Re: Re: Bug in Simplify?
- To: mathgroup at smc.vnet.net
- Subject: [mg33274] Re: [mg33216] Re: Bug in Simplify?
- From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
- Date: Wed, 13 Mar 2002 03:14:42 -0500 (EST)
- Sender: owner-wri-mathgroup at wolfram.com
This almost certainly a bug, although probably a relatively harmless
one. It's easy to imagine how this happens: at some point Simplify first
substitutes 0 for x and evaluates 0/y and at that point decides that the
answer is 0. It seems difficult to prove this conclusively, but
evaluating something like this:
In[45]:=
Simplify[x/y, {x == 1, y == 2}, ComplexityFunction ->
((Print[#1]; LeafCount[#1]) & )]
From In[45]:=
x/y
From In[45]:=
x/y
From In[45]:=
x/y
From In[45]:=
1/y
From In[45]:=
1/y
From In[45]:=
y
From In[45]:=
2
From In[45]:=
1/2
From In[45]:=
1/2
Out[45]=
1/2
suggests that the numerator is evaluated first using the assumption on x.
Andrzej Kozlowski
Toyama International University
JAPAN
http://platon.c.u-tokyo.ac.jp/andrzej/
On Tuesday, March 12, 2002, at 03:47 AM, DWCantrell at aol.com wrote:
> In a message dated 03/09/2002 17:56:36 GMT Standard Time,
> akoz at mimuw.edu.pl
> writes:
>
>> This matter has been discussed in the past a number of times so this
>> time I would just like concentrate on one aspect.
>>
>> The Mathematica evaluator always gives
>>
>> In[3]:=
>> {x/x,"dog"/"dog"}
>>
>> Out[3]=
>> {1,1}
> [snip]
>
> Many thanks for your reply.
>
> As best I can tell, however, nothing that you said explains why
> we should expect Simplify[x/y, x==0 && y==0] to yield 0. This
> still puzzles me.
>
> Regards,
> David
>
>