Newsgroups: comp.object,comp.lang.eiffel,comp.lang.c++,comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!delmarva.com!newsfeed.internetmci.com!news.sprintlink.net!in1.uu.net!world!bobduff
From: bobduff@world.std.com (Robert A Duff)
Subject: Re: Why is one OO language more productive than another?
Message-ID: <DEL5JJ.6s2@world.std.com>
Organization: The World Public Access UNIX, Brookline, MA
References: <26gT74$02p@zoe.pcix.com> <26gjnv$02r@zoe.pcix.com> <DDxH8s.16K@pentagon-ai.army.mil> <41subo$27om@tigger.cc.uic.edu>
Date: Fri, 8 Sep 1995 12:04:31 GMT
Lines: 20
Xref: glinda.oz.cs.cmu.edu comp.object:38062 comp.lang.eiffel:10763 comp.lang.c++:147991 comp.lang.smalltalk:28108

In article <41subo$27om@tigger.cc.uic.edu>,
David Hanley <dhanley@okeeffe> wrote:
>	NULL pointer a kin to type error?  We must have a different
>perception of what a type error is.  What I think of as a static type
>error would be caught before a program ever ran, so I don't think that
>this example applies.

This statement seems a bit circular.  If you define "type error" to be
errors detected at compile time, then Smalltalk doesn't have any.
Certainly *some* null pointer errors can be detected at compile time.
So why is this not a type error?  And regardless of what you call it,
why is it OK for this particular kind of error to be detected at run
time (if at all!).

>...  And rapid prototyping is orthagonal to the issue
>of rapid prototyping.

This seems even *more* circular.  Must be a typo.  ;-)

- Bob
