Newsgroups: alt.lang.design,comp.lang.c,comp.lang.c++,comp.lang.lisp
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!pipex!uunet!allegra!alice!ark
From: ark@research.att.com (Andrew Koenig)
Subject: Re: Reference Counting (was Searching Method for Incremental Garbage Collection)
Message-ID: <D0AnFE.MII@research.att.com>
Organization: Software Engineering Research Department
References: <CzxAv9.HF@rheged.dircon.co.uk> <596@scribendum.win-uk.net> <19941203T221402Z.enag@naggum.no>
Date: Sun, 4 Dec 1994 16:06:50 GMT
Lines: 28
Xref: glinda.oz.cs.cmu.edu comp.lang.c:119220 comp.lang.c++:101602 comp.lang.lisp:15919

In article <19941203T221402Z.enag@naggum.no> Erik Naggum <erik@naggum.no> writes:

> Yourdon basically claims that since they got into the programming business
> much later than American programmers, they use CASE tools, which is
> Yourdon's favorite solution to the software crisis.  there is probably some
> truth to his claims, and from what I have seen of the C++ world, it is nigh
> impossible to write C++ code without a machine doing the work for you.

Really?  That's news to me.

> I have yet to see comparisons of productivity between a CASE/C++ programmer
> vs a LISP programmer, but they claim five times increase in productivity,
> and that's mostly from the dropping number of bugs, for the CASE/C++ combo,
> and ten times increase in productivity with LISP.  such figures are
> probably way too round to be accurate, but if there's truth to them, you'd
> still be way ahead with a real programming language compared to a crippled
> language, however good the crutches.

The closest thing I've seen to a believable study was by Howard Trickey,
who implemented a fairly substantial program (O(10K lines)) in Lisp and
C++ simultaneously and took notes.  His conclusion was that in that
particular context, choice of language didn't affect his productivity much.

Yes, it's only one data point and yes one can find all kinds of flaws
in the methodology, but I haven't seen any evidence that isn't at least as flawed.
-- 
				--Andrew Koenig
				  ark@research.att.com
