Scott will discuss the semantics of the Nuprl type theory. In Nuprl, the semantics have driven system design. In other systems the proof theory seems to be primary.
Scott will compare Nuprl with Feferman's theories with this distinction in mind. Scott may also touch on the notions of equality in types and possibly recursive types.