MathGroup Archive 2007

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: SolveAlways documentation problem

  • To: mathgroup at smc.vnet.net
  • Subject: [mg78773] Re: [mg78745] SolveAlways documentation problem
  • From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
  • Date: Mon, 9 Jul 2007 01:36:51 -0400 (EDT)
  • References: <200707081013.GAA09510@smc.vnet.net>

On 8 Jul 2007, at 19:13, Chris Chiasson wrote:

> "SolveAlways[eqns,vars] is equivalent to Solve[!Eliminate[! 
> eqns,vars]]."
>
> does not work with the first basic example given
>
> In[1]:= SolveAlways[a x+b==0,x]
> Out[1]= {{a->0,b->0}}
>
> In[2]:= Solve[!Eliminate[!(a x+b==0),x],x]
> Out[2]= {{}}
>
> -- 
> http://chris.chiasson.name/
>


The documentation is right:

  Solve[! Eliminate[! (a x + b == 0), x]]
  {{a -> 0, b -> 0}}

If you include x in Solve, than the answer will always be {{}} since  
the equations that you get after Eliminate finishes its job do not  
involve x (actually one could argue over whether the answer should be  
{{}}, which means the equations are valid for all x, or {}, which  
means there are no generic solutions, but this is not the issue here).
If you use LogicalExpand instead of Solve you will get:

LogicalExpand[ ! Eliminate[ ! a*x + b == 0, x]]
a == 0 && b == 0

which is the kind of answer that Reduce returns.

Andrzej Kozlowski

  


  • Prev by Date: Re: Is there anything wrong?
  • Next by Date: Re: Bug in packages with RegularExpression[]
  • Previous by thread: SolveAlways documentation problem
  • Next by thread: Re: SolveAlways documentation problem