The only problem is that most of responsible people are computer scientists and sometimes even only with a BS in CS and therefore have no clue of harmonic analysis and advanced probability theory.
If you project your network system in the C^n- space of markovian probability measures and with to the frequency domain, you can easily see that our system represents a compact manifold of superharmonic measures. And malign overflow is just a upper bound in this set, therefore harmonic. It's well known that the only harmonic functions on compact manifolds are constant. So going back into the time domain this means that you must just analyze the frequency of the packets. All packet streams with a constant frequency are malicious by the above calculation and therefore should be dropped. Of course there are some minor points with the frequency reflection on edges etc. but this is very basic stuff and can be easily solved.
If think there was a paper of Lorgajev and Starniktov in the 80ies about this, but I'm not really sure.
I agree. However I also suggest that
packets streaming into any port under a gaussian bell
curve probability and/or a poisson distribution
also be filtered out. I heard that the newest
version of the linux kernel has mechanisms for
thermodynamically analyzing all packets
for signs of randomness. As all computer scientists
and mathematicians know, humans are not random
and it is therefore unlikely that packets sent
from a client will arrive at any given server
randomly. Richard Stallman in his PhD
thesis ``The Statistical Thermodynamics of
Software Evolution'' says as much. Please read
the paper for details.
Sorry, I don't have the URL. I'm not
a karma whore.
Seriously, out of the enormously huge set of Everything that Can Be Known ("we are all really impressed down here, I can tell you"), we puny humans walk around with itty-bitty subsets of it of Things We Know. But the Things We Know is the easy part.
The hard part is learning to recognize our own ignorance, especially at the boundary between knowledge and ignorance. I think a good measure of general intelligence is the quality of one's bullshit detector. And, really, we all have our own version of that fuzzy area near the boundary where the mixture of knowledge and ignorance means that we are dangerously ignorant. You know, he who is without sin, blah blah blah.
Such floods can be easily stopped. (Score:2, Funny)
If you project your network system in the C^n- space of markovian probability measures and with to the frequency domain, you can easily see that our system represents a compact manifold of superharmonic measures. And malign overflow is just a upper bound in this set, therefore harmonic. It's well known that the only harmonic functions on compact manifolds are constant. So going back into the time domain this means that you must just analyze the frequency of the packets. All packet streams with a constant frequency are malicious by the above calculation and therefore should be dropped. Of course there are some minor points with the frequency reflection on edges etc. but this is very basic stuff and can be easily solved.
If think there was a paper of Lorgajev and Starniktov in the 80ies about this, but I'm not really sure.
Re:Such floods can be easily stopped. (Score:3, Insightful)
I agree. However I also suggest that packets streaming into any port under a gaussian bell curve probability and/or a poisson distribution also be filtered out. I heard that the newest version of the linux kernel has mechanisms for thermodynamically analyzing all packets for signs of randomness. As all computer scientists and mathematicians know, humans are not random and it is therefore unlikely that packets sent from a client will arrive at any given server randomly. Richard Stallman in his PhD thesis ``The Statistical Thermodynamics of Software Evolution'' says as much. Please read the paper for details.
Sorry, I don't have the URL. I'm not a karma whore.
Such moderations can be easily stopped. (Score:1)
Who moderated this +1, Insightful?
Definitely +1, Funny... but insightful? Did the moderator even understand the words yeOldSkeptic was using?
Re:Such moderations can be easily stopped. (Score:2)
The hard part is learning to recognize our own ignorance, especially at the boundary between knowledge and ignorance. I think a good measure of general intelligence is the quality of one's bullshit detector. And, really, we all have our own version of that fuzzy area near the boundary where the mixture of knowledge and ignorance means that we are dangerously ignorant. You know, he who is without sin, blah blah blah.
Re:Such moderations can be easily stopped. (Score:2)