Patrick Cahalan wrote:
> On the other hand, I cannot countenance 1 guilty murderer going free to
> protect 1 innocent parking scofflaw.
>
If they are actually innocent, then they are not scofflaws, eh?
One big problem with convicting the innocent is that it is unjust.
A worse problem is that it causes people to perceive the law as unjust,
and then they stop caring about it, and commit much worse crimes.
So if we allow the law to convict the innocent, we cause more lawlessness.
OTOH, if we allow the law to let the guilty go free, we also cause
lawlessness.
I expect this is a dynamic optimization problem :) but I've said that
before:
"Timing the Application of Security Patches for Optimal Uptime".
Steve Beattie, Seth Arnold, Crispin Cowan, Perry Wagle, Chris
Wright, and Adam Shostack. Presented at the USENIX 16^th Systems
Administration Conference (LISA 2002)
<http://www.usenix.org/events/lisa02>, Philadelphia, PA, December
2002. Postscript
<http://crispincowan.com/%7Ecrispin/time-to-patch-usenix-lisa02.ps.gz>.
or ugly PDF
<http://crispincowan.com/%7Ecrispin/time-to-patch-usenix-lisa02.pdf>.
You get the same problem in intrusion detection systems, where if they
are not sensitive enough, you miss intrusions, but if they are too
sensitive, users ignore them.
This problem occurs over and over when ever you have flaky sensors or
imperfect knowledge.
Crispin
--
Crispin Cowan, Ph.D. http://crispincowan.com/~crispin/
Itanium. Vista. GPLv3. Complexity at work
_______________________________________________
FDE mailing list
[email protected]
http://www.xml-dev.com/mailman/listinfo/fde
