ineq: sum of identical inequalities is not inequality



On Sun, Jan 04, 2004 at 12:13:48PM -0800, Richard Fateman wrote:
> I don't know what you are trying to get maxima to do,

Actually, I try to generate consequences from a set of inequalities.
Of course, the full set of consequences is infinite, 
even if the initial set consists of 
a single inequality (for x>0,  x>y => 2x>y, 3x>y, 4x>y,..., or even better
for any x, x>y => 2x>2y, 3x>3y,..., as maxima does not recognize equivalence
like x>y <=> 2x>2y). I expect that for my task the consequences 
obtained as sums of subsets of inequalities from the initial set would be 
enough. Of course, in the initial set, the OP is always ">", not "<".

> but I suggest you separate out the right and left sides
> of the inequalities before you do arithmetic.  

Yes. If you mean usage of the PART and OP functions, it is possible.
But the ineq package appeard to be very suitable for adding simple 
inequalities. So, I hope to use it with not adding identical ineqs,
where it fails. It seems that nset package will be very helpful for this.

Many thanks,
--
Alexander