On Sun, Jun 10, 2012 at 12:52 PM, Ray Dillinger <[email protected]> wrote: > > Due to the "vagueness," I have (since R3RS at least) > regarded streams relying on delayed functions as a bit > of magic that works, not because the implementors are > following the spec, but rather because implementors are > flatly refusing to follow a spec which permits a > completely useless result in such a way as to actually > *produce* a completely useless result. > > If you claim that the behavior people have been > implementing is something that the spec has required > all along, I think you are wrong. If you claim that > it is something the spec has permitted and ought to > have required all along, I think you are right.
I allow that "primitive procedures like cdr and +" is overly vague, but that no implementation would ever apply this extension in such a way that a portable program would break - i.e. I think the "spirit" of the extension is perfectly clear even if the wording is not. I've filed ticket #399 to clarify the wording. -- Alex _______________________________________________ Scheme-reports mailing list [email protected] http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports
