[Edu-sig] Teaching Middle-School Math with Python
Fri, 6 Oct 2000 18:11:08 -0400
> Tell me more about Python Logo. I know the standard Python comes with
> a turtle class as one of the Tk demos. Is this what you mean?
PythonLogo: I'm glad you asked :-)
Yes and yes++
I thought maybe __someone__ might have developed a very cool PythonLogo
already.. but elif not then why not develop one. And if so, without becoming
a slave to the old, go forward [emulate what is great about it] = make
something new and pythonic [adapted to and in harmony with its environment].
I believe the most important thing about Logo is the idea:
Turtle-centric thinking with two interconnected interpreters for feedback:
one graphic, the other command line.
Given Python's obvious talents for interactive objects, lists, reusable
structures.. and Logo key insights into psychology of learning, and proven
[can we take that as a given ?] effectiveness.
Some diverse brainstorming on this one would benefit CP4E I hope - whatever
[Google results 7,030 for turtle geometry. Search took 0.14 seconds]
There is some valuable research available to be synthesized and reviews, but
not necessarily integrated.
There are some keen and experienced minds to consult with
adhoc to mind come these refs:
- 'Turtle Geometry' the book
- Andy diSessa's subsequent work [up to the strange 'Boxer']
- Microworlds [is that its name - the multi super processing Logo]
- strange mindwarping 3D Game of Life
[Not Logo exactly but somehow quite connected. There was an Amiga and I
think I saw about 7 years ago..]
- GeoLogo = a global and spherical Logo
[This last was my vision after reading 'Turtle Geometry' around 1992 - I
entered into some correspondence at the time about it with a Math teacher /
Musician / JForth Music programmer Nick Didkovsky, based in New York. He
had been actively using Logo with his class to some neat stuff. I saw a
flyer for a recent concert of his band Dr.Nerve. So I know he's around and
thriving. Time to look him up again]
Four key points of the GeoLogo idea were:
1. Action on a spherical world leads to all sorts of valuable geo-spatial
Screen is flat, Cartesian geometry is flat, Logo applies a toroidal logic to
flat screen [Turtles reappear top-bottom left-right..pretty strange!!]
As Bucky noted in 'Utopia or Oblivion' and myriad other places, spherical
Behaviors led to group Turtle interactions - thus turtles could have
properties etc, and also genetic and memetic* traits -leading to learning
and social structure families being the direct genetic trait, society being
the combined genetic plus memetic behavioral ones.
3. Maps and environments
GeoLogo would use basic logo functions to draw environment maps. The maps
would themselves have properties so that when the turtles for example
reached water or a cold region or source of food, or a road/route they could
be programmed to behave differently.
The idea was to have a Logo which could start off very simple. But which
could also scale up to something much richer more collaborative, distributed
hopefully with much more encapsulation possible. Influence of the early
Sim-City and A-life [Sim-World..] but keeping the focus on the core values
interpreter shell, scripts and graphic feedback
4. And among interesting distributed ideas which was much harder then to
implement or articulate was message passing between Turtles. How to network
these together internally ? How to handle multiple sessions across a
Micro-worlds have addressed some of this I think. 3-D graphics displays
system easily handle this now.
One might use something like Pyro to implement distributed message passing.
To handle really effective geo-spherical drawing one needs a toolkit
designed for such.
This way all is Geo-referencing from the beginning.
Then the scale-dependent FLAT world SCREEN world artifacts we handle
everyday are put in their correct context - Zoom in Zoom out. You Zoom in
enough - flat enough for you?
The only and best SDK I know of for doing this is and with great precision
is Hipparchus system.
We would need to do the Python embedding and implementation...Any SWIG
I did not know back then how much was involved to put this together or even
how much of it is be possible now - or advisable. But it seems like it _is_
now possible, and still a good idea.
The trick would be to have a broad enough vision and project map but start
off modestly step at a time building towards an enduring system. Clearly
without curriculum and teachers this is hard. A good system is one of the
teachers. There is a lot of existing Logo material and experience which
could hopefully be used. But perhaps some resistance too. cries of WYANL.
PythonLogo says it plainly enough. Because it is PythonLogo it should still
be able to do all the other cool python stuff too without missing a beat:
>>>import series, precalc
Perhaps could get even it to run on a PlayStation2 with some backing from
Sony to port Python and develop it. Throw in some sound [not necessarily
music, could be multilingual tutorials]
and you have one cheap fast hot global curriculum machine.
[ intriguing article about the PlayStation2 team in WiReD some months ago,
says they might be up for this.]
[*October Scientific American has a fascinating article on memes]