COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	tracer[s85,jmc]		Information tracers
C00006 ENDMK
C⊗;
tracer[s85,jmc]		Information tracers

	Physicists, chemists and earth scientists use tracers all
the time.  A small amount of a ``tracer'' substance detectable in very small
quantities is injected into a system, and the flows within
the system are monitored by measuring the amount of the tracer substance in
different places.

	I have inadvertently used information tracers by introducing
unconventional terminology.  ``garbage collector'', ``alpha-beta
heuristic'', ``conditional expression'', ``circumscription'' are
all my terms, and when I see them in an article I know the ideas
are to some extent descendants of mine.  This doesn't say someone
else didn't do it earlier or better.  I introduced the
phrase ``time-sharing'' for a system of collective computer usage,
but since this term was in use previously in communication
engineering it is less certain that any particular time-shared
operating system owes anything to my work.

	I remember reading an article in Pravda entitled ``pervaya
myaghkaya posadka na lunye'' which translates literally to ``first soft
landing on the moon'' about a Soviet soft landing.  Since ``soft landing''
was in use long previously as an American term, I concluded that either
the Soviet scientists or the Soviet journalists got the idea from the
American literature.

	Information tracers can also be used to find hidden connections
among people.  Inject a bit of useful confidential information at one place
in a system coupled with an interesting and exotic falsehood and watch
for the falsehood to come out elsewhere.  Gossip about a person is
the most plausible tracer, but the best kind would be something that
people won't talk about (perhaps because of irrelevance to the
usual subjects of conversation) but which will be mentioned if
the right topic comes up.  This might work best if the injected
confidential information and the tracer were combined in a document.
Of course, the tracer should not be something people would be inclined
to keep confidential.

	Perhaps this is a standard technique.  In World War II, we
had broken the Japanese naval code except for place names, so we
had Midway send a message that it was short of water, and soon
there was a Japanese message reporting that A20 was short of
water confirming the conjecture that A20 referred to Midway.
This allowed more confident planning of the battle of Midway.