Home Original page

[Numpy-discussion] Re: numpy, overflow, inf, ieee, and rich comparison

Konrad Hinsen hinsen at cnrs-orleans.fr
Wed Oct 25 04:56:00 EDT 2000 
Jon Ribbens <jon+python-list at unequivocal.co.uk> writes:

> My opinion is that floating point is Dangerous and To Be Avoided
> unless you know what you're doing. The language should not create

I don't disagree, but defining / as integer division doesn't solve the
problem. If integer division is not what is called for in a certain
algorithm, its presence is a bug and will sooner or later be detected
and fixed (by substituting float division).

What we are discussing here is not whether you should use floating
point, but what the most reasonable interpretation for a certain
syntax is.
-- 
-------------------------------------------------------------------------------
Konrad Hinsen                            | E-Mail: hinsen at cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24
Rue Charles Sadron                       | Fax:  +33-2.38.63.15.17
45071 Orleans Cedex 2                    | Deutsch/Esperanto/English/
France                                   | Nederlands/Francais
-------------------------------------------------------------------------------
More information about the Python-list mailing list