Re: GNU Extension Language Plans

[KenD@Newton.apple.com (Ken Dickey)]

At 12:37 AM 10/25/94 -0400, ozan s. yigit wrote:
>since when the () vs #f distinction became not-terribly-significant?
>adapting this meant breaking countless programs, a lot of lost time.
>were we fools to adapt it?

ANECDOTE: Just a couple of weeks ago, someone was using a dynamic language
and was testing to see if '() was a member of a collection: {1,2,3}.  The
answer was TRUE.  This was considered a bug and the underlying (Lisp) code
was changed to return FALSE for the above test.

I personally don't like (eq? '() #f) -> #t, but I would vastly perfer to
use an extension language which looks a lot like Scheme with this
particular wart, than to use Tcl.  The extension language could be called
GEL or Schemish, or whatever.  Any language has warts in a particular
context.  The question is one of engineering tradeoff: given a range of
options in a particular problem domain, which of the choices would you
rather use?

I do feel that helping Stallman & Co use a language conformant with
IEEE/ANSI Standard Scheme has benefits, [1] because many corporate
environments allow the use only of languages with a standard by some
recognized organization and [2] it would be good to be able to run much of
the resulting code in the current Scheme environments.  On the other hand,
if the choice is between Tcl, Visual Basic, C++, etc. or GEL...

Take your pick,
-Ken