Principles behind EToys

Hi,

Are there any papers or documents that describe the principles behind Etoys
along the style of John Maloney's article on Morphic[1]?

I am trying to understand the (conceptual) difference between a Menu and
Viewer in an object's halo. They both offer a collection of controls to act
on an object, change its properties and so on. The Viewer offers a richer way
(e.g. phrase tiles) to deal with the properties, so it could be extended to
handle all operations in the Menu halo. Do we really need the Menu?

Are categories hard-wired by design? If users could define their own category
(or move variables and scripts into existing/newly defined category) then
variables (e.g. area) could be moved into an existing (e.g. geometry) or new
category. Generic categories like "scripting" or "variables" will not be
necessary.

TIA .. Subbu

[1]
http://stephane.ducasse.free.fr/FreeBooks/CollectiveNBlueBook/morphic.final.pdf


_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

On Monday 10 September 2007 7:44 pm, Alan Kay wrote:
> Have you seen:
>
> http://www.vpri.org/pdf/NSF_prop_RN-2006-002.pdf
>
> and some of the other writings about this?
> http://www.vpri.org/html/writings.htm
Yes (though not the recent ones). I am glad that you got funds to pursue it
further. I believe we are still pretty early in the computing curve so
audacious goals :-) do have a role to play. I do believe that smallness is a
hallmark of elegance so the search is in the right direction.
> Etoys is a "demo that wouldn't die". We have tried many experiments
> in it besides making an environment for children. You will see in the
> docs above how some of these can be extended in a new from scratch
> system that will be much more comprehensive.
Often, temporary hacks outlive well-designed stuff :-). I suppose for those
who have been involved in EToys from the beginning, the difference between
principles and hacks :-) would seem very obvious. From Self to Morphic to
Etoys some nice ideas seem to have slipped between the cracks. Glad to hear
that such ideas are still in the works.
> The tradeoffs between a universal object with total polymorphism and
> complete extensions are interesting (especially given that most
> computer people are very bad at designing extensions). We are still
> thinking about what the first level of this will be like.
Extensions are indeed a hard part. Good to hear that your team is engaged on
it.

I will catch up with the new papers and see if they bridge my understanding,
Thanks .. Subbu

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland

On Monday 10 September 2007 10:44 pm, Alan Kay wrote:
> Actually ... Etoys wasn't a "temporary hack" or else it wouldn't work
> so smoothly with so many users around the world. It was a model built
> on top of Squeak with a plan to make a more comprehensive version of
> the model also serve as the base after a few years experience. The
> few years have stretched into 10 years ....
I didn't use hack in a pejorative sense but in the sense of a clever piece of
coding by an individual (or small team). Apologies if it gave a different
impression.

The player, costume, tiles, viewers are really neat ideas. The category idea
seemed incomplete. Some of the controls (e.g. fills, paint controls) came
across as expedients. Pen, canvas and color swatch, being such fundamental
and powerful graphical entities, could have been surfaced directly in object
catalog. Whether such omissions were by design or due to lack of time/funding
wasn't clear to me.

But then opinions don't need funding :-). On the whole, Etoys is an exciting
experiment. It brings back fun into programming.

Regards .. Subbu

_______________________________________________
Squeakland mailing list
[hidden email]
http://squeakland.org/mailman/listinfo/squeakland