Provers: Status and ProspectsCombining interactive and
automatic theorem provers is anobvious idea, but it contains
surprising complexities. Theinteractive user can derive little
benefit from merelyquerying an automatic prover: to be useful, the
integrationmust be deep, which complicates the implementation.
Thespeaker will describe the project, ongoing at Cambridge, tointegrate
Isabelle with saturation- based theorem provers,primarily SPASS. He
will outline the potential benefits(including the less obvious
ones). He will also describe theproblems that have been encountered,
including irrelevancefiltering, higher-order reasoning and
presenting lengthyproofs to users. Some of these issues are avenues for
futureresearch.