COMMENT ⊗   VALID 00006 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	This  seminar will  cover many  of  the more  recent  results in  the
C00005 00003	I expect  people to  work in this  seminar. I  will be available  for
C00008 00004	randomness:
C00009 00005	bibliography
C00010 00006	get tarski fix-point paper
C00011 ENDMK
C⊗;
This  seminar will  cover many  of  the more  recent  results in  the
mathematical   theory  of  computation.  The   emphasis  will  be  on
establishing  motivation  and  intuition  rather  than  dwelling   on
technical  results. It  is the  philosophical outlook  which must  be
cultivated;  it is the intuition which must  be build. Once these are
established  carefully,  the   technical  details  usually  hold  few
suprises.  That is one of the satisfying aspects of the field: a rich
and beautiful discipline,  close to the  mathematical surface,  whose
study has  immediate applications to  practical problems  in software
design,   and  whose   mathematical   properties  really   are  quite
clean.

This is another aspect of the strange field called  Computer Science:
Those of  us who enjoy  pure mathematics, but become  dispondent when
the  word "relevance"  is mentioned,  can feel justification  for our
existence. Those of us who enjoy programming, but  explode when faced
with  the  incredible  muddle  of  programming  languages,  can  find
solitude in searching for "a  better way".  If Bertrand  Russell will
pardon me: "one's view of Computer Science should be prompted by love
and guided by intelligence".  I hope to cultivate a little of both in
this seminar. 

I expect  people to  work in this  seminar. I  will be available  for
questions  and insults at either  497-4971 or 941-6510. I  will be on
campus at least twice  a week. If I don't  know the answers I'll  get
them; if you can't find the necessary papers I'll try to find them. 

I propose the following schedule.  An initial session of interminable
duration  to set  the tone;  basically philosophical  (basically bull
shit, but  you knew  that anyway.) I  will outline  several areas  of
current  research  interest  let  you  choose  which  you'd  like  to
concentrate on.  Then the work starts. 

                           General outline

The schools of semantics
 axiomatics
 operational
 denotational

Why bother?
 what is gained

problems of correctness and equiv.
 correctness of interpreters and compilers
 equivalence of functions and algorithms and programs.

general overview

little about logic and model theory

λ-calculus
 syntax
 pragmatics
  machines and implementations
  call-by-name, call-by-value
   lisp's eval and SECD machine
 semantics

self-application
label and fixpointing

extensions to "real" languages.

abstract data structures

clean it up: what have we got.

letting the dust settle:
Darlington
Boyer-Moore
VCG(?)

comparison of operational and denotational


off to the races again


Vienna Def Lang: 

construction of models

continuations
rconcillation of operational and denotational: ligler

interspersed with great amounts of philosphical bull.

how to build systems
lcf
twity
vcg
prover
boyer moore
darlington
sammet(?)

meta bullshit: ACTORS

conjectures: forcing and intuitionism

randomness:
do godel argument in lisp
bibliography
scott 
strachey
ligler
hoare
mccarthy
milner
wadsworth
gordon
von henke
landin
naur
dijkstra
rogers
morris, jim & lockwood
wegner
backus
milne
tennent
de roever
sammet
london aim
newey
luckham, park and paterson
plotkin
reynolds
get tarski fix-point paper;
when do you know there is a solution to domain equation?
simply because you can write it doesn't mean crap.


sexp = atom|(sexp . sexp)

sεS ↔ sεA ∨ s=(s1 . s2) and s1,s2εS

S = A∪S ⊗ S

try sεS ↔ ¬sεS, just as easy to write , but difficult to satisfy.