people like glesica rant about how a one-cell-phone-per-person charitable organization is wasting, as he said, "resources that could have been used providing more basic types of development aid that would have created something to multiply in the future"
Technology is not magic, but it does help, even poor people.
If a school has 50 pupils in one class, the cost of 50 laptops could instead be used to split the class in two (1 extra teacher) and provide students free breakfast for a year, even at a cost of $100 per laptop. Both of these are proven to have big effects on attendance and achievement. 25 well fed 10 year olds in a class vs 50 pekish 10 year olds is a big difference.
I would never say give a laptop to someone that doesn't have food. Or spend money on technology over providing enough calories for someone to live and work. I don't think anyone at OLPC says that either. We aren't talking about failed states like the Central African Republic here. Argentina and Brazil were huge proponents and adopters of OLPC. This is not a "technology" or "food" zero sum game.
But the parent comment didn't say "handing a laptop to someone who has no food and shelter is never a great idea". glesica said "handing a laptop to a poor child."
Technology is a force multiplier for curiosity and ingenuity and for problem solving. "poor" people have curiosity, ingenuity, and have problems that need solving. Yes, technology isn't magic that just solves problems, but to say that poor people can't use or benefit from personal technology is just patently wrong.
Get over yourself. That was not even remotely what was said or implied.
Being in a situation where you have near-zero utility from a computer is not "being equal to zero".
OLPC (org and community) works on the scale of deployments. Villages and sometimes entire countries are organized around educational technology access and infrastructure is found or built. Volunteers orgs stand up to help translate content, support and repair laptops in situ, and help the local teachers, students, and parents.
Also, they make/design the laptops. And Sugar Labs makes the Sugar software. Check out the blogs on http://planet.laptop.org/ for more on all of this, including the OLPC news article tat kicked off this HN thread.
Handing out laptops (or iPads, or whatever) is easy, developing a curriculum that uses them effectively, achieving buy-in from relevant stakeholders, and keeping the hardware updated and in working order are challenges that are orders of magnitude greater, yet they get less attention and funding.
This is why I would be interested in genuine, academically rigorous evaluations of the success of the project using pre-established metrics (not just press releases trumpeting their achievements).
Real work is hard and I agree with your point that the deployments have struggled with reaching their goals around content and infrastructure. I've tried to re-frame the discussion in terms of the goals, rather than the laptops, but I don't suggest it's all been unicorns and rainbows in the field.
And I didn't go out there, so I'm cautious of criticizing their efforts.
I do believe that success has already been achieved in several countries measured by the goals of donors and supporters of the project (and community), but the studies, and detractors, are looking for different goals.
After all, not every child has a (hackable!?) laptop yet, but millions do that didn't before. :D
Standardized tests are being taken on computers now and many of her kids do not even know the simplest commands -- drag and drop, right click for more options, drag to expand, etc. This is an obvious incremental hinderance on their ability to catch up to the rest of society.
Additionally, they actually have funding for computers, but the district pays the lowest vendor for the most hardware. Makes sense from a non-technical bureaucrat's point of view. More computers means more children can have access. However, these computers break down before the end of the year and they mysteriously disappear after being loaned to the children.
My friend's wife now has to bring in her own computer and use that to teach her children how to use the basics of computers because the system is so broken and because people have a generally naive and ignorant view of what is needed and not needed by the education system, even at home.
From what I've heard, the problems they saw were with things like the cost of electricity consumed when doing multi-unit charging at schools and difficulty in integrating OLPC devices into teacher-driven curriculums.
In Cambodia at the moment, it seems like inexpensive Android devices are starting to deliver on some of the promises of the OLPC program. They're rugged, energy efficient, simple for children to use, and have a large variety of educational apps and material available (though fewer are available in Khmer). A family owns it, so issues like theft and damage are their responsibility. I vacationed in Cambodia last year, and was amazed to see how many smart phones were in common use, even in poorer areas.
http://8minutesoffame.com/pressrelease-1/
http://www.manhattandigest.com/2013/08/20/project-phoenix-ri...
http://www.youtube.com/watch?v=5Smmh3D_tXE
It doesn't have to be an expensive research project, but at the very least the key movers and shakers involved need to do some thinking and report back.
