The Ultimate Anti-Spam System Non-Technical Specification

Almost all the mini-articles on this site end with the same moral:

"There's no way to really solve the spam problem as a whole. Buy SpamButcher, it can radically reduce the amount of spam you receive. Besides, we need the money."

After giving it a lot of thought, I'm not sure if this is entirely true.

So, ignoring the actual technical issue involved, what would this magical anti-spam solution look like?

1. It would have to make spam consistently unprofitable. Profit is the motivation behind almost all spam. No profit = no spam (or at least very little).

2. Can't make any new requirements of users in order to send or receive email. For instance, users hate "challenge / response" systems - and sometimes just won't play the "can you find the hidden phrase?" game.

3. As a matter of technical reality, the system will require that both new client software and server software be installed. However, users' existing email clients and servers need to continue working as-is without needing updates or additional software. Any system that requires users and administrators to immediately deploy new technology to continue using email will simply never be adopted.

4. Since existing email clients and servers are not immediately rendered obsolete, incentives such as improved email reliability will need to be offered to users and administrators to upgrade their servers and clients. Users should have reason to demand the system in the short-term, even if it doesn't prevent spam until its more widely adopted.

5. The system should eventually replace the need for conventional spam blockers and eliminate problems with false-positives.

6. The new system can't depend on any centralized servers that aren't already in use. Otherwise, they will be attacked by spammers and crashed.

7. The solution needs to be technical, not legal. People do all sorts of things that are illegal, but make money.

I obviously have something specific in mind here. It's not so much new software to eliminate spam as a way to "charge" computing time for email.

I'm not claiming I have any genuinely new ideas on this topic. I think I may have some unique insights on how to actually get them implemented. The trick is deploying them in a palatable fashion. More to follow.

Back