From: Alan Kay <Alan.Kay at squeakland.org>

>P.S. Arguments that something bad but long established (such as MS
>Windows conventions) should be catered to don't have a lot of force
>for me.

I agree there is no strong need to cater to Windows UI conventions. But the
bigger issue is to whether to build upon an existing environment or build
one from scratch. Let me try a biological analogy. Squeak seems to be about
finding a newly emergent island and populating it with plants and animals.
The alternative is to find old islands that already have a rich ecology and
strive to coexist. It is true that the old islands (like MS Windows) may
have many warts (rats, mosquitos, posionous spiders, etc.) but they also
have plants you can eat, animals you can domesticate, trees for shelter etc.
Making the new islands habitable is a much larger task. The resulting island
may have a nice rationale design while colonizing old islands is a more
chaotic distributed process that lacks the elegance of a top-down design.
But maybe it is a richer, more adaptive environment.

To push this analogy further, back when Smalltalk was being designed most
islands were bare or nearly so. Today over 90% of the "islands" out there
are running MS Windows. Sure, islands can be sterilized and recolonized but
the point is that there already is an existing ecology one can join and
build upon.

I'm not arguing that Squeak will fail only that it is trying to do something
very hard and risky. I hope it succeeds.

Best,

-ken kahn ( www.toontalk.com )