
1) Motivation
a) Traditional Logic: A success BUT ...
b) first order logic - minimal so necessary
claim: if 'contextualized' it is sufficient
no longer outside the first order vision
Logicians: Intuitionistic, Pi-1-1 ...
AI: Model, Temporal, direct, RDF, ...
c) Hilbert's Program
Hilbert - Reductionist
NEW - mutual recursions + model
practical - build a mind? - heads are finite
d) history - later
Building a program for recording proofs
You can't expect a program if find a proof if you can't even tell it the proof JMC
I tried this here is what I found
Formalize
Arithmetic
Induction is a schema
Set Theory
Early Axiomatizations
No finite set of axions (needs meta theaory immediately
SO Godel-Berney (?)
Needed many sorts
BUT sets were Classes SO Wang Many Sorts were NOT usable
So see early FOL sort theory
Mendleson
Group Theory
a subgroup is mot mentionable in the elementary theory of groups
how do we describe jumping between ele ns non-ele theories
Hand in math homework
Intersection of Bounded nested closed intervals is not eembty
? why is the bondedness criteria necessary?
usually answered by giving a counter example
NOT a the proof of the existence of a counterecanple
SO mentioning elements of a model are necessary
so simulation structure
Wilson' theorem
The usual proof uses the igeon hole problem
to solve ele arith problem
of course the IS an elementary proof (ever seen it?)
reasoning on the side
What are disgrams used in Geometry Proofs.
Wht
Needed many sorts
and representations
meeded context
needed formal metatheory
mukltiple contexts
e) Strategy: replace infinite notions with finite data structures