Subject: Re: Theory #51 (superior(?) programming languages)
From: Erik Naggum <>
Date: 1997/01/24
Newsgroups: comp.arch,comp.lang.lisp,comp.lang.scheme
Message-ID: <>

* David Hanley
| Then why did't they flunk out of the SE class?

why do you think they had an SE class?

haven't you noticed that those who have software engineering and have a
computer science education don't usually write software in C?

| If I know integer overflow will not happen in a number crunching loop,
| I don't want to pay because you happen to like defaulting to bignums!

I'm sorry to see the level to which people feel the need to resort, now.

the myth you're believing and trying to perpetuate is that Common Lisp
cannot be efficient because it defaults to safer than C.  however, you can
compile with safety 0 and speed 3 and you can declare all your variables
just like you have to in C.  the result is most likely faster than C, and
at least as safe.

you should investigate something which stirs up so much emotion in you, and
try to learn some of the history and the facts surrounding your case.  if
you attack others for flunking SE classes, maybe it's time to re-read you
programming language textbook, or perhaps look further into the matter.

| I'm guesing you like scheme or lisp.  What hapens in these when the
| dynamic type is not what you expect?

you're thrown into the debugger unless you did as I suggested above,
compile everything with speed 3 and safety 0.  then you will just wind up
with junk values from the program.  it may even crash.  this is no worse
than C.  however, no sane Lisp programmer compiles everything with speed 3
and safety 0.  that is done for well-tested functions that need the extra
performance, in the way I demonstrated.

writing efficient code in Lisp takes some effort.  writing working code in
Lisp is relatively easy.  the reverse is true for C.  Peter da Silva will
have some words of wisdom to add to this statement, no doubt.

1,3,7-trimethylxanthine -- a basic ingredient in quality software.