Instead, Negroponte just abandons it? Totally irresponsible.
Yes, the founder has, after several years, moved on to start other things as those skilled at getting things moving often do once things are well underway. Yes, the HQ isn't in Boston any more. Yes, the original XO-1 series has been replaced as the flagship product by the XO-4 series. None of these are even loosely equivalent to the project being dead, especially the last.
I'd recommend reading all of his OLPC articles.
http://news.slashdot.org/story/08/01/12/1424209/intel-employ...
The comments there are enlightening. It seems the website (the very website linked here) was established in order to spread mostly-negative news about the OLPC. The main author was in a partnership with Intel. Intel had (has?) a competing product aimed at education: The Classmate PC.
So it's no wonder this is the first site to report "OLPC is dead." Maybe it is, but it was a worthy humanitarian project and some paid shills were strafing it from this blind for its entire lifecycle.
Edit: Seems like about half the articles on the site are laughably negative. Pretty great wolf-in-sheep infowar here.
It was truly a horrible machine, and the software was worse.
It's a nice enough idea, to make a computing device that is rugged enough, powered by humans, and mesh networked. That's a nice idea.
But the vastly more important idea was to TEACH with computers. And I feel OLPC completely under-served that need. Putting all of their eggs in the Constructionism basket enraged me.
I'd much rather see a team take existing hardware, existing operating system, and develop an app for it that acts as much as possible like the "Young Lady's Illustrated Primer" from Neal Stephenson's "The Diamond Age."
It's not the total solution, but I think it would have been far better to take Khan Academy, and cram it in a computer. A little bit like "Khan Academy on a Stick." http://mujica.org/khan/ (Translated to each language, and culturally adapted where necessary.)
I love the idea of a laptop being the cheapest possible way to give a kid a ton of books. ...but then they didn't cram a ton of books in each OLPC! Hand out enough books, and you'll get more Srinivasa Ramanujan's.
And for full disclosure, I have a Masters degree in Computer Science, but no background in education, so my opinion is just my opinion. But as a CompSci geek, the OLPC always felt to me like "Computer Nerds Save the World." Rather than "Educators cram what they know about teaching into a computer, so every child can have their own, personal, education box."
https://onebillion.org.uk/malawi
Why do you say that? The content packs used in the deployments are customized to local language and culture. Translating and sourcing the included materials was and continues to be a huge undertaking and the burden of that fall mostly to the local team and government liasons.
You'll hit language barriers pretty quickly trying to keep up with most of those teams, but the links are around. EG: http://www.olenepal.org/, http://www.olpcmexico.org/
well, I do. apologies for projecting. Looks like Mexican school kids might get XO-4 before me. #firstworldproblems
For Spanish, Science and Math there are: http://activities.sugarlabs.org/es-ES/sugar/browse/type:1/ca...
For XOs with network access (including just to a classroom network) included activities allow students to add books. http://activities.sugarlabs.org/en-US/sugar/addon/4304
And the Vivtar tablet include non-free educational content from many licenses in English and Spanish (for the Americas): http://xotablet.com/#features
Wikpedia subsets in local language are included in the XO images I've seen, though I don't have detailed links to those handy (It's on my XO).
Does that help?
Maybe I'm just missing it, but I see learning activities, not books.
I'd like to see lectures and/or text books. For instance, Pre-Algebra, Algebra, Geometry, Algebra 2, Trigonometry, Pre-Calculus, Calculus, Differential Equations, Linear Algebra.
I think Khan Academy has a nice map of their materials:
https://www.khanacademy.org/exercisedashboard
Wikipedia subsets are nice to have, but they're not Course Material. It's a good reference, not a teaching book, filled with problem sets, etc.
I think that creating educational content is an incredibly important thing, and that it should be freely distributable to all and translated and localized broadly. It's a huge undertaking.
There are a few projects around trying to do things like this (against the impressive might of the textbook publishing industry) and I wish them the best of luck. OLPC will be only one benefactor of such efforts, and KA is certainly a great project.
