New eToys [Was: Re: [Etoys] EToys question -- clearly I'm missing some key concept]

[Milan Zimmermann]

On 2007 January 26 09:19, Alan Kay wrote:
> Hi Milan --
>
> At 11:07 PM 1/25/2007, Milan Zimmermann wrote:
> >Hi Alan,
> >
> >Thanks for the article, very interesting. For the eventual delivery
> > mechanism it would be a great "trojan horse" (in a good sense of course)
> > if the plugin, when not yet installed, could install itself as 1-click on
> > the page, without having to download and install an .exe or .rpm. (That
> > is really not relevant to OLPC because it can come with the system there,
> > but great for rest of browser users).
>
> Unfortunately browsers don't work that way. If JS and DOM were as
> capable and fast as Squeak this would be easy. But the very reason we
> have to download an executable plugin to run Squeak/Etoys is that the
> capabilities of the browser are still lacking. 

I was not really advocating not using Squeak I like it :) - just a 
no-separate-download/1-click install. I guess one way to achieve 1 click (or 
no click) install of Squeak would be to make vendors ship Squeak with the 
browser like Java, but that's hard. I thougt there must be another way to 
achieve 1-click install, I wonder how Flash installs on Windows, it seems to 
install only with a click without separate download (if the plugin is not 
originally installed) but I do not know how it's achieved and whether same 
could be done for Squeak.

> To me, this is a huge 
> problem that needs to be resolved (in part because many American and
> other school children are prevented from any downloads that involve
> plugins and/or executables by the sysAdmins of their school district).

Also in the industry, so most development is defaulting on what is guaranteed 
to be "there" - HTML, JS and perhaps Flash lately ... those 3 own the 
railroads :( but it should not be that way.

>
> Last year, we did do a Logo entirely in the browser JS and DOM to see
> what could be done. This worked. So the answer is: some things but not
> enough.

Yes.

Thanks Milan

>
> Cheers,
>
> Alan
>
> >I played with Tinlizzie a few months ago it was quite captivating, and
> >amazing.
> >
> >Thanks and later, Milan
> >
> >On 2007 January 25 12:53, Alan Kay wrote:
> > > Hi Milan --
> > >
> > > Yes, it is much too early to talk about the new architecture (we are
> > > still thinking). However, it will be along the lines (but quite a bit
> > > further) of the WYSIwiki that we did as an experiment last year (and
> > > is written up in an excellent paper by Takashi Yamamiya, Yoshiki
> > > Ohshima, and Scott Wallace that was just presented at C5 in Kyoto.
> > >
> > > Cheers,
> > >
> > > Alan
> > >
> > > At 11:03 PM 1/24/2007, Milan Zimmermann wrote:
> > > >On 2007 January 23 16:01, Alan Kay wrote:
> > > > > Right --
> > > > >
> > > > > Well, the next version of Etoys ... (heh heh) ...
> > > >
> > > >trap for the impatient :)
> > > >
> > > >Alan,
> > > >
> > > >Is it too early to share some details about the architecture (I am
> > > > wondering whether the new version will be based on Morphic, Tweak or
> > > > something else, also whether existing projects are planned to be be
> > > > loadable etc)
> > > >
> > > >Thanks Milan
> > > >
> > > > > ... will make it much much easier to expose new functionality to
> > > > > the children via the Etoys interface. Currently, this is doable,
> > > > > and people do it all the time (especially in Japan, Germany and
> > > > > Spain) but it is not a smooth process and reveals that Etoys
> > > > > started life as a demo and never got re-done as a real system.
> > > > >
> > > > > The OLPC machine, besides being an impressive result just on its
> > > > > own, is also a strong forcing function for us to expand Etoys to a
> > > > > wider range of users and also to make a better architecture
> > > > > underneath (and these are in progress), Meanwhile, we have to hit
> > > > > the build deadlines of OLPC with the system we have.
> > > > >
> > > > > However, it would be great to hear from you about the project you
> > > > > actually want to do in Etoys.
> > > > >
> > > > > Cheers,
> > > > >
> > > > > Alan
> > > > >
> > > > > At 11:50 AM 1/23/2007, Steven Greenberg wrote:
> > > > > >On 1/23/07, Alan Kay
> > > > > ><<mailto:alan.kay at squeakland.org>alan.kay at squeakland.org> wrote:
> > > > > >Hi Steven --
> > > > > >
> > > > > >What you are trying to do is not Etoys, but to do something in
> > > > > >Squeak Smalltalk using one of its graphics systems (called
> > > > > > Morphic). Etoys is a UI that rides on top of Squeak Smalltalk.
> > > > > > Its objects are called "Players" and the associated graphics of a
> > > > > > player is called its "costume". Using Squeak Smalltalk, you can
> > > > > > talk to the costume of a player by saying "self costume blah
> > > > > > blah", where blah blah is a message that morphs understand.
> > > > > >
> > > > > >
> > > > > >
> > > > > >Hi Alan, thanks for the answer.  I think I actually do want EToys
> > > > > >because I want my objects to be generically scriptable using
> > > > > >tiles.  That example I asked about was chosen because it's
> > > > > > something I already know how to do in squeak :-).  It's not the
> > > > > > actual project, just a learning exercise.
> > > > > >
> > > > > >     Regards,
> > > > > >     Steve
> > > >
> > > >_______________________________________________
> > > >Etoys mailing list
> > > >Etoys at laptop.org
> > > >http://mailman.laptop.org/mailman/listinfo/etoys
> >
> >_______________________________________________
> >Etoys mailing list
> >Etoys at laptop.org
> >http://mailman.laptop.org/mailman/listinfo/etoys