...Property

This work was sponsored by Office of Naval Research grant nr. N 00014-95-1-0660.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...260

Step I in [134] can be viewed as superfluous for the theoretical development. If step I is not used the goal is to remove all edges, if step I is used the goal is to produce an empty graph. Nothing else changes. Removal of vertices should however speed up the actual computations.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...ordinal.

In this section we thus talk about algorithms/reductions with potentially infinitely many steps. For finite sets the reduction procedure is an algorithm in the sense of Definition 2.1.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...points/cliques/simplexes:

The following short argument was shown to the author by J.D. Farley.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...out.

Thus clearly we are facing infinite ``algorithms" once more.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.