Hacker News Reader

Comment by bryanwb | Hacker News Reader

I worked on the one laptop per child project for three years of my life. I co-founded olpc Nepal and its succeeding organization OLE Nepal together with rabi karmacharya who is referenced in the article. I was also co editor of olpc news for a time. It was a great experience if not altogether successful. The most problematic aspect of olpc IMHO was that it was dominated by male autodidacts who looked down on professional educators. Ole Nepal from the very beginning worked very closely with Nepali educators.

Rabi continues to do great work and I am rooting for him. You can find out more about OLE Nepal here http://www.olenepal.org/

View on HN

Can you say more about the autodidacts and the conflict with professional staff? I'm interested in specific areas that were problematic, and if you have any learnings from mitigating it.

Olpc leadership saw government education bodies and curriculum groups as potential road blocks best avoided. This is with some reason as it can take years to work through then. That said, they can't be advoided in my experience if you expect to have any impact. They also disregarded the concerns of every day teachers because they presumed every kid could be an autodidact just like them once they had the magic laptop.

It all stems from Seymour papert's original vision that inspired olpc. This theory is called constructionism and held that children can develop knowledge of the world through experiential learning. They can "construct" their own knowledge. Unfortunately, the role that a teacher might need to pay on this process was often underestimated.

To counteract this, we engaged the most creative Nepali teachers (of which the re are a large number) to create educational activities that aligned with the nationial curriculum and addressed content areas where nationally Nepali kids were struggling.

>every kid could be an autodidact just like them once they had the magic laptop

So, if I understand you correctly, OLPC went in with a constructionist perspective and it didn't work out? This is hyper-relevant to what I'm working on, so: did a proper theoretical constructionist framework fail you, or was underestimating the role of teachers the main problem?

Did engaging Nepali teachers give any insights in the above?

> went in with a constructionist perspective and it didn't work out

I have no knowledge of OLPC, but it seems obvious to me: successfull autodidacts in societies where self-directed-learning is not commont (probably most societies...) tend to be internally motivated, so higher proportion of them are introverts, also higher proportions have at least little bit of aspergers-like traits etc. Extroverts on the other hand tend to learn most from human teachers they physically interact with!

If you'd crunch the numbers and compare them with the personality traits of "influencers", you'll likely see they are opposite. So any chance that a positive view of the device will spread via word of mouth is low, aka any change of whatever the equivalent of "going viral" would be amongst African villages is low!

By targeting the self-learners you're basically going anti-viral... you're doing anti-marketing! You'd need to try and hit the "micro-influencers", and probably only chance of that is by hitting teachers and some community leaders and local "celebrities".

We techo-focused hyper-individualistic self-learners only thrive in societies after they've been properly wired up both socially and technically. Drop us in a borderline-medieval society and we're useless and have zero influence on the people around us. Heck, "geeks" started to thrive in medieval Europe after the church managed in a primitive way to network part of the world. Probably similar patterns happened in China and the Arab world too. Most underdeveloped societies today totally lack that kind of useful networking ...otherwise they likely wouldn't be underdeveloped in the first place!

internal motivation == introvert == aspergers? got some stats on that?

no == there, not that strong of a claim, just "if [A], then more likely [B] than without prior [A]" ...if you have endless time (or a few PhD students you can task to research literature for free) and patience search for studies about "associations between [A] and [B]" dig and dig through things.

I avoid making a stronger claims bc it would require too much work to research it, do it yourself if you want do (dis)confirm, I'm just "throwing a bone here", too lazy to think or research more about it :)...

Ahh, so you are making these claims because you think they are true. You presented them as some sort of fact by stating that this is obvious to you. I just don't really get why it was relevant to lay out that chain of relevance. I don't see the obvious connection between self motivated and introvert, but I guess it could be possible. The other connections just seem like random thoughts. I was trying to figure out why something so obvious to you doesn't explain itself when presented without evidence...

We had to show results in a short period of time and focus on the top priorities of the parents and teachers which boiled down to belong their kids not fail out of school in the early grades. You can focus on nurturing autodidacts but you will never have more than a handful of anecdotal successes in the short run at best. And you need more than anecdotes to keep educational projects alive. Further, educators run the school. If you want an educational project to be successful, You live and die by their support.

