3 ms·
I think this is an interesting attempt, but will fail when it meets a persistent spam group, like most anti-spam systems. First, captchas can and will be broke
by Perceval 17y ago
I think this is an interesting attempt, but will fail when it meets a persistent spam group, like most anti-spam systems.
First, captchas can and will be broken/gamed or otherwise circumvented.
Second, trust networks are not new and are certainly not invulnerable to attack. Advogato has had one for a long time, and it was subject to a pretty run-of-the-mill but effective attack: http://www.squarefree.com/trust/trust.pdf http://www.squarefree.com/trust/trust.pdf
Combining several different anti-spam methods into a kind of layered approach is good, but it will hardly make it "spam proof."
- stcredzero 17y agoSome general (perhaps fictional) said something once about the useless of walls without men to defend them. (This is one of those things I can't Google easily. Perhaps someone else can give a try?) I've been watching the "web of trust" concept since the days of Phil Zimmerman and PGP. Seems like it would be vulnerable to balkanization and cliques, just like the non-virtual webs of trust that predate electronic networks. The only thing that works against spam is to destroy its economic utility. Perhaps, instead of a free-net, what's needed is a for-a-nominal-charge-net? The problem with the current net, is that much of the stuff that has economic utility, like advertisements, works best against the least savvy segment of the population. It's the too-dim-to-block-banner-ads, too-clueless-to-avoid-phishing, stupid-enough-to-fall-for-spam segment driving so much of the business of the internet. I hope we find a way to counteract this tendency.
- ajb 17y agoLycurgus of Sparta: "The city is well fortified which hath a wall of men instead of brick."
- Perceval 17y ago>Perhaps, instead of a free-net, what's needed is a for-a-nominal-charge-net? We had those. They were called things like Compuserve and Prodigy. And they didn't do well, as you recall. The problem with spamming is that it is very profitable. And it's not clear that a micropayment for each email sent, each facebook message, each blog-comment-spam is going to significantly cut into the fat margins of the most profitable spamming rings. At most, a surcharge will weed out the small fries, leaving only the big successful spamming rings without any competition. Not exactly a big victory. And like you say, the spam targets "the least savvy segment of the population." If only there was a way to directly link the credulous fools to the scruple-less cheats, we could achieve a worldwide system that more effectively fleeces the idiots while leaving the rest of us in peace. Alas, such a system is not on the horizon.
- stcredzero 17y agoIf only there was a way to directly link the credulous fools to the scruple-less cheats I suspect that some black-hat entrepreneur is already working on such a database and the tools for automatically harvesting such user's information. Or was that last line sarcasm? We had those. They were called things like Compuserve and Prodigy. And they didn't do well, as you recall. But those things had a fee for entry. To directly address spammers, without creating undue friction to participation, all you need is some sort of a fee for posting. But alas, the problem is with collecting the payment. One could argue that stackoverflow's karma/privilege system demands a non-monetary sort of payment, but they don't have this as a barrier to posting.
- DavidSJ 17y agoThe only thing that works against spam is to destroy its economic utility. Perhaps, instead of a free-net, what's needed is a for-a-nominal-charge-net? Sounds like you want hashcash (http://hashcash.org http://hashcash.org).