« And So It Begins |
Main
| Steam kills »
Spam and Cyberterrorism
There's a brief bit by Bruce Sterling in Wired online on cyberterrorism and spam and what should be done. Worth a read, though I don't agree with all his conclusions.
Posted by Andrew Case at August 09, 2004 08:58 AM
TrackBack URL for this entry:
http://www.transterrestrial.com/mt-diagnostics.cgi/2791
Listed below are links to weblogs that reference
this post from
Transterrestrial Musings.
Comments
Ok, I've known about the spam problem for 10 years- I received the very first spam on USENET; over a decade ago. I quickly worked out the ramifications of that; and I really did not like the non scalability one little bit.
So far as I can tell, the only long-term solution to spam is pushing back the spam filtering to the sending ISP rather than the receiver. For that to work, it requires cryptographic ISP sender identification; so we can blackhole any irresponsible ISPs that aren't filtering properly.
Once that works, spam should fall away; but there's always going to be some naive people that haven't worked out that spam is a bad thing, and happen to make it past the sender ISPs spam filtering for a while before getting jumped upon, so a tiny residual spammage will remain. That can't be helped. The important thing is to destroy the economic incentives to spam that exist currently.
At the moment, it looks like Microsoft is dragging their feet over it; they seem to have some patents, and they are doubtless looking at using them to stick it to the open source Apache. Somebody is probably reading them the definition of 'business ethics' as we speak, but I don't expect them to remember the definition.
Anyway, that's the solution, pretty much; prefiltering at source.
Posted by Ian Woollard at August 9, 2004 09:31 AM
The important thing is to destroy the economic incentives to spam that exist currently.
Duh, maybe we could change the pricing model and charge by the mailgram--the cell phone folks already do it coming and going. I'd be happy to read ALL my spam at $0.25 a pop! Give the developers a 1% rakeoff on that, and it'll get implemented a hell of a lot faster than cryptographic ISP sender identification.
Posted by at August 9, 2004 10:46 AM
Micropayment systems like that aren't very well developed; and more expensive than generally assumed.
Attempting to monetarize something like email that inherently costs tiny, tiny fractions of a penny to deliver shouldn't really fly. Then again the cell phone companies did it already with sms messaging, and the public bought it, so I guess the world isn't totally sane.
Posted by Ian Woollard at August 9, 2004 06:42 PM
Post a comment