My algo has a long way to go. It really has to be tailored to accomodate all stray cases, dividing by a reciprocal is certainly one of them. In any case, I thought it was a fairly structured way of solving, if not all, at least say some 40% of the cases.
In any case, if you want a complete algorithm that can do this, then are we not bordering on the NP complete satisfiability problem. We are actually making matters worse as variables are not just binary valued, but could take any value. So in my opinion, a complete algorithm that can compare any 2 given equations is not possible in polynomial time.
-Vijay
On 12/15/05, pramod <[EMAIL PROTECTED]> wrote:
Vijay,
Your algo is not clear to me.
Let's take the original example given by moop.
(a+b)*(c-d) whose lexicographic sorting will give (a+b)*(c-d)
and second _expression_ is (c-d)/(1/(b+a)) whose lexicographic sorting
will give
(c-d)/(1/(a+b))
These two are not the same, so what you propose to do?
