Re: [boost] [Boost-users] Ann: Floating Point Utilities Reviewstarts today

From: Paul A Bristow (pbristow_at_[hidden])
Date: 2008-03-03 05:15:23

>-----Original Message-----
>From: boost-bounces_at_[hidden] 
>[mailto:boost-bounces_at_[hidden]] On Behalf Of Steve Ward
>Sent: 03 March 2008 06:18
>To: boost_at_[hidden]
>Cc: boost-users_at_[hidden]
>Subject: Re: [boost] [Boost-users] Ann: Floating Point 
>Utilities Reviewstarts today
>I have just one quibble.  I would substitute the word "mantissa" with
>"significand" in these files:
>boost/math/detail/fp_traits.hpp
>boost/math/fpclassify.hpp
>
>See my rationale here:
>http://en.wikipedia.org/wiki/Significand#Use_of_.22mantissa.22
>
>> "However, this use of mantissa is discouraged by the IEEE 
>floating-point
>> standard committee and by some professionals such as William 
>Kahan and
>> Donald Knuth, because it conflicts with the pre-existing 
>usage of mantissa
>> for the fractional part of a logarithm (see also common logarithm)."
>>
>
>Additionally, this Wikipedia page <
>http://en.wikipedia.org/wiki/IEEE_754-1985> referenced in the 
>documentation
><libs/math/doc/fpclassify.htm> never mentions "mantissa" -- only
>"significand".

I strongly support this 'quibble' - I think it is quite important to be consistent about this and 'significand' is clearly the
'trendy' work, even though many people understand mantissa.

The link above should be included in the docs to say why the significand' name is used.

The time to make this change is before this source gets into general use - that is - now!

Paul

---
Paul A Bristow
Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB
[phone] & SMS, Mobile [phone] & SMS
pbristow_at_[hidden]