[Edu-sig] K-16 CS/math hybrid

[Dethe Elza]

On 10-May-05, at 5:47 AM, Arthur wrote:

> If in talking about programming and children we can only be  
> realistically
> talking about a small subset, we cannot really be talking about  
> Python as it
> is.
>
> Isn't talking about Python as it is more on-topic.  This isn't a  
> list about
> a subset of Python, after all.

When teaching Euclidean geometry, you don't start with how to  
calculate the tangent of a sphere, you teach a small subset, then a  
bigger subset, possibly taking detours to explore interesting side  
paths, then a bigger subset.  Everyone uses a subset of Python, just  
different subsets.  This is also true of C++, but at least in Python,  
once you know a decent subset, your chances of understanding other  
subsets when you encounter them is greater, and there are fewer (not  
none) ways to shoot yourself in the foot than C++ (for example).

I've said this before, but perhaps it bears repeating.  Programming  
is hard.  There are many, many things we can do to make it easier,  
and Python does a lot of this (but still has room to do more, of  
course).  But making it easier is different than making it easy.  The  
programming languages which go out of their way to protect us from  
ourselves end up being too limited to be of much interest.  The  
default Lego Mindstorms environment does this: it's too dumbed-down  
for kids.

When my kids express an interest in learning to program, I'll  
introduce them to Python the same way I learned it: in small,  
incremental steps. Which is still the way I'm learning it, five or  
six years later.  But right now my eight-year-old is more interested  
in learning POVRay (any pointers I should know about, Kirby?) and ABC  
(the text-based music notation).  And my four year old is content to  
play with the programs I write for him, or draw rocket ships.

--Dethe

values of β will give rise to dom --Unix prehistory