[plt-scheme] 3rd-8th Grade

On Mar 19, 2006, at 8:09 PM, Matthias Felleisen wrote:

>
> On Mar 18, 2006, at 12:33 PM, Richard Cleis wrote:
>
>> I dropped in on the fifth meeting of a robot club for children in 
>> grades 3 through 8...
> Here are some _thoughts_ based on experience: drag-and-drop 
> programming has little or no positive effect on reasoning skills as 
> far as I have been able to make out in my freshmen courses. Time and 
> again, I encounter freshmen who have a variety of programming 
> experiences with such limited resources, and they are all over the 
> spectrum. It is my distinct impression that the kids who use those 
> things just don't truly understand what programming is beyond "giving 
> commands to a computer/robot/whatever". Furthermore, these systems 
> have a hard time scaling up. I have yet to encounter a good facility 
> in such systems for abstraction, which in the end is the only good way 
> to accomplish it.
>
My programming experience began on a number cruncher that looked like a 
typewriter and read pencil 'bubble' forms.  That was entirely different 
than what these youngsters were doing, so I couldn't tell how they were 
thinking (there were 9 teams of several kids).  I thought the argument 
went something like this: I wasn't learning abstract thought since I 
was only filling in bubbles to program a sequence of 
calculations...vs...The graphical approach relieves the kids of the 
details of the computer, so they can more readily explore different 
behaviors in a program.  It turns out that few new ideas were tried, 
but I am not sure that words would have made any difference.

> [Note: You could imagine a pictorial representation of combinators, 
> and that would produce abstraction with pictures but I am not sure 
> you'd feed this to kids or even ordinary programmers. S, K, I, Y and 
> friends are complex.]
>
> ;; ---
>
> I must admit that I am totally confused by Carl's response. He is one 
> of my PhD students. Perhaps he has been out of the classroom for too 
> long; perhaps the deadline stress of OOPSLA was getting to him "-)
>
> ;; ---
>
> Here is some _reasoning_ and if you want to call it prejudice, I won't 
> object. People used pictures to communicate before they had a written 
> language. I think the codified written language was a step forward, 
> allowing people to reason abstractly. This is what we eventually want 
> with "programming for kids", which should really be an attempt to get 
> them to do what math has failed us for in the last N years. At the 
> same time we shouldn't throw out everything that math has done; there 
> is just too much positive stuff there. FP, well staged, is just about 
> the right compromise in my mind.

In the robot lab, would pictures (before the written language) that 
create words be useful?  For example, would it be better if they chose 
an icon that represents a left turn, then they immediately saw 
something like

(left-motor stop)
(right-motor go)
(wait 1)
(right-motor stop)

and were then required to use the words?  It is not clear to me what 
causes reasoning to occur and develop (pictures vs words vs ?), so I 
don't feel qualified to affect how the club works.

rac

> ;; ---
>
> Last but not least. Just to make this clear, I don't object to 
> exposing kids to computers via commands and drag-drop, blob-and-bop 
> "programming" (I can't type this w/o quotes). It's better than 
> watching tv, playing video games, etc.
>
> -- Matthias
>
> _________________________________________________
>  For list-related administrative tasks:
>  http://list.cs.brown.edu/mailman/listinfo/plt-scheme