From: ichimal@...
Date: 2003-02-04T08:15:39+09:00
Subject: Re: Local variables & blocks

Hi,

In message "Re: Local variables & blocks"
    on 03/02/04, matz writes:

> Hmm, do you mean #<non-initialized> value should be true?
> Like Scheme's #<undef>?

  What's #<undef> in R5RS?
  Is it virtual unspecific value such that "result of (if #f #f)" in
the definition of "do" in the section derived expression types?

  I thought it should be a value like "result of (if #f #f)".

Points are below:
    (*) first-class object,
    (*) not as false, at least,
    (*) to check equivalency is supported;
        (**) ==, != and nonInit? methods may be defined.

c.f.
(define x (if #f #f)) => #<unspecified> ; This external representation of
(define y (if #f #f)) => #<unspecified> ; unspecific value is conformed
(eqv? x y) => #t                        ; to scm.
(if x 1 2) => 1


>|  Maybe, I have over reactioned to the words "reasonable" and
>|"save additional test". Usually, I cannot belive those words.
>
>whose words are they?

  Not yours. These are Mr.Klemme's. You did just affirm it.

Thanks.
----
1002.(ichimal)
SUZUKI Shingo