Re: Trouble understanding define (!)

[Michael Livshin <mlivshin at bigfoot dot com>]

Marius Vollmer <mvo@zagadka.ping.de> writes:

> me> generic functions are *not* ordinary objects,
> 
> They should be, from the point of view of the module systems.

[ snip ]

well, I guess we should define what we are talking about.  me, I'm
talking about generic functions that I would want to actually use.  I
do agree that named generics are not very Schemely, but I find the
namedness important.  and I can't see what's wrong with deviating from
"core" Scheme principles where the deviation is convenient, not
conflicting with anything else and well defined (the "well defined"
thing is very important here).

> In what way are they namespaces?
> 
> They are about dispatching to different functions depending on the type
> of the arguments.  These types don't have names.

I wasn't very rigorous in my definitions, you see.  names as such
don't have to be symbols.  you can view a generic function as a kind
of "applicative namespace" where the names are lists of class objects
(or indeed any objects that can serve as specifiers).  that ok? ;)

> me> if
> me>   0. GF0 and GF1 are generic functions.
> me>   1. (eq? (generic-name GF0) (generic-name GF1))
> me>   2. (eq? (class-of GF0) (class-of GF1))
> me> then
> me>   (eq? GF0 GF1)
> 
> [ I don't think you can define `eq?' in this way, generally.  You can
>   define `eqv?' and `equal?`, but not `eq?'. ]

yes, sorry.  it should be `eqv?'.

> In Scheme, objects do not have names.
> 
> Either we stick to this, or do no longer pretend to be implementing a
> Scheme.  Which would be OK, too.  I find the thought of just leaving
> Scheme behind and following a more fleshed out standard like ANSI
> Common Lisp quite appealing.

well, I'm interested in Guile just as in an attempt to design and
implement an extended Scheme with a CLOS-like object system.  GOOPS is
the thing that got me not to lose all interest in Guile a year or so
ago.  I'm not so interested in another Common Lisp - Clisp is there,
after all.

> Guile is not a research project in
> language design, but it more and more turns out to be one.

but of course.  there are enough "just Schemes" out there already.

> - Marius

--mike

-- 
A CONS is an object which cares.
                -- Bernie Greenberg.