[Localization] XOs for Cambodia (was...)

[Javier SOLA]

Walter Bender wrote
>> Bringing 20.000 XOs in English into Cambodia now (without
>> teacher training, and without maintenance) will only create
>> problems for everybody, and will not help one single students,
>> because they will not be able to learn how to use them.
>>     
>
> Who is intending to bring in English laptops to Cambodia? 
The information about 20.000 keyboards coming soon comes from Mrs 
Negroponte.
> The whole
> point of this thread was to find resources to help ensure we bring
> Khmer laptops into Cambodia. I still await feedback regarding the
> Khmer keyboard layout: it is currently a blocker for production of
> these machines.
>   
I don't know about this request, I just heard of this thread when 
somebody insulted me out of the blue.

There is a government standard Unicode keyboard, we manufactured the 
first 1.000 in China a couple of years ago, and them transferred the 
method to all manufacturers, who now manufacture and sell them. Graphs 
for the letters are available if you want them. There is also a standard 
driver for Linux. It does have some issues on Pango, because some of the 
characters in the keyboard (standard vowels)  generate two unicode 
code-points, and this is done through conversion to pseudo code-points 
in the locales. Most distribution should already have this, maybe not 
Fedora, I am not sure.
> Regarding the assertion that children will not be able to use them, I
> would love to see the evidence. In all of our trials over the past two
> years in almost every corner of the planet, in English and/or in a
> local language, children were able to learn to use the laptop with
> little if any supervision. (I have every confidence that children in
> Cambodia can learn just as readily as children in Thailand or Peru.)
>   
It is hard for me to believe that meaningful learning can happen in a 
foreign and unknown language . Children might be able to handle the 
laptops, but you must interiorize concepts for meaningful learning, and 
relate them to something that you know already, which is iin your onw 
language (Piaget's constructivism theory, often mentioned by 
Negroponte). If you have to learn a new concept, and it has a new name 
that does not relate to your knowledge (an English word), you might 
memorize it, but you will not learn it (it will go away very quickly).  
This is the whole theory behind localization of software.

If you think that this is not necessary, then why is XO doing localization?

I have not seen any studies on results from Peru or Thailand. Are they 
available somewhere?
> -walter
>
> (PS: the children around the world do most if not all of the
> maintenance themselves as well.)
>   
Repair a broken screen? Reinstall a system? With what resources? What 
spare parts? Who will pay for replacement batteries? It is already hard 
to repair computers in rural areas, in spite of the 20 hours of 
maintenance training that we give teachers...  for 8 year olds without 
training it MUST be a little more difficult.

Again, maybe I am wrong. Are there any written evaluations on this?

Javier