> I haven't looked at the implementation in OpenBSD extensively, but at
Well, perhaps you should, instead of commenting before you do. > a basic level there are two portions, the greylist function, and the > "waste their time" function, yes? I'm talking about bypassing the > first, not the second. Neither cost us. Neither is bypassable. > Even in the second case, if the spammer notices they're connecting to > something that will waste their (bot's) time, they can simply > disconnect and use the bot's resources to do something else. No spam was delivered. Again, what is the problem? > Not the > the spammers really care about wasting resources *that* much since > they don't have to pay for them (or very little for a bot herd > compared to "bulletproof hosting"), but it could make them a little > more efficient. No spammers care about wasted resources? I didn't know you were a spammer, and knew what they cared about. I guess their lack of wasted resources must be why they retry, like SMP demands. Except they don't. Perhaps it is not so simple? > The history of fighting spam has tended to show that if any form of > combating spam becomes too effective (and wide-spread), spammers will > invest effort figuring out how to defeat it. You're right. We should not try. This whole conversation is totally stupid. You don't use spamd, yet you want to discuss it. I think you just want to see your words on mailing lists.
