|
Motion: Make the draft consistent in distinguishing
between numbers and non-numbers (NaNs). Seconded: Jim Thomas Thanks: To Jim Thomas and Michel Hack for their significant
help in completing the text of this motion. Rationale: Much of the draft distinguishes between
numbers and NaNs. For example, see sections 7.3.1 ("if one operand is a
number and the other a 7.5.1 ("treating numbers and
NaNs alike") 7.10 ("totalOrder(number, + and 9.4 ("nextAfter is infinite
and differs from the number x") which become incorrect if NaNs are considered to be
numbers. The proposal is that we change the draft so that it
is consistent with this distinction, and that we adopt
the term “floating-point value” to refer to
the collection consisting of numbers and NaNs. One alternative,
where a the interpretation of the definition of This motion is consistent with the following
classification (prefix floating-point is assumed): value number finite
zero nonzero
normal
subnormal infinite and here is the adjusted glossary entry: 3.2.16 floating-point value: a member of a
floating-point format. Floating-point values include finite and
infinite floating-point numbers and nonnumbers (NaNs).
A finite floating-point number has three components: a sign,
an exponent, and a significand. It's numerical
value is the signed product of it significand and its radix
raised to the power of its exponent. In this standard, a
floating-point value is not always distinguished from its format
encoding. ------------- Here are changes that would clarify uses of
"number" in the draft, in line with the classification above. p10 1.2 first bullet: delete "number" p10 1.2 fifth bullet: change "numbers" to
"values" p14 3.2.15: change "number" to
"number or p14 3.2.16: change as above p15 3.2.37: change "number" to
"number or "number" to "finite number",
depending on intent p19 Table 2 title: change "numbers" to
"finite numbers" p20 5.4 para 2: change "Numbers" to
"Values" p21 5.5 para 5: change "Numbers" to
"Values" p24 Table 7 title: change "numbers" to
"finite numbers" p25 6.2 para 1: change "number" to
"finite number" p34 7.5.2 change "numbers" to
"values", twice p44 7.12 style-review change from Jim and Mike
pending p47 next to last para: "numbers" to
"numbers and NaNs" p51 i) "number" to "value" ------------- Here are changes that would clarify uses of
"value" in the draft, again in line with the classification
above. 5.3: "For any variable that has the
value zero, the sign bit provides an
extra bit of information": Jim and Warren Replacement: "For any variable
that has a zero value ..."
Michel Replacement: "For any variable that has
the numeric value zero
..." Aside: Style review should consider use of variable
here. 5.4 text for 2nd bullet c), "If 1 <= E
<= ..." text for bullet d),
"If E=0 and T^=0..." "the corresponding
representable entity value v is..." Replacement: "the representable entity v =
..." 7.3.2 description of quantize(x,y)): the result
"is a number in the same
format which has the same value as x and the
same quantum as y." Replacement: For finite decimal operands ... which
has the
same numeric value as x ...". 7.8 para 3 and 4: floating-point value (2) and
operand value (1). Jim and Warren Replacement: None needed, use of
floating- point
value is what is intended. Michel Replacement: "When the
rounded-to-integral numeric value
..." . The rationale is that NaNs are
implicitly excluded here, but included in the proposed
meaning of floating-point value. 7.12.3 3rd-last para: implied numeric,
but implication is
subtle. Jim and Warren Replacement: Ok as is; the whole
section is about
finite numbers. Michel Replacement: "As a consequence ... :
increasing the numeric
value of an internal ..." 7.12.3 2nd-last para: "source
value": implied numeric, but subtle. Replacement: In the second sentence: "The
result format's
numeric values are the numbers ..." |