Changes at OLPC [LWN.net]

Changes at OLPC

Posted Jan 8, 2009 2:11 UTC (Thu) by csigler (subscriber, #1224)
Parent article: Changes at OLPC

Well, (and please feel free to flame me as appropriate,) this is, at last, the final sign of the failure of the OLPC project. I found I had interest in this project as an outsider, and what I perceive to be its laudable goals. But I never took them up on G1G1, and when I saw the production problems and delivery delays I was relieved I didn't. After their Xmas 2007 push, I lost interest. Am I, and others, at fault for not getting involved, becoming an insider, keeping up interest, investing in the project through G1G1, etc.? To be sure, yes. But this is the way the real world works, and that is what I perceive to be a major OLPC problem from its inception. In an ideal world, it might have worked....

It's my experience that idealistic projects like this, in general, don't succeed. It takes pragmatism to succeed. And Negroponte's play to the Windows hegemony was _not_ pragmatism. It was likely a flailing, last-chance-type grasp at a lifeline to rescue OLPC once it was seen that their goals were probably unreachable.

To compound the lack of pragmatism, they displayed poor strategy and planning. The laptops cost more than the promised $100 figure, which had been a _major_ selling point. In fact, they were nearly double. They should've been able to predict that from fairly early on. Even if $100 was the planning target, had enough experienced large-scale engineering and manufacturing people been involved they could've drawn a line in the sand and said, "Um, guys, $100 is an ultimate fail. The real cost will be closer to $200 in the beginning, even in volume." Then there was the production/shipping/customer management disaster that was G1G1 2007.

So, now they're "[p]assing on the development of the Sugar Operating System to the community." Since we're a FLOSS community, this is a good thing, right? Well, if you consider the cessation of Sugar development good, then it is. Sugar likely will not evolve past this point, except maybe for a few baby steps. And it's a pity, because as the world adapts to daily computer use, it's crying out for a new UI paradigm. (Sorry for buzzword overload; and anyone who argues that current UIs are the best we can reasonably achieve prior to the advent of everyday 3D computing displays may be right.)

The Windows blunder was just that, and an epic one. Microsoft won that round and is now laughing on its way to the bank. Now the majority of EEEs even come with some variant of Windows CE, right?

So, the XO-2 laptops are, I guess, in final design and pre-production stages. They look amazing. But they won't work in laptop mode worth a damn. Nothing comes close to clicky-key feedback. A touchpad keyboard will lead quickly to frustration, which will lead even more quickly to a lot more force being applied to the touchpad surface than it was designed to handle, I imagine. This major shortcoming of expectations aside, it looks to be an amazing learning/textbook/interaction platform, and very much along the lines of the early goals of OLPC. But now they have to implement it with a 50% RIF, and all others on reduced salary (i.e., reduced work hours). As Dr. Phil would say, "How's that workin' out for ya'?"

Combine all these factors, and it's easy now to predict that XO-2 will never be produced -- unless the design, tech, manufacturing capacity, etc. are bought by a commercial entity for production. And that seems very unlikely with the current downturn and its effect on the tech sector. Sugar is now at a dead-end. Then the only thing left is XO-1 production, and one wonders how long that will hold out. Is funding still coming in? Do they still have working contracts with international education departments?

And then there's that new "no-cost connectivity program." Huh? I'm sure that'll be a piece of cake to set up in sub-Saharan Africa and Afghanistan. Right? (Do you hear the knocking at the door? That's reality calling....)

I guess if anyone out there is still interested in G1G1, you better get 'em while the gettin's good.... *sigh*

to post comments

Changes at OLPC

Posted Jan 8, 2009 3:21 UTC (Thu) by csigler (subscriber, #1224) [Link] (18 responses)

Bad form to reply to oneself and all that, I know, but I forgot to make Yet Another Point:

OLPC is no longer developing Sugar. Now it's been turned over to the community. Well, with the RIF they probably don't have the manpower to develop it themselves any more. This is understandable from that standpoint.

