Limited palettes (was RE: I want to document but I need to learn first!)

Alan Kay Alan.Kay at
Fri Mar 14 08:19:04 PST 2003

None of the limitations in etoys have *any* effect at all on really 
high quality projects by 10 and 11 year olds. That's who we wanted to 
test with over 3 years and (a) really good results happened, and (b) 
only about a third of the stuff we came up with easily covered a 
whole school year -- so there is plenty more that can be done.

I think one problem here is that you, like many adults, really want 
the next version of Hypercard with lots of features and wide range. 
This is good. That's what we want to do also, and we have been 
working on this for a few years. But this is not what etoys are 
about, as I've said many times over on this list. Etoys are an 
experimental authoring environment for kids around the age of 5th 
grade, done solely to allow us to test a bunch of ideas that seemed 
fruitful and needed testing. We made the work open source to attract 
potential colleagues, not to be a vendor (Squeak and etoys are not 
products, we are a nonprofit public benefit corporation operating on 
a shoestring for the public good, etc.)

Forgive me for saying this, but there's a certain amount of  special 
pleading and rhetoric in your recent remarks. At one point you're 
using LOGO as "something that can't easily be learned", at another 
point you're invoking Seymour against BASIC. Neither of these have 
much to do with etoys -- in part because neither has a powerful 
dynamic object system with automatic graphical update. They simply 
aren't comparable and shouldn't be compared. The real heart of the 
matter is that children with pretty minimal help can do a wide range 
of projects that are engaging and empowering to them and that we 
think are intellectually interesting in the context of "real 

The one place I agree with you is that "a new thing like Hypercard" 
(with even wider scope and higher ceilings) is what is eventually 
needed. But until that comes, a fabulous range of ideas can be pretty 
easily explored with children using etoys. (I.e. you shouldn't wait 
for the "76 Trombones" before you start a music program in a school. 
The children can sing and make instruments and a musical adult can 
bring them to very above threshold musical experiences with just 



At 9:56 AM -0800 3/13/03, Jerry Balzano wrote:
>[Alan Kay, 3/11]
>>>  So the people who have been very successful at etoys are those that
>>>  are comfortable with any limited palette they are given. This has to
>>>  be widened before we can declare etoys to be a real release.
>[Dean Swan, 3/11 (later)
>>>You know, this could also be interpreted as a feature rather than a
>>>limitation.  It's been my experience that constrained resource
>>>environments encourage "better" design.
>Dean -
>I guess I have to weigh in on the side of those who would want a wider
>palette of resources.  Interestingly, the smaller number of primitive
>commands in BASIC vs LOGO back in the 70's was used as an argument for
>BASIC (e.g. "BASIC is easier because there are fewer commands to learn");
>in fact, your comment reminded me of a rebuttal of this argument provided
>in "Mindstorms" which I quote here:
>>An example of BASIC ideology is the argument that BASIC is easy to learn
>>because it has a very small vocabulary.  The surface validity of the
>>argument is immediately called into question if we apply it to the context
>>of how children learn natural languages.  Imagine a suggestion that we
>>invent a special language to help children learn to speak.  This language
>>would have a small vocabulary of just fifty words, but fifty words so well
>>chosen that all ideas could be expressed using them.  Would this language
>>be easier to learn?  Perhaps the vocabulary might be easy to learn, but
>>the use of the vocabulary to express what one wanted to say would be so
>>contorted that only the most motivated and brilliant children would learn
>  >to say more than "hi".  This is close to the situation with BASIC.  Its
>>small vocabulary can be learned quickly enough.  But using it is a
>>different matter.  Programs in BASIC acquire so labyrinthine a structure
>>that in fact only the most motivated and brilliant ("mathematical")
>>children do learn to use it for more than trivial ends. (Mindstorms, p. 35)
>Let me, however, add the following provisos:
>(a) The "minimalist" exercise of "making do" with an artificially reduced
>set of resources to accomplish particular (selected) tasks can be quite
>valuable and satisfying, even if it is not something you would wish (esp on
>novices!) as a constant state of affairs.
>(b) Access to the relatively large set of resources should be controlled or
>at least managed so as not to overwhelm novices, and there are a number of
>ways to go about this.  HyperCard's five "user levels" constituted an
>interesting attempt at this "shielding" of beginners from unwanted
>resources, although I'm still not sure what I think of how well it worked.
>And etoys does have its language resources organized into panes, but here
>too I keep thinking a better way is on the tip of our collective tongues.
>			- Jerry
>P.S. I know I am invoking Papert a lot in my postings, but I am trying both
>to celebrate his insights and take proper warning from (what I believe to
>have been) his mistakes.  Not to confuse necessary and sufficient
>conditions, I am nonetheless hoping that we Squeakers, by being more aware
>of history (of both events and ideas), won't be doomed to repeat it.


More information about the Squeakland mailing list