Excellent. Appreciate the answers.

as an observer from the sidelines (i was contributing to the OLPC development but never involved in any deployment) my guess is this: if there was a proper theoretical constructionist framework then it failed because of the resistance of the existing educational system.

in other words, a constructionist model may work in a situation where kids are left to themselves, but it may not work if there is a competing educational system that doesn't allow the kids to take advantage of the opportunities the constructionist model provides.

personally though i am skeptical that a pure constructionist approach will enable the children to achieve their maximum potential, and any successes of it are in comparison to a failing traditional model.

There were many reasons the OLPC failed, but I don't think constructionist education was one of them, when it's succeeded in so many other places.

EA donated SimCity to OLPC because of its relation to constructionist education, thanks to Maxis's collaboration with Doreen Nelson, who wrote the SimCity teacher's guide, and developed "City Building Education" and "Design Based Learning", in which kids built cities out of cardboard instead of pixels:

https://news.ycombinator.com/item?id=20329281

>SimCity can be used educationally, but not in the sense of literally training people to be urban planners or mayors. It's more useful for "Constructionist Education" and "Design Based Learning", as practiced by Seymour Papert and Doreen Nelson.

>[...] One of the teachers Curtin hired was Doreen Nelson, a brilliant and innovative educator who had developed a pedagogy called City Building Education, in which students collaboratively built cities out of craft materials and role play. Nelson become a regular visitor to Maxis, and Curtin made some trips to Los Angeles to see City Building in action, where she found the experience of “watching a classroom actually go through a couple of days worth of creation” to be “very inspiring. … I will never forget that experience” (Curtin 2015; Nelson 2015). [5]

Chaim Gingold wrote a section about Doreen Nelson's work in his dissertation on "Play Design":

https://pqdtopen.proquest.com/doc/1806122688.html?FMT=AI

what i meant was not that constructionist education was a failure in itself but that it was rejected by the established education system.

Ok, I understand you now!

It's not straightforward just how to apply those theories, and takes a lot of experimentation and adaptation. And getting the established education system to change is a Sisyphean task.

Instead of thinking of SimCity as a way of directly teaching urban planning, or financing, or building construction, you can use it to motivate and indirectly develop language, logical argument, and debating skills. You can ask students to write about their cities, by describing their aspirations, proposals, platforms, campaign promises, then promoting and defending and discussing the issues, then holding elections and voting for which plans to implement, then implementing them, then discussing and writing about how they turned out.

Thanks for digging out the links on Doreen, pocketed both.

But am I reading GP wrong that it wasn't constructionist education proper that failed, but a "let's do things with laptops" being retconned onto constructivism?

at the time i had the impression that not much thought was given on how to implement education using the laptops. looking at it now, i have the impression that the expectation was that the mere existence of the laptops would allow for constructionist education to happen all by itself.

i don't know enough about constructionist education to know what factors ought to be present, but i can't imagine that an education system and teachers that do not understand or do not want constructionist education aren't a problem.

i also believe that OLPC didn't sell constructionist education to the buyers, nor that the buyers wanted to start using constructionist education.

i have the impression that it was rather hoped that constructionist education would not only happen by itself but also undermine the existing education system in that it flourishes despite the existing system.

in the end the failure was that the expectations of parents, teachers and others in the education system were not met.

one might argue that the failure was to not educate the buyers on what to expect, but then i can't imagine that OLPC would have successfully sold anything if they had tried to sell constructionist education along with it.

Here's an unboxing video of the SimCity Classic "School Edition" Lab Pack, which includes the teacher's guide by Doreen Nelson and Michael Bremer:

LGR - SimCity Educational Version Unboxing & Overview

An overview of the "School Edition" Lab Pack of SimCity Classic by Maxis. Unboxing, first impressions of the package and testing of the radically rad software ensues.

https://www.youtube.com/watch?v=edXRNtuAGTga

LGR has done many other extensive reviews of SimCity, and here's his most recent retrospective:

SimCity 30 Years Later: A Retrospective (Feb 1, 2019)

https://www.youtube.com/watch?v=TrScy1icWjI

I don't have personal knowledge of the project, but it was very heavily constructionist (or they said it was - can't comment on the implementation) so if you look around it should be easy to find talks and papers about their approach. Lots of links from here: http://wiki.laptop.org/go/Constructionist

There is a bi-annual conference about Constructionism:

https://www.constructionismconf.org/

http://www.constructionism2018.fsf.vu.lt/

http://e-school.kmutt.ac.th/constructionism2016/

https://confdriver.ifs.tuwien.ac.at/constructionism2014/

http://constructionism2012.etl.ppp.uoa.gr/

http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=607...