But, if they aren't developing Sugar, how much longer will they be _using_ Sugar? How can they use Sugar as their major UI platform if they can't depend on needed improvements or even fixes being made to it in a timely manner? It seems obvious to me that, like it or not, in short order every OLPC shipped will be loaded with Windows.

Changes at OLPC

Posted Jan 8, 2009 7:38 UTC (Thu) by dlang (guest, #313) [Link] (17 responses)

they haven't been running sugar development for several months already.

if they drop sugar I would expect them to switch to a plain linux desktop, bot switch to windows.

switching to windows increases their costs (the microsoft license fees, the cost of the SD card), and limits the useability of the XO (since the SD card is in use for the OS it can't be used for content), never mind the maintinance and support issues (anti-virus/spyware software for everything, and keeping it up to date, etc)

normally I would also refer to performance, but the current (and older) Sugar builds are bad enough that it is a wash, once new builds get in place that have far better performance I expect this to matter again

Changes at OLPC

Posted Jan 8, 2009 9:34 UTC (Thu) by ngiger@mus.ch (subscriber, #4013) [Link] (16 responses)

If OLPC (the foundation) would switch to Windows they would loose at least two third of the community.

There are people inside OLPC which consider the turnout as a chance to get the community much more involved. See what Samuel Klein thinks about it under http://wiki.laptop.org/go/It%27s_a_New_Day

Changes at OLPC

Posted Jan 8, 2009 10:20 UTC (Thu) by metasj (guest, #56000) [Link] (1 responses)

That article you link to on the olpc wiki was written by Yama Ploskonka, not by me! I do appreciate that perspective and have pointed to it since. --Samuel

Changes at OLPC

Posted Jan 8, 2009 10:41 UTC (Thu) by ngiger@mus.ch (subscriber, #4013) [Link]

Sorry, Samuel. I was looking at http://wiki.laptop.org/go/User:Sj, found the article. Just now I see, that you put in the "Wiki schedule" all new articles, not just yours.

Changes at OLPC

Posted Jan 8, 2009 16:24 UTC (Thu) by Alterego (guest, #55989) [Link] (13 responses)

The current concern at OLPC is olpc survival, no more Children and human Development. This is sad, but shows how big is the failure.

The choice of "one machine for each children" sounds good to rich westerners, but shows the ignorance of what is really needed in developing countries, and what are the human develpment processes.

http://www.gapminder.org/ (great slides with UN stats about the world)

Also, one very bad hidden idea is that rich westerners think they know how to educate poor southern children.

And oh, don't forget to get back to basis in economy:
- primary development (agriculture) is needed before
- secundary dev (industry)
- then tertiary (services)

=> stop US an EU agriculture subventions which are the root cause of misery in the world, by preventing local primary economy to work.

Sorry to be a bit hard, but if we want to stop misery (that causes terrorism ...) and help our poorer brothers, we need to learn, and make clean in front of our own house before anything else.

Changes at OLPC

Posted Jan 8, 2009 17:54 UTC (Thu) by dlang (guest, #313) [Link] (12 responses)

education helps primary development

'give a man a fish and you feed him for a day, teach a man to fish and you feed him for a lifetime'

the OLPC project isn't a services or industry project, it's an education project using technology.

take a look at this video if you haven't already seen it.

OLPC Mission: Part 2 The XO Laptop
http://www.youtube.com/watch?v=qMeX2D4AOjM

Changes at OLPC

Posted Jan 8, 2009 20:40 UTC (Thu) by Alterego (guest, #55989) [Link] (11 responses)

I have spent 3 years in Senegal, Gambia, Burkina Faso, working with small NGO and schools in villages and tried to do something with IT there. Mostly i teached them the basis of windows 98 administration, and ms word, because it was the only choice, and the _service_ was incredibly useful to them, no mater of the technological choice. (eg, doing nice looking document to get funds from micro-credit banks for development project in agriculture or health...)

