Re: s-exprs + prototypes

[Steve Dekorte <address@hidden>]

On Saturday, June 21, 2003, at 03:08 AM, Michael Vanier wrote:
> Your comment brings up a question that has been nagging at the back of 
> my
> head for a while: are some language paradigms "more general" than 
> others?
> If so, which are the "most general"?  For instance, I *feel* as if the
> lisp/scheme language paradigm (essentially untyped lambda calculus) is
> "more general" than the single-dispatch OO paradigm at the heart of
> Smalltalk, but obviously both are equally powerful in a strict sense 
> (you
> can write mappings between each of them and the other).  So why do I 
> feel
> this way?  I guess it has to do with the fact that the receiver has a
> privileged status in ST, but one could argue that the first argument 
> in an
> s-expression in scheme has a privileged status (it must be a procedure 
> or
> special form).  Or else it's that ST sort of locks you in to one object
> system, whereas scheme is agnostic.  However, I could see arguments 
> that
> (say) the paradigms in prolog or haskell are in some senses more 
> general
> than those in scheme as well.
>
> Does this bother anyone else?  Or do I just need to up my medication? 
> ;-)

ST does lock you into a particular object system, but ST is a language, 
not a paradigm. There are more general object models (such as 
prototypes) that are more flexible than traditional class-based models.

OO != ST :-)

I see the OO paradigm of communication via messages between things that 
encapsulate data as more general than lists. Everything is an object, 
not everything is list.

LISP's use of a uniform representation of code(code as data) is very 
elegant, but:
1. There's nothing to prevent an OO language from having such 
uniformity.
2. Lists aren't the only choice for the base of a uniform system.

Why not use a tree as the unifying structure, for example? This would 
seem more suitable since code is semantically a tree, not a list of 
lists.

Cheers,
Steve