RE: FW: PPIG discuss: teaching kids to program

2007-08-01 Thread C.Douce 
 

*   I am really asking the question, why try to teach children
programming?

 

I really liked this question.  Perhaps another way of tacking it might
be to ask, 'why do we teach the subjects that we choose to teach
children?'  The main subjects (from my school days) include maths,
language(s), literature, sciences (physics, chemistry), and humanities. 

 

What is so special about maths?  Fearing stating the obvious, it allowed
me to discover (and understand) other phenomena, especially during my
physics and electronics classes (much of which I have sadly forgotten).
A similar argument could be said about language classes - it gave me the
vocabulary to learn the other humanities and literature subjects.  These
subjects were the fundamental tools that facilitated the further
discovery of other things.  One thing that struck me about Logo is that
it theoretically (I chose that word carefully) facilitated the discovery
of mathematical ideas and concepts in a 'fun' way.  

 

I agree that 'what you do with it' is important but I think what you
might be able to 'discover' (or develop) from it is also important.
Perhaps when it comes to tools like lego mindstorms discovering how a
sensor might be accessed by code may not be as important as discovering
new ways to communicate amongst a group of peers.

 

(I've found the following Wikipedia link that was interesting and there
was a whole bunch of other languages that I haven't heard of:
http://en.wikipedia.org/wiki/Educational_programming_language )

 

Chris

 



From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Yishay Mor
Sent: 01 August 2007 15:49
To: Ruven E Brooks
Cc: [email protected]
Subject: Re: FW: PPIG discuss: teaching kids to program

 

What does an economist do when she wants to understand complex
economical systems? She programs a model. What does a physicist do to
understand thermodynamics? Same. So yes, there is something special
about programming: its a medium which allows us to play with ideas. In
one of the papers I mentioned, we argue that programming is a medium
which affords 'mathematical narrative', and given the epistemic powers
of narrative - has a great potential for learning mathematics. 

That's not to say that I see any magical virtues in learning programming
for its own sake. Its fun, which would be enough for me, but in itself
will, of course, have zero effect. Its a question of what you do with
it. 

As for the comparative study (programming vs. Latin), I look forward to
read your report.

- Yishay

On 01/08/07, Ruven E Brooks < [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]> > wrote:


Yes, I am really asking the question, why try to teach children
programming? 
The place where Latin comes in is because Latin was taught in schools
centuries after 
it had any major value in every day life.  The argument for doing so was
because it "disciplined the mind." 

http://links.jstor.org/sici?sici=0036-6773(190504)13%3A4%3C281%3AASIFD%3
E2.0.CO%3B2-0
<http://links.jstor.org/sici?sici=0036-6773%28190504%2913%3A4%3C281%3AAS
IFD%3E2.0.CO%3B2-0> 

Walter Milner questions, as do I, whether there is any general benefit
in other areas to teaching programming. 
Yishay Mor gives some references to work that shows that doing
programming exercises can help children learn 
mathematics.  Is that because there's something special about
programming or just because it meant children 
were spending extra time on mathematics?  I would very much like to have
seen a control condition in which, 
instead of learning ToonTalk, children learned Latin by studying texts
about motion and sequences.  I wonder 
whether they might have done even better on the mathematics than the
ToonTalk group. 

Ruven Brooks 




Walter Milner <[EMAIL PROTECTED] > 
Sent by: [EMAIL PROTECTED] 

08/01/2007 03:43 AM 

To

[email protected] 

cc

 

Subject

FW: PPIG discuss: teaching kids to program

 

 

 




Is this asking WHY try to teach children programming? 
  
A possible answer would be that it does something which has a positive
transfer to other areas - and that there is no evidence that it does, or

  
It produces better commercial programmers whne they grow up - again no
evidence 
  
I'm not sure where the Latin comes in, unless the suggestion is that
trying to handle challenging natural language structures enhances the
ability to deal with formal language constructs such as a computer
program? There is evidence that bilingual or multilingual children on
average do better educationally than others. 



From: [EMAIL PROTECTED] [mailto: [EMAIL PROTECTED]
<mailto:[EMAIL PROTECTED]> ] On Behalf Of Ruven E Brooks
Sent: 31 July 2007 16:30
To: [email protected]
Subject: RE: PPIG discuss: teaching kids to program


Can anyone point me to any research results that show that teaching kids
to

Re: FW: PPIG discuss: teaching kids to program

2007-08-01 Thread Yishay Mor 
What does an economist do when she wants to understand complex economical
systems? She programs a model. What does a physicist do to understand
thermodynamics? Same. So yes, there is something special about programming:
its a medium which allows us to play with ideas. In one of the papers I
mentioned, we argue that programming is a medium which affords 'mathematical
narrative', and given the epistemic powers of narrative - has a great
potential for learning mathematics.

That's not to say that I see any magical virtues in learning programming for
its own sake. Its fun, which would be enough for me, but in itself will, of
course, have zero effect. Its a question of what you do with it.

As for the comparative study (programming vs. Latin), I look forward to read
your report.

- Yishay

On 01/08/07, Ruven E Brooks <[EMAIL PROTECTED]> wrote:
>
>
> Yes, I am really asking the question, why try to teach children
> programming?
> The place where Latin comes in is because Latin was taught in schools
> centuries after
> it had any major value in every day life.  The argument for doing so was
> because it "disciplined the mind."
>
>
> http://links.jstor.org/sici?sici=0036-6773(190504)13%3A4%3C281%3AASIFD%3E2.0.CO%3B2-0<http://links.jstor.org/sici?sici=0036-6773%28190504%2913%3A4%3C281%3AASIFD%3E2.0.CO%3B2-0>
>
> Walter Milner questions, as do I, whether there is any general benefit in
> other areas to teaching programming.
> Yishay Mor gives some references to work that shows that doing programming
> exercises can help children learn
> mathematics.  Is that because there's something special about programming
> or just because it meant children
> were spending extra time on mathematics?  I would very much like to have
> seen a control condition in which,
> instead of learning ToonTalk, children learned Latin by studying texts
> about motion and sequences.  I wonder
> whether they might have done even better on the mathematics than the
> ToonTalk group.
>
> Ruven Brooks
>
>
>
>  *Walter Milner <[EMAIL PROTECTED]>*
> Sent by: [EMAIL PROTECTED]
>
> 08/01/2007 03:43 AM
>   To
> [email protected]  cc
>
>  Subject
> FW: PPIG discuss: teaching kids to program
>
>
>
>
>
>
> Is this asking WHY try to teach children programming?
>
> A possible answer would be that it does something which has a positive
> transfer to other areas - and that there is no evidence that it does, or
>
> It produces better commercial programmers whne they grow up - again no
> evidence
>
> I'm not sure where the Latin comes in, unless the suggestion is that
> trying to handle challenging natural language structures enhances the
> ability to deal with formal language constructs such as a computer program?
> There is evidence that bilingual or multilingual children on average do
> better educationally than others.
>
> --
> *From:* [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] *On Behalf
> Of *Ruven E Brooks*
> Sent:* 31 July 2007 16:30*
> To:* [EMAIL PROTECTED]
> Subject:* RE: PPIG discuss: teaching kids to program
>
>
> Can anyone point me to any research results that show that teaching kids
> to program has any transfer to other areas?
> Last I followed this kind of thing, the results were negative - teaching
> programming doesn't have any more of
> a beneficial effect on, say, mathematics than time spent directly on math.
>
> Can anyone point me to any research that shows that kids who learn
> programming are better at it than those who
> learn it later, after you control for personality/apptitude effects?
>
> Last, but not least, what is the effect of learning Latin on learning to
> program?
>
> Ruven Brooks
>
>
>
>   *"Guzdial, Mark" <[EMAIL PROTECTED]>*
> Sent by: [EMAIL PROTECTED]
>
> 07/31/2007 09:52 AM
>
>   To
> "Enda Dunican" <[EMAIL PROTECTED]>, [email protected]  cc
>
>  Subject
> RE: PPIG discuss: teaching kids to program
>
>
>
>
>
>
>
>
> We're seeing a lot of use of both Alice and the new MIT Scratch with
> children.  We're successfully using Python for media computation with
> children as young as 11 years old.
>
> Mark
>
>



