Re: JACAL, scm

[Thien-Thi Nguyen]

   From: Tom Lord <address@hidden>
   Date: Tue, 2 Oct 2001 17:22:29 -0700 (PDT)

           How much existing Guile Scheme code would be broken by making
           '() and #f, once again, the same?  How difficult would it be
           to fix that code?

my guess is that scheme idioms grown around R5RS interpretation rely on
`null?' to get essentially the latter from the former, so having `null?'
keep track of the current preferred interpretation nets a large majority
of cases (w/ the main point, actually, being that that such idioms are
therefore impervious to interpretation (so to speak, hehe) and so code
need not be changed).

tree walking uses `null?'.  what else is there?!

same thinking can be applied to a small grounp of primitives and syntax,
i figure.  implementation cost is low.

the remaining source that breaks is characterized by using `(eq?  '()
...)' when the intent is better expressed using `null?', etc.  [insert
exhaustive study here.]  for this source, local (as opposed to default)
interpretation policy must be added, and properly handled by guile.  if
being non-invasive is not required, fix can probably be added to some
forthcoming lint for guile.

thi