home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.skeptic
- Path: sparky!uunet!stanford.edu!CSD-NewsHost.Stanford.EDU!Xenon.Stanford.EDU!amorgan
- From: amorgan@Xenon.Stanford.EDU (Crunchy Frog)
- Subject: Re: skeptic-faq V 0.1
- Message-ID: <1992Jul29.232747.23650@CSD-NewsHost.Stanford.EDU>
- Sender: news@CSD-NewsHost.Stanford.EDU
- Organization: Computer Science Department, Stanford University.
- References: <1863@snap> <1992Jul29.204535.20322@CSD-NewsHost.Stanford.EDU> <1992Jul29.214827.13287@ringer.cs.utsa.edu>
- Date: Wed, 29 Jul 1992 23:27:47 GMT
- Lines: 42
-
- In article <1992Jul29.214827.13287@ringer.cs.utsa.edu>
- djimenez@ringer.cs.utsa.edu (Daniel Jimenez) writes:
- >In article <1992Jul29.204535.20322@CSD-NewsHost.Stanford.EDU>
- > amorgan@Xenon.Stanford.EDU (Crunchy Frog) writes:
- >>In article <1863@snap> paj@uk.co.gec-mrc (Paul Johnson) writes:
- >>>[about how we could test an alien, e.g. "Prove FLT"]
- >>Fermat's Last Theorem has the risk that the AB (advanced being) will 'answer'
- >>it by saying "Well, just integrate over the semi-hyperbolictransrealpartially-
- >>dynamic numbers and it is obvious". A better solution, which I saw here
- >>(but I don't remember who suggested it) is to ask for the factorization of
- >>some *huge* number that is known to be composite but whose factors are
- >>not known. Computing the factors of, say a 10000 digit number, is way beyond
- >>us right now but for an alien dude it should be cake. Checking
- >>the answer is, of course, trivial.
- >
- >What if the aliens say "Factoring is intractable." Humans don't know if
- >there is any less-than-subexponential solution to factoring. Factoring
- >could even be NP-complete (unlikely, though). It might turn out that
- >all the computing power in the universe may not be able to factor a 10000
- >digit number in a reasonable amount of time.
-
- Okay, so don't make it 10000 digits (that was just and example fer cryin'
- out loud). Pick a number that would take us, uh, 10 years to factor
- using our computers. If they are really advanced then they should
- be able to crank out an answer in under a week (that would only require
- a computer 500 times faster than what we use).
-
- Anyway, if we are dealing with people who aren't too sophisticated
- mathematically then you should be able to trick them:
-
- ME: Is there an x,y,z,w that fit FLT?
- NewAger: No. Proof by Hughlkjasdjfl's 3rd Lemma
- ME: (nuts) Is there a transfinite between aleph 0 and aleph 1?
- NA: No. Proof by Hughlkjasdjfl's 4th Lemma
- ME: (damn) Is P a proper subset of NP?
- NA: No. Proof by Hughlk...
- ME: Don't bother, please reduce the Hamiltonian problem to polynomial
- time (VICTORY!)
-
- >Daniel Jimenez
-
- C Frog
-