home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.theory:2510 comp.lang.functional:1390
- Newsgroups: comp.theory,comp.lang.functional
- Path: sparky!uunet!usc!zaphod.mps.ohio-state.edu!darwin.sura.net!Sirius.dfn.de!hpux.rz.uni-jena.de!mwj
- From: mwj@rz.uni-jena.de (Johannes Waldmann)
- Subject: complexity measures for the lambda calculus?
- Message-ID: <1992Nov23.082459.24583@rz.uni-jena.de>
- Organization: Friedrich-Schiller-University Jena, Germany
- Date: Mon, 23 Nov 1992 08:24:59 GMT
- Lines: 12
-
- How can I measure the complexity of computations in the lambda
- calculus? Of course I could chose an evaluation strategy,
- simulate everything on a Turing machine and apply housekeeping
- there. But that's not what I want. I thought of counting the
- minimal number of beta reductions (that a given expression needs
- to reach normal form). One could allow parallel reductions (of redexes
- not contained in each other)... Has there been work on this field,
- is it considered useful? Any hints, pointers to the literature
- etc would be appreciated.
-
- I will summarize if I get enough email responses. --- Johannes.
-
-