Subject: Re: lisp as a mutiple team programming language?
From: Erik Naggum <erik@naggum.net>
Date: Thu, 18 Apr 2002 13:54:02 GMT
Newsgroups: comp.lang.lisp
Message-ID: <3228126842018131@naggum.net>

* Jacques Wainer
| I have no doubts that Lisp is a great programming language for a SINGLE
| programmer, in terms of programmer productivity.  (But there was some
| postings some time ago of very competent Lisp programmers declaring that
| they are even more productive in Python.)

  Are you sure they were very competent Common Lisp programmers?  All of
  the largely disgruntled Common Lisp programmers who have argued in favor
  of Python seem to have some serious problems actually _liking_ Common
  Lisp enough to become "very competent" in them.

| But it is not clear to me whether Lisp is even a reasonable language for
| a TEAM of programmers.

  The surgical team model you find described in Frederick P. Brooks, jr:
  The Mythical Man-month appears eminently suitable for Common Lisp, with
  one chief programmer-cum-designer and a support staff who writes what
  would effectively be supporting librararies.  Even though it generally
  takes more time to write library code than application code, writing good
  specifications for them would be the chief designer's task, so he could
  write code using it.  I can imagine this to work in stages, where the
  chief programmer quickly writes a prototype with his design, then hands
  off the task of solidifying and optimizing it to an assistant, which
  would then replace the prototype seamlessly upon completion.

| I think that an important component of team programming is lets call
| deontinc programming, for the lack of a better name.

  I presume that you mean "deontic".

| Constructs that forbids and forces the other programmer to do stuff.

| Lisp is weak in forbiding constructs, and totally lacks forcing
| constucts.

  This is a good thing.  If you believe in deontic programming, it is a
  mistake to believe that (Common) Lisp will ever be able to accomodate
  you.

| Anybody has seen or done experiments, or had some experiences with Lisp
| as a multiple programmer langauge, and is there any one thinking about a
| richer set of deontic constructs in Lisp?

  Like so many others who have come up with theoretical background for
  something else and then blindly wants to force it on everything in sight,
  I think you should explain why _you_ believe in deontic logic and why you
  think it is so obviously suitable for programming teams that you have
  effectively discarded all other means of achieving the same goals.  In
  other words, it looks like you have discovered a solution and are on the
  search for a problem it can solve.  This is generally not a good idea.

///
-- 
  In a fight against something, the fight has value, victory has none.
  In a fight for something, the fight is a loss, victory merely relief.

  Post with compassion: http://home.chello.no/~xyzzy/kitten.jpg