From ...
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news.tele.dk!small.news.tele.dk!129.240.148.23!uio.no!nntp.uio.no!ifi.uio.no!not-for-mail
From: Erik Naggum <erik@naggum.no>
Newsgroups: comp.lang.lisp
Subject: Re: Why no call with current continuation?
Date: 02 Oct 2002 23:21:15 +0000
Organization: Naggum Software, Oslo, Norway
Lines: 64
Message-ID: <3242589675749194@naggum.no>
References: <3D98D523.CE0C2B95@melbpc.org.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Trace: maud.ifi.uio.no 1033600876 1741 129.240.65.5 (2 Oct 2002 23:21:16 GMT)
X-Complaints-To: abuse@ifi.uio.no
NNTP-Posting-Date: 2 Oct 2002 23:21:16 GMT
Mail-Copies-To: never
User-Agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.2
Xref: archiver1.google.com comp.lang.lisp:42799

* Tim Josling
| I am wondering why there is no call/cc in lisp. It seems to be potentially
| useful.

  Please get back to us when it is /actually/ useful.

| For example, in the book 'On Lisp', a significant part of this excellent
| book seems to consists of reinventing various cut down versions of
| call/cc.

  Well, duh.

| (Call/cc is a way to save the current state of the program in a
| variable. You can use the variable later on onto resume processing at
| that point. It can be used for coroutines, exception handling,
| nondeterministic search, etc.)

  None of us know Scheme at all, so thanks for this lecture.

| To my naive point of view this seems to be contrary to the usual lisp
| approach.

  In your next life, learn a real Lisp before Scheme.  This will change
  what you consider "usual" as well as actually help grasp a hell of lot
  more than you do now.

| Three possibilities as to why call/cc is not in lisp occur to me:
| 
| 1. For some reason it is not seen as useful. But to me it seems useful.

  That is because you have yet to learn how useful the Common Lisp
  constructs are.  A common affliction of Scheme users is to believe that
  their ignorance of Common Lisp gives them a right to tell people about
  Scheme, which they know.  This is actually nuts.

| 2. It is too hard to implement efficiently - period. I don't know enough
| about implementing functional languages to say for sure, though it does
| look like it might be a challenge to implement e.g. maybe you need to
| copy the stack.

  You do no implement languages with call/cc with ordinary stacks.  Stacks
  are normally used to contain call frames.  This works well because calls
  are strictly hierarchical.  With call/cc, this is no longer true, so call
  frames need to be allocated from the heap and garbage-collected instead
  of the stack memory being trivially reusable.

| 3. It can be implemented efficiently in scheme, but is too hard to
| implement efficiently in lisp. I can't see any reason why this would be
| the case. Lisp is bigger than scheme but the core of the language does
| not look that much bigger to me.

  Do you actually believe this nonsense to be true?

| Can anyone shed any light on this? If someone wants to use this as a
| springboard for a lisp versus scheme discussion, please start another
| thread.  (In my view lisp and scheme both have good reasons to exist.

  I think both antibiotics and bacteria have good reason to exist, too.

-- 
Erik Naggum, Oslo, Norway

Act from reason, and failure makes you rethink and study harder.
Act from faith, and failure makes you blame someone and push harder.