Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uwm.edu!linac!att!ucbvax!ulysses!kpv
From: kpv@ulysses.att.com (Phong Vo[drew])
Newsgroups: comp.lang.misc
Subject: Re: Kolstad's Konjecture
Message-ID: <14378@ulysses.att.com>
Date: 26 Feb 91 17:48:09 GMT
References: <1991Feb22.211643.12151@linus.mitre.org> <1991Feb25.184544.20726@convex.com>
Organization: AT&T Bell Laboratories, Murray Hill
Lines: 17

In article <1991Feb25.184544.20726@convex.com>, tchrist@convex.COM (Tom Christiansen) writes:
> 
> I guess it's time to throw out an idea I first hear from Rob Kolstad:
> 
>     Kolstad's Konjecture: A new programming language will enjoy widespread
>     success in the UNIX community in direct proportion to that language's
>     resemblance to C; compare the relative success of awk, perl, and C++ 
>     with that of icon, scheme, and smalltalk.
> 
Partial counter-example: csh.
The rule is that if a new tool provides functions that are already
provided by existing tools, it "may enjoy more success" if it makes
"the migration path seems natural". This rule is independent of community.
csh didn't make it because it wasn't compatible to sh.
On the other hand, witness the success of ksh.
C++ and ansi-C will enjoy success not because they are compatible
to classic-C but because they make the migration path seems natural.