Re: What is OOP?

Michael S <already5chosen@yahoo.com> writes:

On Sun, 15 Dec 2024 04:24:20 -0800
Tim Rentsch <tr.17687@z991.linuxsc.com> wrote:
>

But that does not answer original question, so O.T.

>
His comment about activity and so forth is very much relevant to
his view of what "object-oriented programming" means (to the
extent that I understand what he has said and was trying to say
in that comment).

>
How many words (rough estimate) are required to answer the original
question?
>
Your answer to Janis Papanagnou approximately a week ago seems to
suggest that 10,000 words are likely to not suffice, even if those are
words of Alan Key himself.  OTOH, relatively short demos have better
chance.
Or do demos have better chance only due to 10,000 words that preceded
them?

To address the last question first, in the case of the demo that
Fred Brooks saw, he appreciated that /something/ was very different
compared to what he was used to, but he didn't yet know what it was
(and I'm sure he must have realized that).

It's hard to answer the first question, because understanding the
answer requires a change in perspective in the person asking the
question.  A story from my past:  my first exposure to programming
was in summer school, so probably 8 weeks, certainly no more than 10
weeks.  The first half, or maybe even 60 to 70 percent, was FORTRAN
(probably FORTRAN IV).  I got FORTRAN, it was no problem.  For the
rest of the session we did System/360 Assembly Language.  I was able
to get things done in assembly, and thought I understood what was
going on (and mostly did, but not everything).  Then a question came
up: go through the elements of an array without using an index
register.  I was baffled.  The mental model I had constructed for
FORTRAN did not suffice to understand assembly.  As things turned
out a short response from the instructor cleared up my bafflement,
but it wasn't just my not understanding, it was that my mental model
was wrong.  An analogy might be the change in perspective that
happened between physics in the 1800s and physics in the first half
of the 1900s (that is, after quantum mechanics).  Of course my
change in perspective wasn't as profound as what was needed for
quantum mechanics, but still a change in perspective was needed.
Fortunately one short remark from the instructor was enough to do
that for me, but it easily might not have been.  My thinking was
more fluid and malleable when I was a teenager.

The problem with trying to answer the question "What is object
oriented programming?" is that people want to understand the answer
inside a framework that they have built up over decades, and OOP in
the sense that Alan means it is a different perspective, a different
framework for thinking about programming.  For me the change
happened in a moment, and what prompted it was at most a sentence or
two.  On the other hand I had been listening to Alan in "The Alan
Kay mystery hour" for what was probably between 50 and 100 hours,
and I'm sure that made a difference, but it's hard to point to any
specific thing and say that was it.  Incidentally he never did try
to explain what he meant by the term, or really say very much about
it at all;  sometimes it would come up in passing but that was it
(in my very old and not always reliable memory).

There's a paper that was written in the late 1970s -- 1978 IIRC --
by Dan Ingalls ("Daniel H H Ingalls" may be the exact cite) about
Smalltalk, which at that time would have been Smalltalk 76.  The
paper includes a Smalltalk interpreter written in Smalltalk:  short
but with a surprising amount of detail.  I remember a paragraph near
the end where Dan tries to convey how programming in Smalltalk is
different from conventional programming.  I recommend that paper for
anyone who is interested in getting an answer to What is OOP.

One further thought.  I didn't understand the idea of object
oriented programming by having it explained to me;  I understood it
by experiencing programming in a Smalltalk-like environment.  So my
advice is to try programming a substantial project in Smalltalk.  As
much as anything else working in Smalltalk has the benefit that
there are lots of examples in the form of all the classes and
methods that make up the Smalltalk system.  The combination of
working in the environment and seeing lots of examples done by
people who have absorbed the underlying ideas has a good chance
of flipping the switch and allowing one to take a new perspective.