On Tue, 2 Jul 2002 04:26:17 -0700 (PDT)
Phil Carmody <[EMAIL PROTECTED]> wrote:
: --- Lars Henrik Mathiesen <[EMAIL PROTECTED]> wrote:
: > > because
: > > a b
: > > b a
: > > is satisfied by
: > > a
: > > b
: > > and
: > > b
: > > a
: > > in every model.
: > >
: > > i.e. It is explicitly _not_ the case that "no possible order of
: > nodes
: > > [...] will satisfy the input line requirements".
: >
: > Now you're being clever. It's true that if the input is a partial
: > order, you can conclude that a=b; but why do you output the element
: > twice, then?
:
: There's a difference between an element and its value.
: Consider { 1, -1, i, -i } with partial orderings real components.
: There are 4 elements, but only three values being compared.
Perhaps is a matter of interpretation but I don't see
a b
b a
as a pair of axioms and we looking for models, I guess different labels
mean different elements, so
a b
b a
implies a != b, is not a partial order over the set {a, b}, and as I
understand it valid output says so.
-- fxn
