Re: Jensen's device

scott@basis.com (Scott Amspoker)
24 Mar 1998 22:44:26 -0500

          From comp.compilers

Related articles
Call by name in Algol-60 RogerHA@aol.com (RogerHA) (1998-03-07)
Re: Call by name in Algol-60 xxx@info.lv (1998-03-15)
Re: Call by name in Algol-60 lindsay-j@rmc.ca (John Lindsay) (1998-03-18)
Re: Call by name in Algol-60 xxx@info.lv (1998-03-20)
Jensen's device xxx@info.lv (1998-03-22)
Re: Jensen's device wclodius@aol.com (1998-03-24)
Re: Jensen's device scott@basis.com (1998-03-24)
Re: Jensen's device jsm@it.dtu.dk (JørgenSteensgaard) (1998-03-24)
Re: Jensen's device wclodius@aol.com (1998-03-30)
Re: Jensen's device genew@vip.net (1998-03-30)
Re: Jensen's device mslamm@olive.mscc.huji.ac.il (1998-03-30)
Re: Jensen's device u-reddy@cs.uiuc.edu (Uday S. Reddy) (1998-03-30)
Re: Jensen's device Andrew.Walker@nottingham.ac.uk (Dr A. N. Walker) (1998-04-03)
[4 later articles]
| List of all articles for this month |

From: scott@basis.com (Scott Amspoker)
Newsgroups: comp.compilers
Date: 24 Mar 1998 22:44:26 -0500
Organization: Basis International
References: 98-03-074 98-03-124 98-03-160 98-03-193 98-03-204
Keywords: design, PL/I

xxx@info.lv wrote:


>1. When they discovered the mistake, why didn't they simply rewrite the
> document containing the definition of Algol-60? I assume, it was not
> so easy because of all that bureaucratic work :-(


Speaking of language "mistakes": I had the misfortune of taking a class in
PL/I back in the 70's. That was still a time of punched cards. PL/I
statements (on an IBM machine) had to start in column 2 or greater. This
seemed perplexing at first. What was so damn special about column 1 that we
couldn't touch it. A classmate made the connection and pointed out that the
problem was the open-comment delimiter: '/*'. Those with past experience
with IBM JCL (or OCL for you System/3 fans) will remember that a punched card
containing '/*' in column 1 was and end-of-file marker.


I often wondered if PL/I design commitee realized the conflict late in the
project and decided it would be easier to insist that PL/I statements not
start in column 1.


Scott Amspoker
scott@basis.com
http://www.rt66.com/sda
[My dim recollection is that col 1 contained some sort of preprocessor
flag, but I can easily believe that the /* problem was the real motivation.
PL/I and JCL were designed at IBM at the same time, and it would be typical
for the two groups not to talk to each other. -John]




--


Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.