From: Paul Prescod <paulp@...>
Date: 2001-07-31T01:25:45+09:00
Subject: [ruby-talk:18824] Numeric model

Yukihiro Matsumoto wrote:
> 
> Hi,
> 
> In message "[ruby-talk:18781] Re: Ruby as opposed to Python?"
>     on 01/07/30, Paul Prescod <paulp@ActiveState.com> writes:
> 
> |In my opinion, either integers should be substitutable for floats or
> |should not be. In Python and Ruby they are "almost", "mostly", "usually"
> |substitutable.
> 
> I don't understand what you mean by "fully substitutable".  You mean
> seq[1.0] will become legal in Python 3.0?  How about seq[1.2]?

I'm talking about *integer objects* being substutable for *float
objects* in the same OO sense that cows are supposed to be substitutable
for mammals. Your algorithm should not radically change its behaviour
when you pass in an integer versus a float.

As an aside: you see "1.0" as a float but in my perfect language it
would be a rational, equivalent to "4/4" or "1". Also in my imagination
they would all be valid (and equivalent!) sequence indexes.
-- 
Take a recipe. Leave a recipe.  
Python Cookbook!  http://www.ActiveState.com/pythoncookbook