[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Scheme-reports] ballot question #229: EQV? and NaN

On Fri, Sep 30, 2011 at 9:44 AM, Alex Shinn <alexshinn@x> wrote:
> By all means, but the ballot item was unclear and
> will require a re-vote.  In the meantime, the draft
> we submit with the call for formal comments will
> need to revert to either R5RS (unspecified) or R6RS.

Sorry, still catching up on this.  The R5RS description
of `eqv?' is completely unambiguous in requiring the
"different" semantics.

R6RS loosened the semantics slightly to make NaN
`eqv?' comparisons unspecified.

Neither the "same" semantics (used only by SCM) nor
the "same*" semantics are described by any standard.

Pending a re-vote I'll change the draft to the unspecified
(R6RS) semantics, and urge people to vote for this.  As
I stated in my rationale, this is the de facto standard


Scheme-reports mailing list