Re: On the spam problem...

At 17.17 -0800 04-01-31, Fred Baker wrote:
>I would say "yes", because there is not a uniform 
>definition of spam, nor universal agreement that spam is 
>bad. You think it is bad, and I think it is bad, but the 
>guy sending it seems to disagree. I have been to the folks 
>at my company that send spam (he bows his head sheepishly 
>in absolute shame), and they tell me that the email they 
>send is very closely targeted and resides within a vendor 
>relationship, so that (for example) if I download a new 
>driver for an aironet card from Cisco's site, it is 
>*legally* accepted for them to send me occasional mail 
>related to wireless networking. They are my vendor in some 
>sense, and within that relationship the rules are 
>different than the rules for people promoting v1@gr@.

This is very important. Every sender, and every recipient,
has his own defintion of spam. For 95 % of all mail, most
people agree, but not for the rest. That is why spam
filters never work 100 %.

The only solution I see for this is to make senders pay
a suitable chosen amount for sending mail to me. The amound
should be so low, that no serious sender is stopped, but
high enough to make it uneconomical to send spam to
millions of recipients.

Authetication and laws against certain obviously fraudulent
practices will also help. Like intentional misspelling
of spam words to get by spam filters.

This solution will not stop all spam, some are willing to
pay. Another problem is that it will stop mail which I do
want to receive, such as mailing list messages. Therefore,
I as a recipient must be able to say that I accept some
mail, for example coming from mailing lists I subscribe to,
even though the sender did not pay. It is the responsibility
of the mailing list owner to keep spam out of a mailing
list by only admitting messages within the charter of the
list (not too strictly, of course, not dictatorship, just
keeping the obvious spam out).
-- 
Jacob Palme <[email protected]> (Stockholm University and KTH)
for more info see URL: http://www.dsv.su.se/jpalme/