Hi,
here are my votes.
> I'm putting these questions for discussion and vote by R7RS-large WG
> members, which at the moment means "anyone who wants to discuss and vote".
> Votes go to <[email protected]>
I'm using the reformulated numeric tower ballot text from the WG2 list, which I
think is better (I found the requirement of IEEE 64-bit pairs for inexact
complex numbers
to be hard to give a thoughtful vote on):
https://groups.google.com/d/msg/scheme-reports-wg2/TB88tInT5Zc/oLUdMeoMyTEJ
> 1) Should R7RS-large require arbitrarily large (up to implementation
> restrictions like memory) exact integers?
Yes.
> 2) Should R7RS-large require support for exact rational numbers?
Yes.
> 3) Should R7RS-large require support for exact complex numbers?
Yes.
> 4) Should R7RS-large require inexact complex numbers?
Yes.
Rationale: I think R7RS-large should aim high for numerical computation that is
portable across implementations. When voting yes for (1), I find it logical for
an implementation to support (2) and (3) as well, as they could be implemented
in terms of (1). The same goes for (4): If an implementation supports inexact
numbers, I would surely expect it to support inexact complex numbers also,
even if they may not be used often.
Looking at
http://www.r6rs.org/final/html/r6rs/r6rs-Z-H-6.html#node_sec_3.4
it seems that R6RS requires the full numeric tower, but only exactness for
integers and
rationals. That would mean an all-yes vote here should be backwards-compatible
with it, no?
--
Christian Stigen Larsen
_______________________________________________
Scheme-reports mailing list
[email protected]
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports
