On Tue, Mar 04, 2003 at 02:52:06PM -0500, blitz wrote:
The only disadvantage I see, is a single point of failure, and a point for concentration of attacks.
Marc
Also, it centralizes POWER! There are many different lists with different policies and criteria. Some are based on technically verifiable issues (I can prove that x.y.z.q is a promiscuous relay), some are based on the attitude of the owner of the domain name or netblock, some on past record. You can pick and choose which one(s) meet the needs of your network and operation. Using these lists is a policy question for the network, and I would not like some external, probably unaccountable single point of policy.
At 13:14 3/4/03 -0600, you wrote:
Thus spake "Martin Hannigan" <hannigan@fugawi.net>
Not for nothing, but there's so much time wasted with all these diversified spam systems.
Many of these systems have been shown to falsely flag non-spamming sites, and the more reliable ones unfortunately don't catch a majority of spammers. This leads to a system where administrators (or users) can locally tune preferences for the level of paranoia they wish to suffer from. This would not be possible if there were only one model or provider.
I've been reading about Barry Shein's proposals and I have to say I am on board with a centralized -single- system based on his young, but intelligent, model.
If there were any single, centralized organization I trusted to do my thinking for me, I'd agree. This is also the same problem that PKI faces.
S
Stephen Sprunk "God does not play dice." --Albert Einstein CCIE #3723 "God is an inveterate gambler, and He throws the K5SSS dice at every possible opportunity." --Stephen Hawking
-- -=[L]=-