[R6RS] revised draft of record srfi

Michael Sperber sperber
Wed Aug 3 12:53:58 EDT 2005

Matthew Flatt <mflatt at cs.utah.edu> writes:

> I'd prefer that all "unspecified"s be resolved. At this point, I can
> live with either generative or non-generative records (i.e., whichever
> would lead to agreement).

I have no problem with in principle, but I think this is a bad time to
decide this particular issue: if we go with compile-time generativity,
you can't have a portable reference implementation that allows local
record definitions unless you also add local DEFINE-SYNTAX (or some
other mechanism we don't have yet).  Since we haven't decided on the
latter issue, I suggest leaving this open and keeping it in mind until
we do.

Cheers =8-} Mike
Friede, V?lkerverst?ndigung und ?berhaupt blabla

More information about the R6RS mailing list