-- 
___
  Yishay Mor, Researcher, London Knowledge Lab
   http://www.lkl.ac.uk/people/mor.html
   http://yishaym.wordpress.com
   https://www.linkedin.com/in/yishaymor
   http://www.google.com/calendar/embed?src=yishaym%40gmail.com
   +44-20-7837 x5737

Re: FW: PPIG discuss: teaching kids to program

2007-08-01 Thread Ruven E Brooks 
Yes, I am really asking the question, why try to teach children 
programming?
The place where Latin comes in is because Latin was taught in schools 
centuries after
it had any major value in every day life.  The argument for doing so was 
because it "disciplined the mind."

http://links.jstor.org/sici?sici=0036-6773(190504)13%3A4%3C281%3AASIFD%3E2.0.CO%3B2-0

Walter Milner questions, as do I, whether there is any general benefit in 
other areas to teaching programming.
Yishay Mor gives some references to work that shows that doing programming 
exercises can help children learn
mathematics.  Is that because there's something special about programming 
or just because it meant children
were spending extra time on mathematics?  I would very much like to have 
seen a control condition in which,
instead of learning ToonTalk, children learned Latin by studying texts 
about motion and sequences.  I wonder
whether they might have done even better on the mathematics than the 
ToonTalk group.

Ruven Brooks




Walter Milner <[EMAIL PROTECTED]> 
Sent by: [EMAIL PROTECTED]
08/01/2007 03:43 AM

To
[email protected]
cc

Subject
FW: PPIG discuss: teaching kids to program






Is this asking WHY try to teach children programming?
 
A possible answer would be that it does something which has a positive 
transfer to other areas - and that there is no evidence that it does, or
 
It produces better commercial programmers whne they grow up - again no 
evidence
 
I'm not sure where the Latin comes in, unless the suggestion is that 
trying to handle challenging natural language structures enhances the 
ability to deal with formal language constructs such as a computer 
program? There is evidence that bilingual or multilingual children on 
average do better educationally than others.

From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of 
Ruven E Brooks
Sent: 31 July 2007 16:30
To: [email protected]
Subject: RE: PPIG discuss: teaching kids to program


Can anyone point me to any research results that show that teaching kids 
to program has any transfer to other areas? 
Last I followed this kind of thing, the results were negative - teaching 
programming doesn't have any more of 
a beneficial effect on, say, mathematics than time spent directly on math. 


Can anyone point me to any research that shows that kids who learn 
programming are better at it than those who 
learn it later, after you control for personality/apptitude effects? 

Last, but not least, what is the effect of learning Latin on learning to 
program?   

Ruven Brooks 




"Guzdial, Mark" <[EMAIL PROTECTED]> 
Sent by: [EMAIL PROTECTED] 
07/31/2007 09:52 AM 


To
"Enda Dunican" <[EMAIL PROTECTED]>, [email protected] 
cc

Subject
RE: PPIG discuss: teaching kids to program








We're seeing a lot of use of both Alice and the new MIT Scratch with 
children.  We're successfully using Python for media computation with 
children as young as 11 years old. 
  
Mark

FW: PPIG discuss: teaching kids to program

2007-08-01 Thread Walter Milner 
Is this asking WHY try to teach children programming?
 
A possible answer would be that it does something which has a positive
transfer to other areas - and that there is no evidence that it does, or
 
It produces better commercial programmers whne they grow up - again no
evidence
 
I'm not sure where the Latin comes in, unless the suggestion is that
trying to handle challenging natural language structures enhances the
ability to deal with formal language constructs such as a computer
program? There is evidence that bilingual or multilingual children on
average do better educationally than others.



From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Ruven E Brooks
Sent: 31 July 2007 16:30
To: [email protected]
Subject: RE: PPIG discuss: teaching kids to program



Can anyone point me to any research results that show that teaching kids
to program has any transfer to other areas? 
Last I followed this kind of thing, the results were negative - teaching
programming doesn't have any more of 
a beneficial effect on, say, mathematics than time spent directly on
math. 

Can anyone point me to any research that shows that kids who learn
programming are better at it than those who 
learn it later, after you control for personality/apptitude effects? 

Last, but not least, what is the effect of learning Latin on learning to
program?   

