The user documentation tells us that nedistrib is dangerous. But
(%i1) p : - (a+b)/c$
(%i2) q : -((a+b)/c)$
(%i3) p-q;
(%o3) (b+a)/c+(-b-a)/c
To crunch this to zero, we need a ratsimp (or ...)
(%i4) ratsimp(%);
(%o4) 0
With negdistrib : false, the general simplifier crunchs
p - q to zero:
(%i5) negdistrib : false;
(%o5) false
(%i6) p : - (a+b)/c$
(%i7) q : -((a+b)/c)$
(%i8) p-q;
(%o8) 0
I say that the fact that (%o3) isn't 0 is a bug?
Barton