🚀 go-pugleaf

In article <84...@ttidca.TTI.COM> holl...@ttidcb.tti.com (The Polymath) writes:
>In article <60...@wpi.wpi.edu> jwhi...@wpi.wpi.edu (John C Whitson KB2GNC) writes:
>}Basically, imagine the United States Defense computer
>}systems. Then imagine a whole slew of targets appear on
>}long distance monitors, seeming like a Soviet Attack!!!
>}The blips however were not confirmed by other means.
>}Then the blips went away.
>}Just think .... National Defense alertness based on a 20-cent
>}chip in a minicomputer ... makes one wonder a little.

There was also a case where there was a hardware problem that casued a
realistic looking exercise being run on one machine to leak into the
operational systems.  That may have been the same incident you are
talking about.  See other comments below.

>,,.  I remember one incident described how NATO
>went on full red alert when their radar systems detected what looked like
>a full-bore Soviet weapons launch coming over the horizon.  We came within
>20 minutes of a massive nuclear exchange because our system had spotted
>_the rising full moon_.

Nonsense.

It had nothing to do with NATO.  As I recall it was NORAD in Cheyenne
mountain.  Indeed, a very powerful radar system managed to see
reflections from the moon (being "full" would have been of no
significance).  Although the delay to the moon and back was many times
the systems normal interpulse interval, it interpreted these as sporadic
but valid echos.  Word of the whole thing got out very quickly as there
was a tour group in the control room, when the problem started, who were
quickly hussled out.

A display of the number of incoming missles just started counting up and
eventually got to over 1500.  However, there were a number of suspicious
things about this.  For some reason, the number of incoming missles
counted up by twos and a map on which elliptical estimated impact areas
(shrinking as the missles got closer and estimates more accurate) are
supposed to be displayed never displayed any.  The random sporadic echos
were not coherent enough missle tracks for the system to project impact.

I have no idea what you think "full read alert" is.  As I recall this
event caused a conference between the various national alert
(indications and warnings) centers that quickly decided it was not a
real attack.  These centers are hooked together by communications
circuits to make such conferences easier.  There are usually more than a
dozen such conferences a month, for one reason or another.

If all this sounds a bit flakey, consider which you would prefer:

(1) a single monolithic national defense computer systems that "never"
made mistakes so that, when it said so, we would lauch a massive
nulear retaliatory strike with little or no human evaluation of
the system.

(2) a hodge-podge of various incompatible and very sensitive national
defense computer systems that produce a number of false alarms that are
then evaluated in the overall context by human judgement.
-- 
	+1 617-969-9570		Donald E. Eastlake, III
	ARPA: d...@XAIT.Xerox.COM    usenet:  {cbosg,decvax,linus}!cca!dee
	AppleLink:  D2002	Box N, MIT Branch PO, Cambridge, MA 02139 USA