Pascal Constanza has a post that points us to an article by Oleg Kiselyow that argues against the call/cc function. Regular readers know that I am an admirer of
call/cc so I'm not sure I agree with the idea that
call/cc is a bad thing. To be fair, the article doesn't really say that. Rather, the article argues against the idea that
call/cc should be a language primitive that is used to build other control operators.
That said, the article does make good points—especially about memory requirements—and is worth a read by any Scheme lover. I still think that
call/cc is a useful and powerful construct but I'm inclined to agree that it probably shouldn't be put forth as the basis for all other control constructs in Scheme.