[Edu-sig] Some thoughts on RurPle / TurTle

Gregor Lingl glingl at aon.at
Mon Jun 26 07:15:54 CEST 2006

Andre and me, we had an exchange of thoughts on RUR-PLE and xturtle
and their respective goals. Perhaps someone might be interested ...

On 6/25/06, Gregor Lingl <glingl at aon.at> wrote:
> Andre Roberge schrieb:
> > On 6/25/06, Gregor Lingl <glingl at aon.at> wrote:
> >
> > As long as you have a look at rur-ple and tell me what you think. ;-)
> >
> O.k., I had a short look at it, I liked it, especially the very well
> designed
> user interface. I don't have time now (before my talk at Europython and
> before having finished the second edition of my book, which will have
> significant changes because it will use my new xturtle)

Thanks for taking the time to have a look at it.  I understand
completely about the lack of time to do everything that one might want
to do.

> 20 years ago (i mentioned it somewhen) I wrote a Karel descendant
> "The very very tiny robot programming system" (in Turbo Pascal)
> for an Austrian schoolbook, which of course now is way out of date.
> It was a big success then and was used for about 15 years.

You are years ahead of me!... I just started this (and only for fun, I
don't use it myself as I don't teach) less than 2 years ago.

I just noticed your email address (and your reference to Austria) - I
had assume you were from Germany.  I had the pleasure last year of
visiting briefly your country.  I took part in a "trade mission" to
the town of Güssing to have a look at the renewable energy
infrastructure.  I live in a small rural town in Canada and we are
thinking of implementing a similar infrastructure here... but that's
another story.

> In this book Pascal was tought.  At the same  time in  Austria  Logo
> became popular and I gave seminars for teachers in Logo programming,
> and later also for students.
> So I learned  both approaches and imo there is one significant difference
> between them: The Karel-the-Robot approach has a far more restricted
> "universe" than Turtle-Graphics which has a "open universe". This is
> reflected
> (in Karel, in my tinytiny... as well as in RUR-PLE (how do you pronounce
> it?))
> in that they had and have very well designed introductory lessons with very
> specific tasks to solve.

1. I pronounce it like the colour "purple".
2. I totally agree with your statement about "open" vs "closed"
universe.  I had been thinking of including a "fifth notebook page"
within rur-ple, which would have been a turtle graphics one, where the
robot world is replaced by a turtle canvas.  That being said, I
believe that there is an advantage to having a slightly restrictive
universe initially.  I will explain below.

> In  contrast to this in the turtle-graphics approach prevails the attitude
> to let the students find their own tasks, and the programming environment
> must be thus rich that solutions can be found. Therefore I provided
> my xtx-examples, which indeed  comprise  a very  broad range of different
> problem areas.
> There is one important feature in your RURPLE (as you mentioned
> lately)  that you  do not use a special  RURPLE language but Python
> (do I rember correctly, that this is also a difference to GvR?), which
> I consider a big advantage.

This is something I believe in as well.  My goal is to provide an
introduction to programming which is as "smooth" as possible.  We
sometime hear the phrase "steep learning curve" to characterize some
difficult to grasp concept.  I think it is important to have as few
"steep learning curves" as possible in the learning process.   GvR
uses a slightly easier syntax than Python ... but at the expense of
having a "step-like learning curve" when one wants to go from GvR's
world to Python programming.  Since Rur-ple uses Python, there is no
transition to speak of.

Also, because rur-ple basic instructions [move(), pick_beeper(),
put_beeper(), turn_left(), turn_off()] are very limited, they are
easier to learn and build from, I think, than the corresponding turtle
graphics where one has to contend with arbitrary rotations and
arbitrary step lengths right from the start...  This is, I believe,
one advantage of having a "closed universe", at least at the
beginning.   What I would like to do (eventually) is to add
turtle-like features to rur-ple, with something like
move() -->  move(length)
turn_left() -->  turn(angle)
and have an open canvas.  I think this would combine the best of both worlds.

Finally, to go back to the use of Python in rur-ple, I implemented
both a procedural approach [move(), turn_left()] as well as an object
oriented approach
Reeborg = UsedRobot()

again with the idea of providing a "smooth" transition to more
advanced concepts.  I imagine this is what you do with xturtle lessons
as well.

> I, for my part, now prefer this open approach, where I try to solve
> problems that come to my mind with the given tool, if possible,
> (if not search for a better tool), instead of devising problems for
> a given device - the robot.

I agree with having a tool as versatile as possible - and you xturtle
is *excellent* in that regard.  However, not having had the benefit of
reading your book, my *guess*  (and it is only a guess) is that the
more initially restrictive environment of rur-ple provides an easier
learning environment at the very beginning.  But I could well be
wrong!    And I agree with you that, eventually, rur-ple becomes too
restrictive (certainly as compared with xturtle).

> Example: On June, 22th, the day before I released xturtle.py,
> President Bush had a visit in Vienna. He lived in a Hotel approx.
> 300 m from my house. Consequently all the streets in this
> quarter were blocked  and  I couldn't leave my house for 24
> hours without going a long way round. In this situation I felt
> the wish to produce the peace-logo graphics. It was easily
> possible using xturtle and it became the very last xtx-demo.

Peace logo: how appropriate!

> I definitely feel that both approaches are valuable and can and
> should live in parallel. I'm in no way a fundamantalist. (I just had
> to decide what approach to use for my book - and there of course
> I have to propose lessons in a well designed sequence ...)

I think that having different people exploring different approaches
may yield better ideas in the end (as long as some communication takes
place between the two)

> (I wonder if this topic were interesting also for the edu-sig list?)
I think it might be.  You have my permission to forward this message
directly to the list if you wish, and we can continue this discussion
in public, hopefully generating comments from others.

> This all said with only a very superfical impression of your system.
> Nevertheless I'd be intersted in translating it to German, if nobody
> else will undertake it, but i must ask you to understand that I'll
> certainly will not have time to do it before September/October.

I understand and appreciate the tentative offer of a translation.

> I hope that I expressed myself clearly enough - i fear my English is not
> elaborate enough for such more 'philosophical' discussions.

I believe that your English is just as good as mine, and I feel it is
more than adequate to have these type of discussions (however
frustrating it might feel when we are looking for the  appropriate
English idiom to express our thoughts.)

And if anyone complains, challenge them to carry a conversation in
German if they are not happy!  (I'll do the same for French ;-)

Thanks for your comments and your xturtle work!



More information about the Edu-sig mailing list