home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!think.com!spool.mu.edu!yale.edu!ira.uka.de!smurf.sub.org!news
- From: urlichs@smurf.sub.org (Matthias Urlichs)
- Newsgroups: sci.crypt
- Subject: Re: random number generation
- Date: 24 Jan 1993 23:03:49 +0100
- Organization: University of Karlsruhe, FRG
- Lines: 20
- Message-ID: <1jv3o5$jf0@smurf.sub.org>
- References: <18404@autodesk.COM> <Lw7LXB1w164w@jwt.UUCP>
- NNTP-Posting-Host: 127.0.0.1
-
- In sci.crypt, article <Lw7LXB1w164w@jwt.UUCP>,
- bbs-hrick@jwt.UUCP writes:
- > > > I am looking for good ideas for real random number generation. I've heard
- >
- > How would this work... take the least significant bit from each
- > byte of traffic in a Usenet newsgroup?
-
- I'd assume that English text has some redundancy in the lowest bits when
- ASCIIized. Thus, not 100% random given a suitable definition of that word.
-
- You could grab each article and create its MD5 signature. You'd probably
- have to be somewhat imaginative so that knowledge of the sequence of
- articles arriving at your machine won't reveal the random number.
-
- --
- Life's too short, but if you go fast enough, you can live
- it one-and-a-half times.
- --
- Matthias Urlichs -- urlichs@smurf.sub.org -- urlichs@smurf.ira.uka.de /(o\
- Humboldtstrasse 7 -- 7500 Karlsruhe 1 -- Germany -- +49-721-9612521 \o)/
-