Another hub of free educational content and tools is is Moodle. XO Server ships Moodle with content packs, but again, I'm not sure what is in them, sorry :/ You can start near here to check on them: http://wiki.laptop.org/go/XS_Community_Edition links to http://schoolserver.org/
hth and thanks for discussion, adric
It's worth saying that things like the Khan academy just didn't exist at the inception of OLPC, nor did netbooks at all. I'm very critical of the project (I think Negroponte is a smooth talking arse) but things by technologists rarely are what they say they are, and more often than not the purpose is merely the development of more technology not the actual stated aim of whatever they're involved in. OLPC was never an education project at all, and anyone persuaded otherwise was merely deluding themselves.
I guess that's any compsci person's dream when they read that book, including mine.
One thing that I've noticed in recent days is something that is related to Popcorn Time (another top story on HN as of this time) -- young people love streamlined workflows and tend to avoid complex ones.
I think this observation is reflected in many popular apps/services of today (Facebook, WhatsApp, Spotify) but also in many startups -- most ones are like Uber and deliver something that we already know and use (taxis) but with a streamlined process.
I am getting a bit worried that this migration to the simple is killing a lot of projects that are noble, but "clunky". OLPC might have been one of those -- the interface was never as polished as Uber/Spotify, even though they certainly aimed at it being rather simple.
And here's my main point: I am worried Project Gutenberg is exactly one of those projects that will disappear because they're clunky in the eyes of the young.
The website is old-looking (a Wikimedia engine, likely) and it is not that easy (compared to shopping for apps on a phone) to search for the right book -- look for Emile Zola and you will have to look for many, many results, some of them being the same book but in a different language (and there is no sorting based on say your location).
Worst of all -- most consumer ebook devices are pushing their own bookstores, making it more difficult to load and read books from outside sources such as Gutenberg.
Loading up tons of free books to any kid's "device" sounds like the first baby step to a Primer, and yet even the basic foundation of that -- Project Gutenberg -- is as clunky as OLPC (and lacks a lot of old books, too).
Is there any empirical proof that (computer) technology actually improves learning? That kids who used laptops/tablets achieve higher levels of proficiency in maths, reading, writing... than pupils who went to school with blackboard, pen and paper? Or show bigger improvement? By any measure.
It's been hard to find out from outside what's going on at the various OLPC orgs and companies in recent year, but they are doing quite a bit. The Vivitar devices sold quite well and have some of the same potential in the rich world that XO-1 + have had and continue to have in the world at large.
There are millions of laptops in children's hands. They are learning to use computers and networks and how to hack, not to mention to read , write, and cipher. How is that failure? http://one.laptop.org/stories
Sugar is certainly alive. Sugar on a Stick 10 is quite slick and ships with dozens of educational activities. As noted below their development targets are set by the demands of the their users, educators and technologists who are using the system: http://wiki.sugarlabs.org/go/Sugar_on_a_Stick/10
Even if all they'd managed was to design and produce a few thousand XO laptops: open architecture, open hardware , ruggedised laptops running Linux (oh so many years ago), that would be an amazing accomplishment.
Since we're arguing: what have you, or your code, done to make the world a better place? I'm proud of my tiny forgettable contributions to OLPC and wish I could do more.
It wasn't to be. OLPC operated "programs" and the times they let normal people buy were at twice the price (one for you, one for someone else).
To summarize a lot of arguments had internally in the project at that time: it was better to outsource as much of that as possible to keep the organization's focus and few resources on deployments and technology development, and the partners chosen for that role disappointed the staff, volunteers, and donors immensely (twice). Technology development went quite well, the deployments were wildly successful, and the "get XO laptops to rich world children" part didn't really get as far.
The Vivitar tablets seem like a much better plan for the North American market and seem to be pretty successful so far. they are available (or were) at Walmart and can be ordered online from Amazon, et al. Of course the technology (and cost of parts) to do that weren't available then, and can many believe be credited as a success of OLPC.
I just wish they ran Sugar.
I must admit the phrasing of this makes me a tad bit mad. I'm actually at a loss for words to explain why and I believe it fits with why I see OLPC as a failure.
Looking at it, what is the difference between the Vivitar tablets and something like the LeapPad Ultra? They both look like locked down tablets with pre-packaged lessons and lots of parent reporting and control. The LeapPad looks better supported.
OLPC's efforts to link up with US school systems and do work here were hampered by the same local and national politics, entrenched interests, bad laws, and other ugly realities that hamper all such efforts.