In many ways, the OLPC project was like "Stone Soup":

https://en.wikipedia.org/wiki/Stone_Soup

>Stone Soup is a European folk story in which hungry strangers convince the people of a town to each share a small amount of their food in order to make a meal that everyone enjoys, and exists as a moral regarding the value of sharing. In varying traditions, the stone has been replaced with other common inedible objects, and therefore the fable is also known as axe soup, button soup, nail soup, and wood soup.

The infamous crank, the $100 price, and the Sugar user interface, were among the indigestible stones.

Seymour Papert's vision of constructionist education provided some of the most nutritious meat and potatoes that nourished the soup, as did Mary Lou Jepsen's display hardware:

https://news.ycombinator.com/item?id=21045877

>Also, the OLPC's hybrid monochrome/color display that Mary Lou Jepsen designed was truly innovative, power saving, easily manufacturable, green electronics, and it was even quite efficient and crisply legible and under direct sunlight (requiring no backlight for the high resolution 200 dpi reflective grayscale LCD pixels, which could stay on while the CPU was asleep).

We were able to convince EA to relicense SimCity under GPL3 and contribute it to the project, because it was a quintessentially constructionist educational game. Without the OLPC project to rally around and rationalize the virtues of free educational software, it would have been impossible to convince EA to do that.

Open Sourcing SimCity, by Chaim Gingold. Excerpt from page 289–293 of “Play Design”, a dissertation submitted in partial satisfaction of the requirements for the degree of Doctor in Philosophy in Computer Science by Chaim Gingold:

https://medium.com/@donhopkins/open-sourcing-simcity-58470a2...

Micropolis: Constructionist Educational Open Source SimCity:

https://medium.com/@donhopkins/har-2009-lightning-talk-trans...

Contract between EA and OLPC for open sourcing and distributing SimCity:

https://donhopkins.com/home/olpc-ea-contract.pdf

Sliding offtopic, but do you have any more memories around this specific collaboration? I'm deeply familiar with the theory, but anecdotes on such an ambitious project would be useful to a text I'm working on. Thanks.

I provided Chaim Gingold with a bunch of emails and code as material for his dissertation, which he analyzed, wrote about, and quoted in the sections about SimCity. He also interviewed other people who told sides of the story I had not yet heard.

Please send me an email (contact in my profile) and describe what you're interested in, and I'd be glad to answer questions, dig stuff up, and forward it to you.

His dissertation is long and detailed (as those things tend to be) but well worth checking out! Especially all the great stuff about Doreen Nelson's lifelong work.

I don't have a direct pdf link, but if you prod and jiggle it a little bit, this pdf viewer will let you download the entire pdf:

PhD in CS dissertation on "Play Design" by Chaim Gingold, June 2016, UCSC.

https://pqdtopen.proquest.com/doc/1806122688.html?FMT=AI

I may be alittle late to the party here but came here to thank you for your service back then it was all for a good cause even if things didnt go as planned.

+1 with the thinking that content was a big problem as well

Things like wikipedia & khanh academy offline in-a-box and translated into local languages like they have now would have been a better direction of effort for OLPC in general.

[1] https://en.wikipedia.org/wiki/Wikipedia:Database_download [2] https://learningequality.org/kolibri/ (replaced KA Lite)

>>> I co-founded olpc Nepal and its succeeding organization OLE Nepal together with rabi karmacharya who is referenced in the article.

The project looks very interesting. I'm a 35-year veteran developer looking to maybe volunteer my services, and would like to have a chat with someone about how I could perhaps help out. I just sent them an email, but do you still have contacts there?

Thing is, I'm going to be in Kathmandu for 4 or 5 days from Wednesday this week, and Pokhara after that, so if someone would like to have a sit down and a chat, I'd be more than happy to do that.

I recommend contacting them directly by the contact email on the site. It is an amazing team and I am sure they will be very responsive.

Reply on news.ycombinator.com