[IAEP] [Grassroots-l] OLPC in Kindergarten

[Edward Cherlin]

On Sat, Jun 13, 2009 at 7:08 PM, <forster at ozonline.com.au> wrote:
> Edward wrote
>
> "I am working on a version of Turtle Art
> that will use icons rather than text labels on the tiles, and I want
> to test how much of it the preliterate can grasp, and what we can
> teach using that capability. We can demonstrate many topics in Turtle
> Art, even if the children can't yet program such demos unaided."

It's working. You can see the palettes at

http://wiki.sugarlabs.org/images/c/cd/Iconic_Turtle_Art.odt

I'll extract some Journal entries using these tiles tomorrow and post
them, along with a bundle of .svg files of tiles and palettes with
installation instructions.

Text programming is known to work in third grade, starting by showing
children how to change parameters, then how to write expressions, then
how to change logic. I know of such projects in several languages,
including Logo, LISP, SmallTalk, and APL. In a year or two they can
write their own simple programs from scratch. APL also works as a math
language starting with first-grade arithmetic. There is, of course,
the well-known problem of weird APL symbols such as × and ÷. ^_^

Nobody knows the age limit on teaching graphical programming. I make
an analogy with teaching arithmetic and elementary ideas of algebra
(associativity, commutativity, etc.) to the prenumerate using
Cuisenaire rods. I thus predict that a significant realm will work,
and that it will take a lot of patient experimenting to find out how.
I also predict that giving preschoolers a sound understanding of math
concepts without numbers, notations, and formulas will improve matters
later on when we introduce the rest of the apparatus.

> Edward, to progress discussion, I have done a mock up graphic at http://wiki.sugarlabs.org/go/User:Tonyforster
>
> I have only done some blocks. I have concentrated on turtle motion, pen and flow control.

Just so. We have very similar ideas. I used Unicode characters for my
icons in most cases, rather than draw them.

←→↑↓↗↘↶↷↺↻√∡⊥∧∨∼ and others.

> I think the mathematics blocks are best left as is.

I insist on using ÷ rather than /, and I want to try other math symbols.

> Others, eg push and pop heap may be too difficult for pre-literate children to bother with.

We know that young children can manage a stack of dishes, or a stack
of paper. So I intend to try stacks of numbers.

> Likewise the portfolio and red box tabs.

We shall see. I don't expect preschool children to program in Python, of course.

> (Maybe keyboard input is important enough to bother with but the 2 step process it uses, one block to read kyb and another for its value, is conceptually difficult. First the interface should be simplified.)
>
> Questions the exercise has raised for me:
>
> To what extent could pre-literate children use Turtle Art?

Something to test.

> Are graphics on blocks really better than text?

They don't need localization. This is important for children whose
language of instruction is not their native language. English in
Ghana, for example, or French until recently in Rwanda.

Otherwise, something to test.

> Does keeping text labels help develop literacy?

I doubt it, particularly compared with real literacy programs, such as
Same Language Subtitling.

> What graphics are best?

We must ask the children.

> Should the more difficult blocks be omitted or kept with text labels?

Test everything.

> Should they be on different tabs if retained?

Keep the current palettes for now.

> Where can we do user testing and get feedback?

Various people have offered their children for the purpose. Now we
have to ask the children what they like better.

> Tony
> _______________________________________________
> IAEP -- It's An Education Project (not a laptop project!)
> IAEP at lists.sugarlabs.org
> http://lists.sugarlabs.org/listinfo/iaep
>



-- 
Silent Thunder (默雷/धर्ममेघशब्दगर्ज/دھرممیگھشبدگر ج) is my name
And Children are my nation.
The Cosmos is my dwelling place, The Truth my destination.
http://earthtreasury.org/worknet (Edward Mokurai Cherlin)