The difference between the xotablet and the LeapPad are myriad but not nearly as significant as the contrast between either and an XO. The very openness of the XO is intrinsic to its educational potential and that openness is a core objective of the project.
That openness closed some doors and made them some powerful enemies but in exchange they created some things that may last a long time and be of use to a great many people.
In a very real way OLPC's successes in other countries benefit all of humanity, and thereby directly helps children in US schools. Non-zero sum. Just not as fast as I (and I believe you) would like.
It includes the article above, but also includes posts by people in the field deploying laptops -- still rolling along. As 'themodelplumber' pointed out elsewhere, this is just some blogger (who has invested a lot in the community) claiming the project is dead for unknown reasons.
Foremost protection against theft is that the laptops are brightly colored and made of shiny plastic, like toys. In a successful deployment the sight of children with their laptops is normal but an adult with one is quite unusual.
The technical security in the OS (wiki link above) is remarkable. I think it inspired many aspects of other security systems we use today (Gatekeeper/Sandbox, for one).
Back in 2005, when the OLPC project was first announced, a $100 for a laptop price point was virtually inconceivable. I don't think they ever hit that price, but they introduced a lot of interesting funding vehicles that are pretty normal now, like buy one to give one. It also opened up lots of discussion about low power devices and the suitability of traditional screen tech for low-power use cases like e-book reading. The ideas were so good and so ground breaking that one of the partners, Intel started a class of consumer computers around them and the conversation for portables switched, for a time, from "expensive and high end" to "low power and inexpensive".
Moore's law, and the global shift to smartphones eventually overcame lots of this focus, but I think it definitely set a vector for a while that still informs the industry.
While not the intended effect, it was certainly a catalyst for a not insignificant market shift. I'd also argue that netbooks helped pave the way for tablet popularization to a small extent (much less than, say, smartphones, but still not unappreciable).
Netbooks were a truly portable laptop. You could stick in your glove compartment, you could stick it in a small backpack. It just didn't get the respect that a full blown $1000+ laptop would get. Today, the computers that I stick in my glove compartment and backpacks are tablets, and this because the Netbook form-factor instilled certain computing habits that the Tablets later improved.
People forget, but this was the explicit goal of the effort - to drastically lower the costs of portable computing for everyone.
Things that impressed me: the ambitiousness of the project, the fact that they started with some definite opinions and made something that matched those opinions, the group of talented people that flocked to be part of it and the energy in the building.
Things that did NOT impress me: Negroponte's aloofness with the team, his refusal to see the corpo-political implications of the project. I've always wondered what would have happened if had been more aggressive in defending his project by generating bad press about Intel's and Microsoft's attempts to tank the project.
https://www.youtube.com/watch?v=QoiECE3qlDI
This was the first netbook. Asus followed with eeePC a year later.
Hardware was ok'ish for the time. Nokia was already shipping ARM tablets running full linux, but OLPC went with x86 because Nagroponte loved playing golf with AMD and M$ sales people (M$ promised free XP licenses).
Software was a total GARBAGE.
http://lukego.livejournal.com/8760.html
http://lukego.livejournal.com/10847.html
http://lukego.livejournal.com/24833.html
OLPC is actually still going, though it has transformed quite a lot and isn't sexy anymore.
I always thought that OLPC was a great idea poorly implemented. Sugar was interesting but a bit too out there for most people to want hack on. The mesh networking was the best part.
I wish they had used standard PC hardware so that the communities that got the hardware could learn to service the devices and make changes to the software themselves thus building local technology savvy communities that could use technology to solve their problems with in community knowledge thus enable a ground up technology movement.
XO-1 was, "standard PC" x86, in part to get cheaper available components. They didn't switch to ARM until after the price per processor came down enough to make it viable, though it would have been more efficient to start there (cf Intel-Microsoft-Negroponte politics).
Parts were and are shipped in to the locations. The laptops were designed to be field repairable to an extent well beyond any commercial laptop (then or now): resources start here http://wiki.laptop.org/go/Repair and include teardown and repair instructions.
XO-1 series are,were much more repairable than any new or 2nd-hand consumer electronics gear you can get. I only hope they manage it with XO-4. Touchscreens are pretty fragile.