On Sep 10, 2009, at 5:16 PM, Brian Harvey wrote: >> When a procedure is redefined, existing calls may or may not be >> affected; >> (Anybody have problems with these?) > > Yes! If I redefine a procedure (including, btw, primitives) I want > the new > version to apply everywhere, including already-loaded code. This is a > requirement for interactive debugging.
By primitives I assume you mean things like `car' & friends? How does this relate to the requirement that (define car 42) shall not affect the behavior of any built-in procedure? I agree that variable redefinition (including but not limited to procedures) should be the same as `set!', but I am OK with giving implementations the ability to forbid redefinitions of language primitives. > (I think I understand why you're saying no for macros, but it makes > me sad. > It's not very jewel-like to have different redefinition rules for > different > kinds of things. It makes the language that much harder to learn.) Can you explain what this "jewel" language is? I haven't been able to find anything by that name, but perhaps I'm missing something. When I search on Google for `jewel programming language', I find a result for "Oberon – The Overlooked Jewel", but Oberon certainly isn't the language you're describing. -- Brian Mastenbrook [email protected] http://brian.mastenbrook.net/ _______________________________________________ r6rs-discuss mailing list [email protected] http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
