donald (12/08/82)
I don't think the "object orientation" discussion really belongs in net.micro, so this is my last attempt to make my stand clear. I am not against Smalltalk per se -- in fact I think that you can do some really neat things in Smalltalk. What I do object to is the hype surrounding "object orientation" as if it were a new fantastic thing unique to Smalltalk. It most certainly is not. Smalltalk may provide nice facilities for data abstraction (oh well, "object orientation" if you must), but it is by no means the *first* language or *only* language which gives these facilities. Many of the Smalltalk cult speak the Gospel of Xerox as if it was. Leave it to industry to corrupt things. Remember "structured programming" after industry got a hold on it? I've had mail proclaiming Smalltalk to herald a "new and modular" way of programming, and that Smalltalk enables us to program more "abstractly". I agree that Smalltalk's dynamic nature lets it handle many types of abstraction more easily than algol-based languages like CLU or Ada, but that's a function of it's emphasis on delayed/dynamic binding (a la LISP), not its inherent "object orientation". The algol-based languages give up delayed binding in favor of static type checking, which may seem less elegant because you have to know a lot more things at compile time. Not as flexible as Smalltalk, but a *lot* more efficient. If anybody still wants to discuss this, let's revert to mail and leave net.micro before flames and metaflames ensue. In any case, I'm bowing out for the moment. Down with buzzwords! Don Chan