Here is a video of the speech I gave to the DC OLPC Learning Club on December 18th. I don't actually start speaking until about 2:30 into the video. Ahead of the speech I had an awesome meeting w/ Christoph Derndorfer and Aaron Kaplan of OLPC Austria. They are doing some exciting work and there is a lot that Austria and Nepal and Austria can collaborate on. We shared a general frustration about how limited communication is b/w different grassroots OLPC organizations. I have started working on a social networking site for OLPC that will try to address some of those issues.

Just last week we had Bert Freudenberg in Kathmandu and last night I met w/ some more Deutsch-speakers. Who would think the German-speaking countries and Nepal would be some of the first movers on OLPC?

I especially want to thank Wayan Vota of OLPCNews for organizing the meetup and putting me in touch with Christoph and Aaron.

 

Here is the text of the speech for anyone who is interested in reading it.

>>>>>>>>>>>>

I am really happy to see so many people here are excited about the One Laptop Per Child Project. In fact, at this point it is fair to say that OLPC has graduated from being a "Project" to an international "Movement."

For just a few minutes, please humor me and play along. Imagine that you have decided that OLPC is an incredibly exciting and important initiative, so important that you actually move to an impoverished country to implement it. Or, you might move to an impoverished region of a developed country.

Once there, you form an OLPC Learning Group with some locals that are just as excited about OLPC as you are. This group successfully lobbies the government to conduct a pilot of OLPC. Your team actually gets some laptops and you test them with kids. Things are going great. You are living the OLPC fantasy. Cheap laptops, mesh networks, and constructionist education are going to save the world. Then you slam into harsh reality, Specifically you run into the three difficult issues.

# 1: There is hardly any content for the XO. I used to think that giving a child a laptop would turn her instantly into a constructing, expressing, analyzing superstar. Well, that has not been the case from my experiences so far. Kids like working w/ the laptop but kids often get bored after about 90 minutes w/ the laptop, some after 10. There is a real shortage of basic activities for learning basic numeracy and language skills.

# 2: Teachers don't have time for the XO. Teachers everywhere are busy, whether in Kathmandu, Washington D.C., or Vienna. Keeping up with the national curriculum leaves very little time for special projects. Further, parents and school administrators are loathe to spend time on anything that doesn't help their kids do well on national examinations. As it stands, all the existing activities on the XO are outside of the national curriculum and thus fall outside regular schoolwork and grade-level examinations.

# 3: You have to prove it works! You absolutely, positively have to quantitatively prove to policy makers that OLPC improves the overall quality of education. There is no getting around this. You can convince politicians to approve a pilot but you will not convince policymakers and the general public to make integrate OLPC into the larger education system until you can show that OLPC makes a significant improvement in kids' educations.

I have just walked you through what has essentially been my experience working on OLPC in Nepal over the last eighteen months. In the little bit of time I have I will tell you how we are dealing with these three issues in Nepal.

Before I do that, let me tell you a little bit about myself and the Open Learning Exchange Nepal. I was a career member of the US Foreign Service for 5 years, serving in the Middle East and China, before quitting this last September to work full-time on OLPC. I helped start the OLPC movement in Nepal when I arrived in Nepal in September 2006 and I co-founded the Nepali NGO Open Learning Exchange Nepal. Open Learning Exchange Nepal is working with the government of Nepal to implement the OLPC pilot there. I would love to talk later w/ anyone who is interested in learning more about our pilot strategy and implementation plan.

Enough about me, let's get back to those three killer problems.

To recap,
#1 : There isn't enough content
#2 : Teachers aren't going to use XO's if the laptops don't help them do their existing work better.
#3 : We have to prove that laptops actually improve learning.


Here is how we are trying to create a lot of high-quality content, quickly. We are using the tool Squeak, the same platform that EToys uses. With Squeak, we can develop a learning activity within a few days and literally modify a learning activity on the fly without having to recompile. Additionally, EToys is cross platform. It is absolutely essential that people who don't have XO's can develop and test our activities.

We call our methodology "Educator-Driven Development" (Blog post to come on this subject) because we have two seasoned educators on our team that drive the development process. They sketch out learning activities, critique and revise learning activities created by others, and analyze how kids and teachers react to our learning activities. The team develops learning activities in several day long to week-long iterations. We test the activities with kids on weekly basis. Our educators know both Nepal's national curriculum and more importantly how kids learn. By and large, programmers don't know how kids learn. Great Teachers know how kids learn. We really see in this how programmers consistently overestimate the difficulty of teaching advanced skills and underestimate the difficulty of teaching basic skills.

Because Squeak is a graphical drag-and-drop environment we can develop learning activities quickly and non-programmers such as educators and graphic designers can manipulate the elements within a learning activity. There is only one tool besides Squeak that offers this functionality, Adobe's flexbuilder. We are using Squeak because we feel it is inherently more powerful than Flex.

We have avoided developing learning activities using PyGTK because PyGTK applications are not particularly portable and Python code is just not "discoverable" like the graphical scripting environment in EToys or Flex. I do not foresee non-programmers hacking Python code written by others. I can and have seen non-programmers modifying fairly sophisticated EToys, such as the ones we have developed in Nepal.

Alright, on to problem #2: Teachers are busy. They are not going to use the XO unless it helps them do their jobs as specified by the existing curriculum. To manage this, we are aligning all of our learning activities with the national curriculum and for the first year we are focusing on the two subjects that Nepali students struggle the most with: Mathematics and English.

Not to be forgotten, we have to train teachers how to integrate laptops into education. This training has three dimensions:

1. How to use computers in general
2. Learn the principles behind child-centered/constructionist/experiential learning
3. How to use computers for child-centered learning.

Now to PolicyMakers: We have to show them some statistical improvement in the quality of education. This comes right back to content. By focussing on the subjects that students have the most difficulty with, we think we can show significant improvement within a short period of time.

Now I can talk almost indefinitely about the various aspects of OLPC and what we are doing Nepal. I will take your questions now and be happy to speak with you individually. Thanks for your time and thanks to Wayan Vota of OLPCNews for organizing this event.

The fist OLPC game jam was successfully held yesterday Dec 15th 2007, at Prime College in Kathmandu.

The purpose of the game jam was mainly to familiarize students, teachers and enthusiasts with squeak, its history, and how it is used to develop various applications--such as the activities developed by OLE Nepal. The event was intended to be a fun learning session rather than a programming lesson, and I can gladly say that participants enjoyed the event.

About 25 people attended the event but the most interesting and an important part that made this game jam a success was diversity. There were students of different grades, teachers with various background, the extremely talented OLE development team and expert like Bert Freudenberg who talked briefly about the history of squeak and EToys, and showed some intriguing demos that set the tone for the day.

The event was divided into three parts:

1. Introduction (about OLPC and OLE-Nepal, squeak how-to, its interface and using squeak in general).

2. Scripting and Advance scripting in squeak.

3. Exercises and Interaction

There where practice sessions and breaks in between to give everyone a chance to work with squeak, explore its features, and ask questions and interact with the team.

The whole game jam was based on a simple car game that participants had to make during the course of the day, and the simplicity with they could use features in squeak to do something like this kept everyone glued to their screen. Many even took initiative to do more; some participants were trying to control the car that they just got running with keystrokes rather than using the mouse, and another participant was trying to reduce the speed of the car once its strays away from the track that it should be running on, adding a more authentic feel to this simple game that he just made. At the end of the day everyone seems to want to learn more.

What did we learn from this game jam?

A lot ...I will list a few:

1. Everyone, especially the younger bunch was very receptive to new ideas. Many had questions, about the program, the XO, what the laptop can do, about squeak and much more.

2. Once they got the hang of using the application, they wanted more detailed (about scripting in squeak) examples.

3. Many wanted to know how they could improve themselves and what after that.

4. We provided a copy of squeak, our activities, and materials used in the workshop on a CD-ROM to each participant, and few took extra copies for their friends. This shows that a larger group of students might be interested in such events.

 

I also want to add that we intend to host a Game Jam every month for the next few months. We plan to have competitions during the jam where the winner will be decided by a panel of teachers, kids and government officials.

 

Bert answering questions.

Participants using squeak.

The first OLPC Game Jam went really well. The participant turnout was really good, all of them managed to complete the tasks we had prepared and those who were really fast were happy to find other interesting stuffs to keep them occupied.

At the end of the day most of the participants were asking us when the next 'game jam' was going to be held. That was the best complement we got

Here are a few screenshots of our new activities.

They can be downloaded from: http://ds.lahai.com/ole/DecActivities.zip

 

 

 

 

I have put a lot of work these two weeks into building a prototype library for Nepal's pilot of OLPC. It is my understanding from an e-mail conversation with S.J. that OLPC hasn't decided on a system for the library. For the time being I am much more concerned about the back-end of the library than the user interface. I don't think that it will be incredibly hard to design a simple user interface for kids to search a repository. The harder part is to find a powerful back-end that will be able to accommodate our needs as they grow over time.

After some cursory research, there appear to be three leading open-source repository systems

Eprints

Dspace -- used in OpenCourseWare, and

fedora -- not to be confused with Fedora Linux

Being the incredibly lazy person that I am, I did not go to the trouble of installing and testing each one of these repositories. Instead I spent half a day reading reviews, blog posts, and news group discussions comparing various repository packages. After reading this evaluation of the leading three repository systems and watching this video, I decided to try out fedora.

After many painful hours I got fedora set up. Actually, it is quite easy to set up fedora, which is a pure web service. I found installing the most popular UI Fez rather difficult to get set up. I see this decoupling of service and UI as a strong positive in fedora's favor. We need a very simple kid-friendly UI for kids, a more advanced one for teachers, and a very advanced one for the people who will load materials into the library.

I would love to hear from someone who actually knows about repository systems and can explain to me the benefits of one system over another. I will be in the US for 5 weeks starting Dec 16th and one of my goals will be to really understand online libraries so I can build an awesome one for Nepal.

I have put a rough install guide on how to set up fedora with the Fez UI on Ubuntu. There are already install guides for fedora and Fez but I encountered several problems during the install. 

So, I would love to hear from people who actually know about these kinds of systems.

By the way, e-Pustakalaya means "e-Library" in Nepali. The title in the upper-left of the screenshot reads the same in Devnagari script.