is squeak really object oriented ?

Christian Hofer Christian.Hofer at
Fri May 23 15:15:28 UTC 2003

Hi Jan,

>> The questions you ask have been answered a long time ago by philosophers.
>> You are victim of positivist thinking: there is no such thing as a
>> reality of objects. Objects are the abstractions we make to get a 
>> grasp of physical reality. The medium of making abstractions is 
>> language - this is why we can make abstractions while animals don't.
>no i am not victim of positivist thinking (my opinion). objects ARE the 
>abstractions we make to get a grasp of physical reality - i agree with this

>sentence and believe in it. so perhaps, the misunderstanding is that you 
>thought i do not think like that - i think. i consider the statements in
>above paragraph (and paragraphs below) to be correct.

Then I don't understand your distinction between object-orientedness and
textual representation.

Objects are objects because we have a predicate of them. I can experience a
stone because it hurts when I take it for a football. This experience knows
no subject and object, it is just immediately there. But now I will start to
rearrange my predications, I will differentiate between the object that hurts
and the subject that can be hurt. I will learn that my old predicate of the
object was wrong etc.

The means to achieve all this is language (not a concrete language but the
principle of language), because language is the means to construct
hypothetical worlds, e.g. worlds where a stone hit with the foot does not hurt. So it is
quite natural to express ideas about objects by using language. The textual
representation is extremely powerful, because our syntax and semantics is so
unlimited expressive.

But of course this comes at a price. Everyone knows that a glance at a
visual model can sometimes replace reading a difficult explanation. Thus a
language of simple graphical symbols can sometimes be much more adequate for a
description. But at a loss of detail, of course.

But a symbol of an apple does not get more of an apple if it is an image
than if it is the word "apple". Only you can recognize it easier. The same is
not true for a symbol representing a "customer" I suppose.

So, while I think visual clues can be extremely helpful, on the other hand
it doesn't make them more like objects - this they are only on the level of
language - whichever language you choose.

Thus when it's all language, it's easy to copy "glass":
glass glass
or more abstract:
two glasses

Now I performed a symbolic transformation where I lost the immediate
reference to the real world of my desktop, where there is "this glass here". But
that is the power of mathematics, isn't it?

And in fact I gain an understanding of the glass by negating the negation
that a glass would be two glasses. To have this understanding can really be
helpful when you get drunk, though we learn it at a very much earlier stage of
our development, so one should not have problems with that. But it is
important to notice that this is something a child has to learn at a certain moment.

To sum it up:

The computer is a tool that helps us with our mental processes by performing
defined symbolic transformations - there is absolutely no reason that these
transformations by themselves have to be immediate representations of real
transformations. Important is that the result of a transformation is applicable
to reality.

Because a computer is such a powerful tool, it is really hard for every
programmer to understand the symbolic transformations one defines. Thus
metaphors, visualisations and simplifications are extremely important. But they are
not changing the character of the operations being simply symbolic


+++ GMX - Mail, Messaging & more +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!

More information about the Squeak-dev mailing list