Stability of Squeak

Richard A. O'Keefe ok at atlas.otago.ac.nz
Wed Aug 15 06:06:17 UTC 2001


I wrote:
	> It must be acknowledged that a programming language where you can do
	> 	Smalltalk keys do: [:k| Smalltalk at: k put: nil]
	> (which reliably kills Squeak 3.0) is one that is going to be subject to
	> crashes of a kind not common in other programming languages.
	
"Andrew C. Greenberg" <werdna at mucow.com> replied:
	And there's always the good 'ole:
	
		Smalltalk := nil
	
	If that's your metric for instability, guilty as charged.
	
Of course it isn't my (or anyone else's) metric of instability.	
However, this kind of thing (smashing system variables) is one way that
people can get themselves into trouble.  It's one of the best arguments
for having an operating system.  (I have rebooted a Dandelion from a
stack of floppies often enough to be convinced of this.)

Of course being able to replace the engine while travelling at 60 m.p.h.,
or whatever the speed is, is one of the things Smalltalk is _meant_ to
let you do, which is why it does it.

My suggesting was that putting some kind of safety net in the compiler
"This method sets a protected global variable, are you sure
you really want to do that?" in much the way that Interlisp-D eventually
did *might* be helpful.




More information about the Squeak-dev mailing list