home *** CD-ROM | disk | FTP | other *** search
/ ftp.cs.arizona.edu / ftp.cs.arizona.edu.tar / ftp.cs.arizona.edu / icon / newsgrp / group98a.txt / 000143_icon-group-sender _Tue Mar 17 12:27:55 1998.msg < prev    next >
Internet Message Format  |  2000-09-20  |  2KB
  1. Return-Path: <icon-group-sender>
  2. Received: from kingfisher.CS.Arizona.EDU (kingfisher.CS.Arizona.EDU [192.12.69.239])
  3.     by baskerville.CS.Arizona.EDU (8.8.7/8.8.7) with SMTP id MAA17272
  4.     for <icon-group-addresses@baskerville.CS.Arizona.EDU>; Tue, 17 Mar 1998 12:27:55 -0700 (MST)
  5. Received: by kingfisher.CS.Arizona.EDU (5.65v4.0/1.1.8.2/08Nov94-0446PM)
  6.     id AA21249; Tue, 17 Mar 1998 12:27:55 -0700
  7. Message-Id: <350EA1D5.26BB@gte.net>
  8. Date: Tue, 17 Mar 1998 10:16:21 -0600
  9. From: Mark Evans <evans@gte.net>
  10. Reply-To: evans@gte.net
  11. Organization: None
  12. X-Mailer: Mozilla 3.01 (Win95; I)
  13. Mime-Version: 1.0
  14. To: icon-group@optima.CS.Arizona.EDU
  15. Subject: Re: Letter Probabilities
  16. References: <9803162346.AA0130@valinet.com>
  17. Content-Type: text/plain; charset=us-ascii
  18. Content-Transfer-Encoding: 7bit
  19. Errors-To: icon-group-errors@optima.CS.Arizona.EDU
  20. Status: RO
  21. Content-Length: 607
  22.  
  23. I think that we have here one of those classic execution-time vs.
  24. memory-space tradeoffs that they teach about in computer science
  25. classes.
  26.  
  27. Mark
  28.  
  29.  
  30. Paul Abrahams wrote:
  31. > 
  32. > Using the probabilities to construct a weighted string and then
  33. > selecting a random character from the string does have one unaesthetic
  34. > property, in my book: the length of the string grows exponentially with
  35. > the precision of the probabilities.  For that reason I'd opt for a
  36. > binary search, which takes 5 probes no matter what the precision.
  37. > 
  38. > I wonder if there's another way of attacking this problem.
  39. > 
  40. > Paul Abrahams
  41.  
  42.