I would have prefered to install and teach them linux, but the hardware was too weak (often 16MB of RAM, and dsl was not usable at that time). One of our shame is to send them our old computers, when the transport cost are rather high once you include the local transport problem which is incredibly difficult and expensive when there are no roads nor trains (20mph is a good average speed for transport in Africa).

But the main benefit is for us: we transfer pollution cost to them and have good feeling of doing something nice. This is totally wrong because the proportion of working computers ending in educational system is very low.

"teach a man to fish and you feed him for a lifetime"
They know this proverb too, and are fed up by foreigners who knows nothing of their life but want to explain everything and teach them.
Please, try to learn agriculture with a computer, when you have no good iron tool to build irrigation, no money for seeds, nor horse to work in field, no knowledge of local plants and soil, cows that produce one liter of milk per day, and tell me your results.

http://www.youtube.com/watch?v=qMeX2D4AOjM
nice propaganda, but unrelated to real life, at least in Africa.
I comment some points of the video:
- preliminary, most people don't speak english, nor french, nor spanish only several of the 2000+ african tongues, so XO or whatever foreign technology is useless (assuming they know to read and write which is false for 50% of africans). The first education is done in mother tongue, then they learn one or more occidental language, which is one additional cause of scholar failure.
- "connectivity and power with solar panel" .. is nice but does not provide internet acces: when there is no power in the village, there is no internet.
- "when there is XO children go more to school". Hmm, i don't believe it. Children want to go to school, but often their parent just can not let them go, because they need the child job to eat. I saw children studying in the street under the city light because they have no electricity at home, and adult going to school with children to learn writing.
- "they do music, video, pictures" This is bs ! I asked the children in the (not so poor) family where i lived, what could i offer them. I proposed music, they just said "pffff, music" and asked for a calculator, books, paper and pens in order to have their bachelor. (i was ashamed to be so wrong ;-)

Last, OLPC with its partner ship with a closed source editor, totally went wrong, because closed source prevent your "teach a man how to fish", they can learn only to click, nothing more.
This is the biggest fault in my opinion, because feeding computer with good content is, i agree, a key part of education. What i saw, is incredibly poor library (if any) in schools, and giving them a local wikipedia and all documentation+compilers would be the best thing we can do.
Remember bandwidth is nearly zero, so XO dream of internet access shows only their ignorance. If i remember in Niger in 2003 it was 1MB/s for the backbone of the whole country, and i guess it does not have improved a lot. (this is not a typo, one megabyte per second)

To paraphrase linux coding style: XO people are not evil, they were just severely misguided.

Changes at OLPC

Posted Jan 8, 2009 20:49 UTC (Thu) by dlang (guest, #313) [Link] (10 responses)

so what I would take away from your post is that geeks should not try to send any technology or technology based things to anyone else. all we can do is to donate money to organizations that provide food to starving kids and hope that it solves things somehow. (or put their life on hold for several years and go in on the ground, which is nice when it can be done, isn't realistic for 99.99% of people)

if true that's extremely discouraging (in part, since it hasn't worked in the last 50+ years I see no reason to expect it to suddenly start working now)

Changes at OLPC

Posted Jan 8, 2009 21:48 UTC (Thu) by Alterego (guest, #55989) [Link] (9 responses)

hmm mostly yes, situation is very difficult.

Without having spent time on the ground (or maybe huge studies), we cannot help a lot, except through our excellence in our own job/knowledge. So building a cheap, low power IT solution is a good idea, but the content must be their. So sharing our open source educational tools will be very good once they will be adapted to fit local realities. AFAIK OLPC in our rich countries we don't give one computer to each child, so why should it be in poor countires ?

We should help _their_ initiatives and _their_ solutions, by providing support mailing list, or whatever they ask for. (i teached M$ when i was hoping for implementing linux). Wrt to IT, things have evolved, for example Ubuntu is at first glance an African solution, which obviously shows that linux is also suited for Africa.

