Native GUI Squeak?

Alan Kay Alan.Kay at disney.com
Sat Feb 17 21:00:23 UTC 2001


Karl --

At 8:28 PM +0100 2/17/01, Karl Ramberg wrote:
>"Jochen F. Rick" wrote:
>>  Any non-specialized interface that spends too much time trying to
>>  replace written language as the primary means of interaction is bound for
>>  failure.
>
>Most stuff I do in life are either learned by example or by being told
>how to do
>it. Like cutting bread and putting on clothes :-)
>I know that a big barrier for most people is when doing stuff
>that require reading a manual. Like setting the VCR to record at a
>certain time.
>A good GUI should help with consistency, intuition and being very
>tolerant
>of different user styles. Take for a very simplified example: pen and
>paper.
>Once the concept is grasped you expect every pen and paper to behave the
>same way.
>If it does not you usually throw out the pen or the paper and find a new one.

Then, for your first try at writing, try to avoid pens and paper made 
by Microsoft!

>But there are many variables to consider to make a GUI.
>What will the user use the system for ? How much of the available
>information do he want to know ?
>I find most of the GUI stuff in Squeak quite good for what I do today,
>and the level
>I'm doing it, like creating classes and browsing the image. But to bring
>it to the next level
>is the hard part. Lets face it, most people will never learn Smalltalk.

Well, how many people learn all of a big Smalltalk system today? To 
me the syntax of a language is p;art of the user interface design. 
That was a big concern early in Smalltalk's history and had less 
attention paid to it as it matured. But I think the trick is to make 
subset authoring environments for different kinds of users. One of 
the things we have put some thought into over the years is what kind 
of a subset would you make of Squeak to make a 21st century 
"Hypercard"? Or a 21st century "LOGO".

>It's to advanced and to complex for most people to digest and use.
>But it's a great and I think quite essential stepping stone for what
ever the next advance is.

That is our theory. The trick is not to get sucked into the "good" is 
the enemy of "better" black hole that fairly reasonable systems like 
Smalltalk or LISP create ....

Cheers,

Alan





More information about the Squeak-dev mailing list