From meissner@lynx.cs.usfca.edu Sat Oct 23 12:22:10 1993
Received: from dons.ac.usfca.edu by dkuug.dk with SMTP id AA26551
  (5.65c8/IDA-1.4.4j for <SC22WG5@dkuug.dk>); Sun, 24 Oct 1993 03:33:27 +0100
Received: by lynx.cs.usfca.edu (AIX 3.2/UCB 5.64/4.03)
          id AA33880; Sat, 23 Oct 1993 19:22:10 -0700
Date: Sat, 23 Oct 1993 19:22:10 -0700
From: meissner@lynx.cs.usfca.edu (Loren P. Meissner)
Message-Id: <9310240222.AA33880@lynx.cs.usfca.edu>
To: FS300022@sol.yorku.ca, GWARREN@torolab2.vnet.ibm.com, SC22WG5@dkuug.dk,
        aubryp@pbs.dfo.ca, bbuckley@agb.royalroads.ca, dtm@castle.ed.ac.uk,
        jthimer@iras.ucalgary.ca, martin@ocfmail.ocf.llnl.gov,
        melanieu@vnet.ibm.com, ranga@enel.ucalgary.ca,
        supy@nanonic.ucs.ualberta.ca, waiello@vega.math.ualberta.ca
Subject: Re:  (SC22WG5.444) Just Honour for Fortran Developer, John Backus - Lest We Forget
X-Charset: ASCII
X-Char-Esc: 29


I am not truly in love with IBM either, but I think you are over-reacting.
Early Fortran was indeed a very ugly and ad-hoc language, as viewed in hindsight
from the perspective of 25 years or so of theoretical computer science and
such eminently beautiful (but rather useless) alternatives as Algol-68 and
Pascal.

By the late 1970s, however, Fortran IV was hardly under IBM control at all.
IBM's input to Fortran 77 standardization was probably less than a lot of
others -- one could mention Bell Labs as a prominent example; there were
several other major players.

Fortran through F77 was still pretty bad. However, an objective look at
F90 seems to me to reveal a lot of really good decisions. The main problem
remaining is how to get rid of the accumulated trash.

-Loren Meissner