A brief follow-up on IEEE floating-point.

[Doug Gwyn (VLD/VMB)]

It should be noted that the C standard applies already to IEEE f.p.
implementations.  A set of principles for floating-point representation
were adopted by X3J11 in response to extended questions about this
environment.  One point of note is that there can be multiple
representations of the value zero, e.g. "minus zero" and "plus zero",
but they must compare equal.  Questions about "the nearest representable
value" can also be resolved by referring to these principles, which are
included in the first "C Information Bulletin", still so far as I know
caught up in the standards bureaucracy but available from the X3J11
mailings.