From: Charles Oliver Nutter <charles.nutter@...>
Date: 2007-11-25T05:17:29+09:00
Subject: Re: retry semantics changed

Yukihiro Matsumoto wrote:
> Perhaps.  I just wanted to know what he meant in the word "dangerous".

By dangerous I mean "undetectable and unexpected side effects". The 
original behavior is "most dangerous" because retry in a piece of code I 
call can cause my code to behave differently (drastically so...I DO NOT 
want code I call to ever cause reevaluation unless I tell it to do).

The block-based behavior is less dangerous, but largely only causes 
reevaluation to the level of the retry, which is more predictable. I'm 
not sure that it *needs* to be removed, but I'd like a formal verdict on 
what the "correct" behavior should be.

- Charlie