[olpc-software] age target?

[Alan Kay]

Good question --

I believe that the "OLPC project" is still thinking about its specific 
aims. Earlier is better, but governments sometimes deal with politically 
hot problems (such as teenagers, job markets, etc.)

There are tradeoffs between most desirable ages for starting with children 
and what's required of the mentoring environment (usually adults) to get 
best results. The younger the better is most desirable for children, but 
the adults have to be very good to make best practices work. As the 
children get older, they start to get more set in their commonsense views 
of the world, and some of the special viewpoints (that science requires for 
example) start becoming harder and harder to acquire. OTOH, the older 
children can start to help bridge some of the gaps between what adults are 
willing to learn and what children are willing to learn. We've tried most 
ages over the last 35 years, and have gradually homed in on "4th to 6th 
graders" (about 9 - 12 years old) as a pretty good compromise. They can do 
quite a bit with less than ideal adult guidance. But depth experiences are 
always a problem without good mentors.

So most of the stuff on http://squeakland.org is aimed at this age group. 
To bring you up to date on what we've been doing with children the last 10 
years or so, here are two white papers about Squeak Etoys that you can go 
to directly: http://www.squeakland.org/pdf/etoys_n_learning.pdf , 
http://www.squeakland.org/pdf/etoys_n_authoring.pdf . There is an excellent 
book by Kim Rose and BJ Conn that covers a sequence of about a dozen 
projects that were tested for three years on 5th graders. The system is 
multilingual, and there are installations in many parts of the world.

Interestingly (in general and wrt the HDLT project), a very large number of 
the actual uses in the US and the world do not follow the "learn science 
and math" stuff, but are used as ways to solve the often more limited aims 
of the particular teachers (for "computer literacy", "creativity", etc.). 
We believe this happens primarily because most teachers of this age range 
have very limited science and math backgrounds. I think this will be a huge 
problem with the HDLT as well. It will be immediately used for "human 
built-ins" (such as communication, stories, games, etc,) but the use for 
learning "non-natural knowledge" (like science, mathematics, etc.) will 
languish. We've seen this pattern very strongly in the use of the web in 
US, Europe and Japan, and there is no reason to suppose that things will 
somehow automatically be better in the 3rd world. So I think that 
"mentoring" (in all the forms that we can devise) is the most important "UI 
problem" to be solved on the HDLT.

Cheers,

Alan

At 04:38 AM 3/21/2006, Joshua N Pritikin wrote:
>A million units is a drop in the bucket for a country like India.  Is
>the OLPC project aiming for a specific age group for the first year?
>For example, sixth grade?
>
>I didn't see this question in the wiki or FAQ.  Answering this question
>might help focus discussion of other questions.
>
>--
>Make April 15 just another day, visit http://fairtax.org
>
>--
>olpc-software mailing list
>olpc-software at redhat.com
>https://www.redhat.com/mailman/listinfo/olpc-software