Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!wugate!uunet!psivax!torkil
From: torkil@psivax.UUCP (Torkil Hammer)
Newsgroups: comp.software-eng
Subject: Re: Question on programming languages
Message-ID: <2865@psivax.UUCP>
Date: 3 Oct 89 00:30:10 GMT
References: <8720001@hplsla.HP.COM> <89267.181118UH2@PSUVM.BITNET> <5219@eos.UUCP> <1989Sep26.114433.515@IDA.ORG>
Reply-To: torkil@psivax.UUCP (Torkil Hammer)
Organization: Pacesetter Systems Inc., Sylmar, CA
Lines: 33

In article <1989Sep26.114433.515@IDA.ORG> rwex@IDA.ORG.UUCP (Richard Wexelblat) writes:
#
#The first ALGOL compiler I used was implemented at the U. of Grenoble.
#All "regular" keywords in English but all I/O statements in French (lire
#& ecrire, etc.)
#
Because Algol was designed as a publication language.  The idea was that
if some guy figured a really smart and reasonable machine independent
way of coding a math function (in local assembler, of course) he could
rewrite the algorithm into an international language, publish it in
BIT (or whatever) and some other guy across the ocean could pick it up
and recode it by hand for his local machine.

The publication language was Algol.  If you look at its definition, you
will find that it is truly portable - independent of any implementation.
Since I/O is implementation-only it is not needed and was not defined in
Algol.

Later, some wise guy figured out that if a human can translate accurate
and fully defined info into machine code so can a program - and the
concept of a true compiler was born.  I/O was left as an exercise
for the local gurus, and in the spirit of the 1950's it was done in
local natural language including local national alphabetic characters.
"skrvtekst", "tryktekst" and "l}stekst" come to mind.

In 1968 Algol became completely defined as a computer language,
with standard character set and standard reserved names.  The publication
history still lingers.  The syntax is free from any forethoughts on how
to parse it on a computer.  Data definitions are bitless and do not reflect
whether calculations are to be done on a digital or analog computing
device, and no assumptions are made about Algol compilers.

Torkil Hammer