[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Alan Kay on the meaning of "object oriented"



> I didn't know that Alan Kay originally coined the term "object
> oriented programming". It seems he was *very* close to our way of
> thinking back in those early days (late 60's).
>
> ----
> - I thought of objects being like biological cells and/or individual
> computers on a network, only able to communicate with messages (so
> messaging came at the very beginning -- it took a while to see how to
> do messaging in a programming language efficiently enough to be
> useful).
> ----
>
> Too bad it went so wrong
>
> http://userpage.fu-berlin.de/~ram/pub/pub_jf47ht24Ht/doc_kay_oop_en
>
> Tom

Hi Tom,

Whoever put in method libraries, polymorphism and inheritance (with the
resulting focus on structures, and not functioning processes, as objects)
was the one who diverted it. It makes it OK for GUI programming, as long
as you are not too demanding (e.g. of consistent behavior), but it results
in the equivalent of trying to define building architecture using the
categories of interior decorating. (I laughed when I recently read an old
Maigret mystery, "Maigret sets a trap," where a major character was an
"architect-interior decorator" which meant that he could not do
architecture.)

I too am puzzled about how to get out of the wrongness. There is terrific
pressure for conformity, but I find that trying to work in those
categories makes me crazy. It's as if they offered a "sports communicator"
object which included sports sections of the newspaper, sports channels on
the TV, sports radio, sports games on game consoles, and distinguished it
from a "news communicator" object which included news sections of the
newspaper... You have to fight the design to get anything done.

Larry Dickson

>
>
>