Justin wrote:

> It is a fun language as well as being a great scientific tool.
> In fact it is too perfect, too specific:

Well,IMHO: Squeak is not perfect, it is fun fun fun, 
that's why we choose to live with these little bugs.
Well, sometimes squeakers challenge eachother to kill the most bugs:
Ï did five with one hand.."
But I also saw an email where someone felt a pity,
because someone else killed "his" adopted bug earlier.

> The DESIRE for action transcends the NEED for DESIGN.

No, IMHO: designs become to complex to keep the overview in your head,
That's why we want to live in a toolbox that helps to organise our problem,
and delivers us views from the perspective we choose, 
leaving out the distracting detials from other views.
...
Most of all we want GOOD FEEDBACK ON OUR ACTIONS inside our problemspace: 
only then we can learn quicker from our mistakes,
that's why this forest of helpbrowsers did grow: 
beginners see them as the cages in the Squeak-zoo...
And then someone jokes that we should enter the cages...
And even if you do not enter a cage but just feed one animal,
the llama of Squeak spits you in your face: Message not understand:
blablabla
But then at last we experience the euphoria:

              YES YES, I SOLVED IT MYSELF.


You still have a point that Squeak invites you to action,
even before you think of organising your newdesign


> The clue to smalltalks specific failure lies in its very 
> specific strength; poly-morph-ism (parent to child).

WHat do you mean by that(see also lower on this page)?

> In the bad ole days, the command to hate, was the "goto".
Blame that other Dutchman Dijkstra

> Now, it seems some people need to restore some thing that 
> looks like a goto.
> Why? Is it because they often find themselves on a wrong 
> "branch", cannot afford to backtrack

Problem with backtracking: you do not know when you made THE wrong choice,
becuase most of the time it is a combination of wrong choices.

....

> Now Gary that is a veeerry serious design problem.
> The designers of smalltalk overlooked iso-morph-ism (auties 
> and uncles).


Maybe I inderstand these concepts wrong:
For me polymorphism is a kind of (design)management by exception:

- I cry to some objects draw yourself and they all run to the canvas and do
it.

(I know from studying their pedigree that somewhere during Squeak-evolution,
a branch grow that can draw itself. Well, that's for me one of the big
problems:
I only have time to study the tree in freetime. I also discover
that there is more then one way - not only my way - to organise such a tree,
and I get suprised by places where objects can hide in the tree, puzzling
why there...)

- Now, ONLY when some of these objects draws itself very slow or ugly on the
canvas,
ONLY THEN I am interested in the inside of that object: 
I look at its structure and the resemblances in internal methodstructure 
of it's parents and siblings. (Is this what you mean with isomorphism?)
I focus on the small differences in structure and wonder which of these - or
worse -
which combination of differences did cause the drawing disaster....