home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky sci.physics:14722 sci.math:11518
- Newsgroups: sci.physics,sci.math
- Path: sparky!uunet!gatech!news.byu.edu!ux1!mica.inel.gov!guinness!opal.idbsu.edu!holmes
- From: holmes@opal.idbsu.edu (Randall Holmes)
- Subject: Re: Computability of the universe
- Message-ID: <1992Sep15.230401.16386@guinness.idbsu.edu>
- Sender: usenet@guinness.idbsu.edu (Usenet News mail)
- Nntp-Posting-Host: opal
- Organization: Boise State University Math Dept.
- References: <1992Sep11.181552.416@prim> <1992Sep11.181736.5324@galois.mit.edu> <BuHM6A.4Gz@news.cso.uiuc.edu>
- Date: Tue, 15 Sep 1992 23:04:01 GMT
- Lines: 62
-
- In article <BuHM6A.4Gz@news.cso.uiuc.edu> chappell@symcom.math.uiuc.edu (Glenn Chappell) writes:
- >In article <1992Sep11.181736.5324@galois.mit.edu> jbaez@riesz.mit.edu (John C. Baez) writes:
- >>In my opinion this fairly pervasive notion that you can only
- >>"really" measure lengths that are rational numbers is an outdated
- >>remnant of one of the Greeks' less bright moments. For example:
- >>
- >>Say we all measured distances with a circular roller that was one foot
- >>in diameter. (I think they do something like this sometimes.) We
- >>measure distances by rolling this thing along and counting the number
- >>of revolutions. Okay, now we are measuring things in units of pi feet.
- >>If our driveway looks to be 40 revolutions long, our best guess is
- >>that it's 40pi feet long.
- >.
- >.
- >.
- >>Rationals are nice in many ways, but the idea
- >>that measurements of lengths always give rational numbers is just
- >>plain silly.
- >
- >Strictly speaking, yes, but the idea *is* useful in practice. After
- >all, in real life, no length measurement is made without some
- >experimental error. The result of a length measurement is actually a
- >probability distribution - or, if you like, an interval. Since plain
- >ol' numbers are easier to deal with than intervals (and since, in
- >practice, they're usually good enough) any dense set of real numbers
- >will do for expressing length measurements. The rationals are dense
- >in the reals, and they're also generally easy to deal with, so why
- >not use them?
- >
- >Well, interestingly enough, we usually don't. We usually use proper
- >subsets of the rationals.
- >
- >E.g. traditionally, when the English system is used, measurements are
- >expressed using rationals that can be written with their denominator
- >being a power of 2.
- >
- >Thus, English system rulers are marked in inches, 1/2's, 1/4's, 1/8's,
- >maybe 1/16's and sometimes 1/32's. The same goes for liquid measurements
- >(1/2 gallon), etc.
- >
- >Now, the set of rationals that can be written as a/2^b is still dense in
- >the reals, so it's still good enough.
- >
- >More modern practice tends toward using only those numbers than can be
- >expressed as a/10^b. (When was the last time you heard anyone say
- >"1/8 meter"?) The set of such numbers is also dense in the reals, so
- >it's also good enough.
-
- No, no, use the dyadic rationals (a/2**b). Their special status is
- revealed in Conway's construction of the surreal numbers. :-)
-
-
- >
- > Glenn Chappell <><
- >
-
-
- --
- The opinions expressed | --Sincerely,
- above are not the "official" | M. Randall Holmes
- opinions of any person | Math. Dept., Boise State Univ.
- or institution. | holmes@opal.idbsu.edu
-