Slashdot Mirror
Spamhaus Guru Steve Linford Profiled
BenLev writes "The New York Times has an article profiling Spamhaus Project director Steve Linford. The feature goes behind the scenes at Spamhaus, 'one of the leading groups that is trying to make the world safe from junk e-mail', showing that it operates from Linford's houseboat on the Thames near London, spammers don't like him, and his volunteer corps likens itself to the X-Men."
What happened to that proposal to add records (as comments, so the DNS protocol wasn't broken) to the DNS saying that a domain was authoratative for the envelope 'From ' header ? That sounded like a good idea, so long as the MTA's took it up...
Simon
Physicists get Hadrons!
My particular server (a dedicated box) was innocent, but my hosting facility had spammers on other dedicated boxes.
Isn't blocking a /20 like swatting flys with a hand grenade?
I'm surprised no-one has thought through the logical conclusions of where we're going with spam.
... since there is nothing serious happening against any of these directions, the conclusion seems unavoidable. What I'd like to say is that
Spam filters work only for those able to configure them. For the vast majority of Internet users, they are just a dream.
Spam blacklists are unsustainable in a world where most net connections come across DHCP, and most spam is/will be sent from owned home computers.
Spam merchants will continue to harness the 'dark side of the force', paying crackes and virus writers to create the networks of owned machines they need to operate from...
the Net will split into two halves, an "infected" and a "clean" part, and every single transaction from the infected part will be treated with scrutiny and suspicion.
But this is impossible too.
Conclusion: the purity of the net is a thing of the past. We will come to understand that traffic is bad until demonstrated good. Emails will be 99.999% junk, virus, and trojan, and the art will come not from filtering out this junk but from detecting the signal within the noise.
Clearly, whitelists are part of the solution but they are limited since you can't form a network of whitelists, it's a one-to-one solution that does not scale.
I see only one solution that is scalable. Data clearing houses. You register with me, I'll vouch for all your data, and pass it on to those who need it, along with my signature. A trust network, if you like.
Data clearing houses will rate each other, creating a system of moderation in which data is never guaranteed good, but at least you get a measurable index of confidence.
Ceci n'est pas une signature
I doubt that any progress will be made in fighting spam until Microsoft/Apple include authentication options in their default mail applications.
Unfortunately, authentication is unlikely to do much to stop spam unless people use it with a personal whitelist of permitted senders. It is currently straightforward to track a spam email (SpamCop can do this if you paste the email in with full header information) but nowadays it typically comes from a cable/DSL user whose machine has been hijacked.
We (the dev firm I work for) was approached to do spamming. I went apeshit and said I would quit and report it to the FCC but they were offering real money and my bosses were tempted. So I turned off the blacklists and let them swim in spam a few days till they came to their senses. When he had 100+ spams everyday, he saw the light.
If you don't say anything, you won't be called on to repeat it. -- Calvin Coolidge