Hello,

I’m also against changing this. I never noticed a problem with naming the 
colours, when i asked for pink i accepted as a valid answer different shades of 
purple etc... Children like to play. Also, it would be a pity to add 
differences between simple and advanced mode.
I think the argument to make it more “evaluable” is not so relevant; with 
robotics we say we want to promote creativity and experimenting, not 
right/wrong exercises. Also it sounds to me it would be more like “over design 
the experiment until you’re certain to get the result you want”; I’d rather 
leave this strategy to people who like to evolve neural networks :P

Cheers

Fanny

Le 2/10/14 6:04 PM, « Michael Bonani » <[email protected]> a écrit :

Hi,

I found this reduce to much the feature. Playing, mixing colour is one of the 
first thing I teach to children. I prefer to ask "to switch a color , then to 
another color after some seconds". Letting them some choice is primordial for 
letting them feel they can control the robot, understand the technology.

Regards



Bonani Michael
Tél.: +41 (0) 21 693 78 69
Mobile: +41 (0) 76 349 72 14

Association Mobsya
Av. Vinet 7
1004 Lausanne
www.mobsya.org <http://www.mobsya.org>



2014-02-10 17:47 GMT+01:00 Stéphane Magnenat <[email protected]>:
Hi,


If I understand correctly, the idea would be to have "red or not",
"green or not" and "blue or not" (and the mixes)? I don't really see how
this would be an advantage. Playing with the colours can be fun, finding
the blue you want instead of having only one can bring some creativity
and can be a good pretext to make students play with VPL. I would vote no.

The point of simplifying VPL is to make the introduction to computer science, 
and its evaluation, as simple as possible. Quoting. Prof. Gärtner:
"Imagine a problem where you ask children to switch the color to blue, then to 
red after some seconds. What you want to check here is whether they get the 
logic right, and not whether they can correctly mix colors. And if for example, 
a kid does a mixture of red with some other color (they do such things for fun 
if they can), do you consider the solution correct?"

I have not seen this problem but it might be because I was not paying 
attention. What is clear is that, if we wish to test our tool in a 
scientifically-sound way, we have to be attentive of not testing something else 
than we think we test.

That said, I currently feel that this change reduces too much the feature set, 
but I would like to hear from people who have more educational experience than 
me.


kind regards,

Stéphane
_______________________________________________
Aseba-edu mailing list
[email protected]
https://mail.gna.org/listinfo/aseba-edu

Reply via email to