[plt-scheme] Re: trying to understand PLAI route-producer code...

From: keydana at gmx.de (keydana at gmx.de)
Date: Thu May 7 12:52:13 EDT 2009


thanks for your replies.

<< Do a full trace by hand using the lambda^ notation to represent
<< continuations. In fact, do this for the buggy code that precedes  
it, and
<<make sure you thoroughly understand what goes wrong there first.

I did this for the buggy code, drawing environment diagrams, and I  
think I understood why the non-mutating version does not work (in fact  
I rewrote the let/cc with call/cc to make it easier, and so I have a  
lambda expression to evaluate some environments below the one where  
real-send is bound, and when I later invoke the continuation, I get  
the old real-send closed over in this anonymous procedure... ).

So I think I understand now why mutation is needed, just unfortunately  
I don't understand how it's done (the argumentation step on the bottom  
of page 189, and the interaction between the two boxes, somehow just  
"escape me" at the moment :-;).
Perhaps I'll have to retry later, anyway at least for me the  
understanding of continuations seems to be a very "continuous" rather  
than "discrete" thing...


Posted on the users mailing list.