Compiler history (was "BREAK in C")

On Jun 29, 2009, at 9:16 AM, Olin Lathrop wrote:

> Bare Pascal is essentially unusable as a real language.

So perhaps L&R weren't so stupid to ignore it after all?  I think that  
certainly in those days one tended to dismiss or accept languages  
entirely rather than picking and choosing among the features/
constructs.)  Almost certainly Pascal would have been rejected as a  
whole (assuming K&R were even aware of it) rather than anyone thinking  
"well, pascal has a nice CASE structure that we could adopt..."

(MORE likely, IMO, is logic that went like "We're not willing to give  
up 'goto' and we are willing to live with the ugliness that implies  
without having lots of rules about how it should operate.")

==========

But that's not what I want to talk about, really.
The history of computer languages is usually presented as we've been  
doing here.
Algol, C, Pascal as near contemporaries (and following Fortran, BASIC,  
and Cobol.)
That's fine, but the piece that is missing is the "compile  
environment" itself.

The original C grew from the B compiler running in 8k of memory on a  
PDP7.  The original Pascal ran on a CDC 6000 mainframe, one of the  
largest/fastest systems available in that timeframe.  It shouldn't be  
a big surprise that the language designers had drastically different  
viewpoints on language design, or that C was willing to trade off  
"dangerous" for "simple."

There ought to be included in language histories some indications of  
what sort of computer was required to run the early implementations,  
and perhaps some additional history of when the language "made it" to  
other architectures (and with what limitations.)

BillW

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist