home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!pagesat!spssig.spss.com!hosken
- From: hosken@spss.com (Bill Hosken)
- Newsgroups: comp.lang.c++
- Subject: Re: self-reproducing C++ program
- Message-ID: <1992Nov19.201547.10241@spss.com>
- Date: 19 Nov 92 20:15:47 GMT
- References: <1919@alcbel.be> <DECHC00.92Nov17210448@tohi.DMI.USherb.Ca> <c164-aa.722076580@po.berkeley.edu> <1992Nov18.113006.12935@jyu.fi> <mg.722138417@tyrolia>
- Sender: news@spss.com (Net News Admin)
- Organization: SPSS, Inc.
- Lines: 19
-
- In article <mg.722138417@tyrolia>, mg@tyrolia (Michael Golan) writes:
- > sakkinen@jyu.fi (Markku Sakkinen) writes:
- >
- > >There is one possible problem, though.
- > >When you start generating all possible legal C++ programs
- > >in the order of ascending length, before the first self-reproducing
- > >one you might run into an item about which you can prove _neither_
- > >of the following alternatives: [...stuff deleted]
- >
- > C++ only runs on finite state machines (computers). So the halting question:-)
- > has nothing to do with it. All C++ programs terminate or go back to the
- > same state.
- >
-
- The language, though, is independent of the machine on which it runs.
- One can imagine asking for more and more floppies to be inserted, simulating
- a Turing machine tape, and then the finiteness would be debatable. Turing
- machines have only a finite number of 'processor' states, but have an
- unbounded tape.
-