No subject

[Fergus Henderson]

Received: from mundook.cs.mu.OZ.AU (mundook.cs.mu.OZ.AU [128.250.37.154]) by mulga.cs.mu.OZ.AU with ESMTP
	id OAA17360 for <mercury-users at cs.mu.OZ.AU>; Tue, 15 Sep 1998 14:29:12 +1000 (EST)
Received: (from dgj at localhost) by mundook.cs.mu.OZ.AU (8.8.5/8.7.3) id OAA14730; Tue, 15 Sep 1998 14:29:11 +1000 (EST)
Message-ID: <19980915142911.59223 at mundook.cs.mu.OZ.AU>
Date: Tue, 15 Sep 1998 14:29:11 +1000
From: David Glen JEFFERY <dgj at cs.mu.OZ.AU>
To: mercury-users at cs.mu.OZ.AU
Subject: Re: [mercury-users] re: gsat, Mercury vs functional languages, nondeterminism
References: <199809142245.KAA28050 at borg.cs.waikato.ac.nz>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.88
In-Reply-To: <199809142245.KAA28050 at borg.cs.waikato.ac.nz>; from Don Smith on Tue, Sep 15, 1998 at 10:45:11AM +1200
Status: O
Content-Length: 1087
Lines: 23

On 15-Sep-1998, Don Smith <dsmith at cs.waikato.ac.nz> wrote:
> 
> OK, so how about a pure, eager functional language?  Pure ML with linear
> types/monads. Or eager Haskell.  (Is it possible? How essential is 
> laziness for monads?)

In one sense, laziness and monads are quite orthogonal concepts. However, I
would imagine that laziness is extremely important if you are using a list
monad to handle nondeterminism. In this case, you only want solutions (list
elements) to be created as they are needed. Otherwise, for example, having an
infinite number of solutions could be really bad. Even with a finite number of
solutions, the performance hit could be arbitrarily bad, I guess, if you are
needlessly performing computation.


dgj
-- 
David Jeffery (dgj at cs.mu.oz.au) |  Marge: Did you just call everyone "chicken"?
PhD student,                    |  Homer: Noooo.  I swear on this Bible!
Department of Computer Science  |  Marge: That's not a Bible; that's a book of
University of Melbourne         |         carpet samples!
Australia                       |  Homer: Ooooh... Fuzzy.