Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxl!ihnp4!drutx!houxe!hogpc!houti!ariel!vax135!cornell!uw-beaver!tektronix!hplabs!hpda!fortune!amd!decwrl!decvax!cca!ima!ism780b!jim
From: jim@ism780b.UUCP
Newsgroups: net.lang.c
Subject: Re: The Standard / Re: Problems With Typ - (nf)
Message-ID: <18@ism780b.UUCP>
Date: Thu, 2-Aug-84 00:17:18 EDT
Article-I.D.: ism780b.18
Posted: Thu Aug  2 00:17:18 1984
Date-Received: Sun, 22-Jul-84 03:08:49 EDT
Lines: 17

#R:fortune:-378600:ism780b:25500009:000:790
ism780b!jim    Jul 17 21:43:00 1984

> Dennis talked about this a bit at the Usenix session.  What he said was
> roughly like this:  "I had a choice of two ways to put enums in.  One was
> to make them first-class citizens, so that one could do things like
> defining arrays with enum subscripts and so forth.  The other was to
> fudge them in as a restricted form of integer.  I did it the easy but
> ugly way, and made them restricted integers."

Had Dennis dealt with his semantics in a more formal fashion, he would have
realized that his choices were to make them full class citizens or to
make them *un*restricted integers.  I sure wish he had bitten the bullet
and added subranges and arrays with range types, but I can understand
(but not condone) his reluctance to do so.

-- Jim Balter, INTERACTIVE Systems (ima!jim)