If you want to get some encouraging analyse, watch
http://www.gapminder.org/videos/ted-talks/hans-rosling-te...
(UN data, explained by a clever economist who spent a lot of time in Africa)

And we can help a lot by lobbying our governments to stop the root cause of their misery = our (US and EU) agriculture subventions that prevent their agriculture to develop, and this is the first mandatory step for sustainable development. Historically all (no exception) so called "developped countries" have grown through theses steps, first primary economy and food self-providing, then secondary and last tertiary, with "protectionism" of the local economy by taxes on importations and state intervention in social welfare and infrastructures.

Probably many other actions are possible from here (see debt problem, UN funding...), but i think the best we (geeks) can do is to share what we use for our children, and help them (poor but clever southern guys) to adapt it to their needs.

Changes at OLPC

Posted Jan 9, 2009 17:32 UTC (Fri) by tzafrir (subscriber, #11501) [Link] (8 responses)

We can't solve all of their problems. So let's not even try to help them.

Changes at OLPC

Posted Jan 9, 2009 19:16 UTC (Fri) by Alterego (guest, #55989) [Link] (7 responses)

For first-aid, if you don't _really_ know what to do, the only thing to do is :
- try to avoid the worsening of the situation by signaling the accident
- call competent services to get help
- stay with the victim and speak with her while rescue is coming

A positive approach, is then to learn "first-aid" during a stage at hospital, or with firemen... and forgot all the nasty things that incompetent neighbours told to you.

Changes at OLPC

Posted Jan 9, 2009 20:13 UTC (Fri) by dlang (guest, #313) [Link] (6 responses)

so you are saying that trying to assist anyone with technology or education is worse than doing nothing.

all we should do is provide you (and people like you) with money, we aren't even qualified to evaluate if the money is being spent efficiantly.

Money and help ?

Posted Jan 10, 2009 0:16 UTC (Sat) by Alterego (guest, #55989) [Link] (5 responses)

You transform what i said into ridiculous things.

Did i ever asked for or speak of money ?
Man, i spent 3 years of my life, and nearly all my economies in western Africa for IT in small ngo and schools. I did it on my own money because it was too time consuming and difficult to get the small funds i needed, without going througth insanely dumb sponsorship, who wanted things to be done in sponsors way, when it is needed to do them in african way.

But as you speak of money, one other problem of aid to developing countries is that the funds quickly goes back to the original donators country, and only very little is spent in local economy. OLPC is a huge example : the paid developpers are in US, the hardware manufacturer in developped countries. The target country can only say "thanks, you are so kind", when it would have prefered to have local developpers doing the job. In OLPC it seems some parts are localised, so i guess there have been paid translators.
You should ask your beloved olpc team how the money was spent, i missed it on their site too.

The main things i try to share is:
- from here we understand nearly nothing, so we do bs and think we are right.
- The good way to help them is to do what they ask for. Nothing more, nothing less.

Money and help ?

Posted Jan 10, 2009 0:33 UTC (Sat) by dlang (guest, #313) [Link] (4 responses)

when you say 'do what they ask for an nothing more' who is 'they'

if it's the government, then frequently all they want you to do is to give them money to spend on luxuries.

in every case where OLPC is doing a deployment, someone in the receiving country has asked them to do so.

I commend you for taking the time out of your life to go in on the ground and work to make things better, but telling people that if they don't do that they are better off doing nothing won't make anything better.

Money and help ?

Posted Jan 10, 2009 3:10 UTC (Sat) by Alterego (guest, #55989) [Link] (2 responses)

When "they" is a OLPC leader, should we blindly accept what is said, don't ask to watch their budget, and trust their wonderfull success story ?

OLPC is now tackling sub-saharian problems.
But it seems they have not even read a 3 page document describing _THE_ official continental plan for Africa. (NEPAD.)
http://www.nepad.org/2005/files/documents/30.pdf

Does olpc fit Nepad Objectives ? No.
I cannot explain better why i say OLPC is (now) severely misguided.

