Biep scripsit: > But "traditional code" will never use the ref cell itself, only > its contents. Only if the cell itself is used reasoning goes down > the drain, so if one doesn't use the features there is no penalty. > Let's explore the limit..
As I said, what it limits is not the user but the implementation. An implementation has to be naive and run slow, or clever and run fast (but possibly compile slowly). Of course, you can roll your own cells and create simulated mutable variables with them. The resulting implementation will be naive. -- On the Semantic Web, it's too hard to prove John Cowan [email protected] you're not a dog. --Bill de hOra http://www.ccil.org/~cowan _______________________________________________ Scheme-reports mailing list [email protected] http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports
