gordon@warwick.UUCP (Gordon Joly) (04/23/86)
In article <1800@mtgzz.UUCP> Mark leeper says:- > I talked to Clarke about 2001 in 1969 and he brought up the HAL/IBM > question himself. He said that it was just a surprising coincidence. With odds of 1/8788 against, maybe it was more a case of subconscious reasoning, as in Kekule's discovery of the structure of the benzene molecule, and in the deciphering of Samuel Pepys' diaries. And the odds must also take into account strings like S*X or BCA, musn't they? Gordon Joly {seismo,decvax,ucbvax}!mcvax!ukc!warwick!gordon School of Mathematics, Queen Mary College, London. P.S. Why are the odds (26**3)/2 anyway?
vis@mit-trillian.MIT.EDU (Tom Courtney) (04/26/86)
In article <276@euclid.warwick.UUCP> gordon@euclid.UUCP (Gordon Joly) writes: >In article <1800@mtgzz.UUCP> Mark leeper says:- > >> I talked to Clarke about 2001 in 1969 and he brought up the HAL/IBM >> question himself. He said that it was just a surprising coincidence. > >With odds of 1/8788 against, maybe it was more a case of subconscious >reasoning, as in Kekule's discovery of the structure of the benzene >molecule, and in the deciphering of Samuel Pepys' diaries. And the odds >must also take into account strings like S*X or BCA, musn't they? > The odds probably weren't so bad. Suppose Clarke was looking for a three letter acronym that was actually a name? Then he's restricted to a much smaller set of possibilities (HAL, SAL, SAM, ART, etc... [not ETC]). Furthermore, after the fact, lots of things look like the odds were too horrendous to be coincidence.
flaps@utcs.uucp (Alan J Rosenthal) (04/28/86)
The odds are most certainly considerably less than one in 26**3/2. The reasoning for 26**3/2 is that there are 26**3 possible three letter combinations, out of which we count 'HAL' and 'JCN' both as hits, as certainly it doesn't matter for impact whether the letters immediately precede or follow 'IBM'. But likewise, wouldn't a hit on 'DEC' or 'VAX' or 'CBM' suffice? And also, any of a number of sf stories could make a hit. In other words, if you had 26**3/2 sf stories, you would expect a hit, and the existing number of sf stories therefore reduces the surprise value of this hit. note: many of the keywords in this article are trademarked company names or product lines. disclaimer: the word "impact" is an unintentional pun.
crowl@rochester.ARPA (Lawrence Crowl) (05/01/86)
In article <1210@utcs.uucp> flaps@utcs.UUCP (Alan J Rosenthal) writes: >But likewise, wouldn't a hit on 'DEC' or 'VAX' or 'CBM' suffice? At the time the movie came out, how many computer companies had general public name recognition? DEC did not have name recognition. VAX did not exist. CBM did not have name recognition if it existed. Probably the only other "name" the public would have recognized an associated with computers is UNIVAC. -- Lawrence Crowl 716-275-5766 University of Rochester Computer Science Department ...!{allegra,decvax,seismo}!rochester!crowl Rochester, New York, 14627
vis@mit-trillian.MIT.EDU (Tom Courtney) (05/02/86)
In article <17754@rochester.ARPA> crowl@rochester.UUCP (Lawrence Crowl) writes: >At the time the movie came out, how many computer companies had general >public name recognition? DEC did not have name recognition. VAX did not >exist. CBM did not have name recognition if it existed. Probably the >only other "name" the public would have recognized an associated with >computers is UNIVAC. I Think CDC might have managed it, because of the big lawsuit (which they won) with IBM.
drew@ukma.UUCP (Andrew Lawson) (05/03/86)
In article <339@mit-trillian.MIT.EDU> vis@trillian.UUCP (Tom Courtney) writes: >possibilities (HAL, SAL, SAM, ART, etc... [not ETC]). Furthermore, after the fact, >lots of things look like the odds were too horrendous to be coincidence. Yes, remember that the odds of a given deck of cards being in exactly the order that they are is 1 in 52! (:-) Don't abuse probability. -- Drew Lawson cbosgd!ukma!drew "Parts is parts." drew@uky.csnet drew@UKMA.BITNET