Its hard to explain in small (and emotional) posts what i learnt in several years of travels, discussions, experiences and many books readings.

Maybe google will give you some hints for nepad, gapminder, Jean Ziegler work at UN, and "right for food" (refused only by one country in the world, i let you find out which one), or Joseph Stiglitz, or digg and you will find lots of dramatic total failures, or misleading projects.

If you want a sad real story where it would have been infinitely better to do nothing than impose our rules (but it was supposed to help them, it had nothing to do with our business of course), read this :
http://aithne.net/index.php?e=news&id=201&lang=0

And please, keep on trying to help other people, that's great. But get deep informations, and don't misunderstand my criticism. It's necessary and good to have criticism in order to improve.

IT can provoke big civilisation disasters, by erasing culture, that's why it is important that we (foreigners) provide only help for tech, and that the local people create the socio-cultural-educational content.

Money and help ?

Posted Jan 10, 2009 4:08 UTC (Sat) by dlang (guest, #313) [Link] (1 responses)

anything that claims to be _THE_ plan for a continent is suspicious to start with. and just because it wasn't in that plan (whenever it was drafted), doesn't mean that it's not a useful idea.

you need to investigate how OLPC is working, not just read the press reports from NN's speaches. there is heavy involvement from people in-country, they are working with and through the ministry of education (or equivalent), they aren't just drop shipping laptops to schools and declaring success

Money and help ?

Posted Jan 10, 2009 9:37 UTC (Sat) by Alterego (guest, #55989) [Link]

>> anything that claims to be _THE_ plan for a continent is suspicious to start with

Are you a troll ?
Did you check Nepad site, at least the home page and the url i gave wrt IT ?

They are poor, but certainly not stupid nor ignorant.

Money and help ?

Posted Jan 10, 2009 4:11 UTC (Sat) by Ze (guest, #54182) [Link]

>>I commend you for taking the time out of your life to go in on the ground and work to make things better, but telling people that if they don't do that they are better off doing nothing won't make anything better.

What if it's the truth? Is making yourself feel better more important than helping them? The simple fact is until people accept that aid programs have a cost on the developing country and that not all aid is helpful they will continue to be unsuccessful in the long term. (BTW this can also be applied to social services.)

My personal view when it comes to charity is that I don't donate anything (time,money,goods) unless I know where it is going. It's a philosophy I intend to follow for the rest of my life. I'd love to push some of the responsibility off onto others but mountain of historical evidence tells us that it's far too easy for aid to be wasted , or harm them. I've mainly donated time and skills , and goods that aren't needed to charity so far because I've chosen a different path than most but I'm facing the situation where I could become reasonably wealthy from my own hard work and a bit of luck , so it's something I've thought about a lot. In the long term finding out that my help was harmful even though it had the best intentions would more than counteract the temporary joy from doing it. I'd feel responsible for the harm even though it was unintended (and I would be responsible for it). So looking to the long term and taking a careful involved approach is really a win/win for everybody including me.

There are many reasons why aid fails but a lot of it has to do with exploiting the countries for resources and labour. In the long run I suspect we would be better not to exploit these countries AND give them education through information and helping them construct things using basic skills. We can shorten the time to bring them up to our technological level by showing them some of the mistakes on the way up and letting them chart a better path.

If we exploit these countries we aren't helping them and we aren't letting them choose the way they want to be ,often we are supporting a Govt that the people don't want or that isn't sustainable just so we can get what we want.

It's also no use teaching them how to build or use something if they can't replicate and maintain it themselves. There are numerous times in history when we've donated technology only for it to be useless in a couple of years when the program ends because they can't maintain it (or it's too costly to maintain it). If we give them aid in the form of goods to supplement a short term shortfall we have to be careful we don't wipe out their existing industry or make them dependent on it.