Deposing Dictators

Robin Becker robin at jessikat.fsnet.co.uk
Tue Aug 7 04:42:46 EDT 2001


In article <459umtohbc1q062j2sgflcc0pmond739vv at 4ax.com>, Stephen Horne
<steve at lurking.demon.co.uk> writes
....
>How many people around here have explicitly admitted they are wrong
>when they are wrong?
>
>I have on several occasions. Care to name anyone else?
....
well I was clearly wrong to have referred to the MDFL as it was taken
seriously and clouded the real issue which was the complete wrong
headedness of 238. Actions against my best interest can hardly be
regarded as benevolent though. 

As for fields etc I have the following

                         Abelian Group
                               |
                               |
                              Ring
                             /    \
                            /      \
                           /        \
                     Commutative    Division
                         Ring        Ring
                          |           |
                          |           |
                     Integral         |
                       Domain         |
                          \          /
                           \        /
                             Field

but apparently not everyone thinks an Integral Domain (xy=0 ==> x=0 or
y=0) should be commutative. Since not every integer is invertible the
integers are not a field. On the other hand probably only Tim Peters
knows what the set of IEEE 754 floating point numbers is so mathematical
arguments are clearly pointless. 

We're discussing programming languages and the arithmetic of integers
has oscillated between the floating point and discrete form in various
languages. My belief is that the current 'classic' system is pretty
optimal for programming purposes and that it is wrong to add the extra
complexity that the PEP proposes. Everywhere else in python we are
exhorted to OO thinking, why is it different for this operator? I don't
see any plans for a special divide by string operator (to make 3/'3' OK)
and clearly there is some desire for that amongst perlers.
-- 
Robin Becker



More information about the Python-list mailing list