Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!rice!uupsi!mstr!mstr!jcm
From: jcm@mstr.hgc.edu (James McKim)
Newsgroups: comp.lang.eiffel
Subject: Re: Functions without side effects (was Old confusion)
Message-ID: <1991May31.143320.1861@mstr.hgc.edu>
Date: 31 May 91 14:33:20 GMT
References: <1991May30.141218.3446@mstr.hgc.edu> <1991May30.195249.16103@leland.Stanford.EDU>
Sender: Usenet@mstr.hgc.edu (Action News Central)
Reply-To: jcm@mstr.hgc.edu (James McKim)
Organization: The Hartford Graduate Center, Hartford CT.
Lines: 43
Nntp-Posting-Host: sc3.hgc.edu

In article <1991May30.195249.16103@leland.Stanford.EDU> craig@self.stanford.edu writes:
>In article <1991May30.141218.3446@mstr.hgc.edu>, jcm@mstr.hgc.edu (James McKim) writes:
>|> The point is pop and top are independently useful and so deserve to be separated.
>
>I agree; I never meant to suggest that Top shouldn't exist.
>
>|> And this independent usefulness comes about precisely because one changes
>|> the state and the other merely queries.
>
>No, my Stack class would have a Top method that returned the top of
>the stack and a Pop method that both removed the top element from the
>stack *and returned this element to me*.  Since I don't follow any
>dogma stating that procedures should not return values, I make my
>procedures be as helpful as possible, in this case by also returning
>what the top of the stack was.
>
>-- Craig Chambers
Well, I'm not sure how to define "helpful". As a user of a class I find
it helpful to be able to assume functions have no side effects. I don't
find it helpful to be in a situation where features A and B return 
the same value but B has a side effect (or is it A?).

As the implementer/maintainer/designer of a class I find it helpful
to be able to separate the concerns of changing state and reading state.
A feature that does both just has to be more complex than a feature
that does just one. (At least I can't think of a counterexample)

On the other hand, if there's a really compelling reason, such as
efficiency, I'll bite the bullet and design a procedure that returns
a value. (Funny, the phrase 'procedure that returns a value' sounds
so much better than 'function with side effects' :-))

So you lean toward adding return values to procedures whenever you think
it makes sense and I lean, heavily, toward functions with no side effects.

-- Jim




*------------------------------------------------------------------------------*
Jim McKim  (203)-548-2458   | _Give_ people fish and they eat for a day.  
Internet:  jcm@mstr.hgc.edu | _Teach_ people to fish and they eat for a lifetime.