home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pageworks.com!world!eff!sol.ctr.columbia.edu!spool.mu.edu!agate!stanford.edu!rock!concert!duke!news.duke.edu!neuro.duke.edu!jfw
- From: jfw@neuro.duke.edu (John F. Whitehead)
- Newsgroups: comp.sys.mac.hardware
- Subject: Re: LaserWriter sucks up Cat
- Message-ID: <7944@news.duke.edu>
- Date: 14 Dec 92 17:51:02 GMT
- References: <1992Dec10.095921.29120@lin.foa.se> <1992Dec12.065032.4870@news.acns.nwu.edu> <1gd0piINNdis@MINERVA.CIS.YALE.EDU>
- Sender: news@news.duke.edu
- Organization: Duke University; Durham, N.C.
- Lines: 34
- Nntp-Posting-Host: neuro.neuro.duke.edu
-
- In article <1gd0piINNdis@MINERVA.CIS.YALE.EDU> bell-peter@yale.edu (Peter Bell) writes:
- >In article <1992Dec12.065032.4870@news.acns.nwu.edu> eepmatt@casbah.acns.nwu.edu (Matthew Friedman) writes:
- >>Wait a minute. This guy is having problems writing bug free code that works as
- >>intended the first time? And he works for the GUIDED WEAPONS division!?!?!
- >
- >Oh, don't worry. The stuff is beta tested. Anyone read the article in Sciece
- >about the guidance computers for the Patriots we lent Israel and used in Saudi
- >Arabia? (would have been out a year or so ago)...
- >... The problem was, there was some bug in the clock the
- >clock they were using, such that they had a tendency to not recognize incoming
- >ordnance 'cuz it was in the wrong place on a subsequent check, according to
- >their clock.... the solution? Reboot the computers about once every eight
- >hours while the folks at the pentagon wrote the patch....
-
- It wasn't the programmers' fault, though. They designed a program meant
- to be used on a Patriot that was not on for long periods of time. The
- clock, due to round-off errors, would cause the computer to think that
- the missile was a couple of millimeters off from where it really was.
- No big deal for short periods, but after running the computer for eight
- hours the cumulative error was off by enough to mean that the Patriot
- would miss the incoming missile by > ~10 meters -- enough that it would
- be ineffective. It was an error not of programming but of the
- anti-missile system being used outside its specifications, and of old
- technology that meant that the clock error became significant after a
- relatively short period of time.
-
- If anyone wants to continue this thread, let's move it to
- comp.sys.mac.missles.
-
- --
- ________________________________________________________________________
- | John jfw@neuro.duke.edu Duke University Medical Center |
- | Whitehead jfw@well.sf.ca.us Department of Neurobiology |
- |________________________________________ Durham, North Carolina ________|
-