OpinionIt is technically feasible to dramatically reduce the amount of spam received whilst still accepting messages from strangers, without relying on message profiling. The difficulty is getting everyone to move to a new, agreed standard.
      – Lee J Haywood, 2008-11-10 at 18:04:07   (4 comments)

On 2008-11-10 at 18:10:42, Lee J Haywood wrote...
My first ISP account gave me my own domain name, and since e-mail was still being pushed by SMTP (prior to POP) I had an infinite number of e-mail addresses. So I could hand out a distinct address to every contact / site and block the ones that became obsolete/spammed. One issue with blocking spam is that your friends/contacts are vectors - if they get some spyware reading their contacts list then your address gets harvested. So then you have to block the address you originally gave them and issue them with a new one. This is, of course, extremely antisocial, so the process would need to be automated. Of course, the downside to my idea is that addresses are machine-generated and so typically difficult to remember. But even if you publish an address on your web site, you can expect people wanting to contact you to get the latest address from the page whilst reserving the right to change it if needed. Whitelisting old contacts would help a lot.
On 2008-11-10 at 18:19:55, BorgClown wrote...
That's an interesting idea, but far too complicated for the normal person to implement and keep updated. The root of scams is the ignorance of people who fall in them, so the ultimate solution would be to have a class in basic school about fraud and scams. That would help to combat spam, stock frauds, credit frauds and the like.
On 2008-11-10 at 18:27:11, Lee J Haywood wrote...
Well the issue is that the only way to eliminate spam is to change your address, which I'm sure plenty of people do on a regular basis anyway. Those who don't will receive a disproportionate amount of spam. My point was that a new scheme would allow the process to be automated, so even a 'normal' person could use it. You'd still have a relatively meaningful user name, but a set of keys associated with it to allow access until you decide it needs changing (one key per contact). This topic has been done to death, however - e.g. on Slashdot. No one scheme is likely to gain widespread approval.
On 2008-11-10 at 19:04:06, BorgClown wrote...
Yes, because automated solutions are treating the symptoms, but not the cause. What makes scams works lies outside of technology.