Ruven Brooks 





"Guzdial, Mark" <[EMAIL PROTECTED]> 
Sent by: [EMAIL PROTECTED] 

07/31/2007 09:52 AM 

To
"Enda Dunican" <[EMAIL PROTECTED]>, [email protected] 
cc
Subject
RE: PPIG discuss: teaching kids to program






We're seeing a lot of use of both Alice and the new MIT Scratch with
children.  We're successfully using Python for media computation with
children as young as 11 years old. 
  
Mark

Re: FW: PPIG discuss: teaching kids to program

2007-07-31 Thread Linda McIver 
I'm surprised that no-one has mentioned John Pane's PhD work yet.
John actually did research (gasp) on how kids naturally specify
problem solutions, and came up with a system that allowed them to do
all kinds of neat programming in what was, according to his research,
a fairly natural way.  HANDS, I think it was called, or that might
have been a precursor - I haven't looked at this stuff in years.

I don't know if any follow up work has been done, but the system John
came up with was pretty cute.  I don't have the references to hand,
but it was part of the Natural Programming Project - I dare say a
search of Brad Myers' pages would find it!


-- 
Dr Linda McIver
--
Ride to live...live to ride!
 
--
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Re: FW: PPIG discuss: teaching kids to program

2007-07-31 Thread pgut001 
Walter Milner <[EMAIL PROTECTED]> writes:

>A while back Papert in Mindstorms proposed that young children be taught to
>program in Logo, and that this would enhance their cognitive development and
>expose them to 'powerful ideas'.

I don't know if the programming language has as much of an impact as the kids'
age and their desire to learn programming.  From my (limited) knowledge of
child psychology I wouldn't want to try and teach anyone under the age of
about 12-13 programming, because you'd end up having to use a very constrained
concepts and programming techniques, and more or less run into a brick wall in
some areas until their mental development advances.

Secondly, you can't just turn anyone into a programmer.  In the 1980s there
were kids learning to program using 6502 and Z80 assembly language, hardly a
friendly environment, but what made it work is that they really *wanted* to do
this.  You can lead a kid to a computer, but most of them will only want to
play Counterstrike.

Peter.

 
--
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Re: FW: PPIG discuss: teaching kids to program

2007-07-31 Thread Alan Blackwell 

[EMAIL PROTECTED] said:
> A while back Papert in Mindstorms proposed that young children
> be taught to program in Logo, and that this would enhance their
> cognitive development and expose them to 'powerful ideas'. A
> key idea was recursion.

I see that "a while back" was in 1980. Can anyone offer any
research evidence from the past 27 years that might cause us to
think differently today, or did Papert get it exactly right back
then?

Alan
-- 
Alan Blackwell   Computer Laboratory, University of Cambridge
http://www.cl.cam.ac.uk/users/afb21/   Phone: +44 (0) 1223 334418


 
--
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

FW: PPIG discuss: teaching kids to program

2007-07-31 Thread Walter Milner 
Well, on
  http://blog.wired.com/geekdad/2007/07/how-to-teach-a-.html
someone named Allen suggests OpenGL.

I suggest that those contributors don't know that much about children,
nor vector calculus. ;-)

A while back Papert in Mindstorms proposed that young children be taught
to program in Logo, and that this would enhance their cognitive
development and expose them to 'powerful ideas'. A key idea was
recursion. There is a lot of evidence that many undergraduates find
recursion difficult, and that kids thought that in something like

TO PATTERN :SIZE
REPEAT 4 [ FD :SIZE RT 90 ]
PATTERN ADD :SIZE 50

that that final tail-wise recursion was a (shock horror) GOTO. Neither
could they cope with the reverse Polish ADD : SIZE 50. They enjoyed the
pretty patterns that the programs their teachers' wrote produced, but
they did not understand and could not write them themselves. So don't
teach kids to program.

{stands back and awaits the flames} 


-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf
Of Paola Kathuria
Sent: 30 July 2007 19:26
To: PPIG Discussion List
Subject: PPIG discuss: teaching kids to program

I'm came across on Lifehacker:

  Teach a kid of program (how to?)
  http://tinyurl.com/3al4o8 (howto.wired.com)
  http://blog.wired.com/geekdad/2007/07/how-to-teach-a-.html

Have any of tried to teach your young kids to program?


Paola
--
http://www.paolability.com/

 
--
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

--
PPIG Discuss List ([email protected])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/