spam is a good thing (was Re: Spam IS Free Speech)

[Adam Back <aba@dcs.ex.ac.uk>]

John Deters <jad@dsddhc.com> writes:
> No, spam DOES cost the victim.  I'm not saying that the victim is the spam
> recipient.  Think of the ISPs that are the victims of Spamford's
> CyberPromotions.  Yes, victims.  [...]
>
> This situation is the exact situation that has entangled my ISP for the
> last couple of months.
> 
> Are you telling me that being subjected to a jack-boot investigation for
> running pyramid schemes *and* having your customers leave because they
> can't get mail services is a reasonable expense to bear because of
> Spamford's "right to free speech"?  Spamford's speech (or that of his
> "customers") isn't even directed at the sendmail operator or his customers.
>  The sendmail operator above is merely being used by Spamford as a
> megaphone to broadcast the message of spam to other people (who really
> don't want it, either, but that's beside the point.)
> 
> It's no longer the same as shouting down the marketing researcher.  
> 
> [your unmetered rented mega phone analogy]

The problem is that you and most of the rest of the internet world are
renting your mega phones/accounts out without charging for usage
volume.  You are also allowing completely free use of your account as
a recipient, and completely free use of your sendmail as a mail hub
service.

If this causes you grief, you need to start metering, and charging
postage to receive mail, and metering mail hub usage.

If I offered to supply a completely unmetered water supply, and there
were no clause in the legal agreement prohibiting it, I could use the
water supply to drive a mini-turbine and draw free electricity from
it, just pouring the water back down the drain.

If you have a cell-phone where it costs you to receive calls, and
people call you lots with junk marketing calls, I'd argue that it was
your problem.  You'd need to switch to a cell phone provider which
puts all the call metering charges on the caller.  (Cell phone
tarriffs are structured this way in the UK, I understand some/most
cell phones in the US, the receiver pays part or all of the call, this
seems a dumb arrangement).  If it's still economical for junk callers
to call you put a tariff on your line for non-designated callers.

It's a mean world, not every one plays nicely, if you offer free
services where the user gets some value, it _will_ get abused.  The
quicker crypto technology is used to fix this on the internet the
better, otherwise we get dumb government laws governing email usage
instead.

See:
	http://www.dcs.ex.ac.uk/~aba/hashcash/

for some thoughts on a short term solution to metering.  Long term
solution use ecash.

> In this particular case, of course, there was a technical solution:
> install a sendmail to prevent routing of incoming mail.  ISPs and
> corporations around the globe will need this new hardened sendmail to keep
> the spammers away.  Restricting the speech they carry.  Turning off the
> megaphones.

Long term solution charge for your IP packets per packet.  If you get
lots of "business" use the proceeds to buy more bandwidth, or put up
your prices.

Of course I sympathise with your plight on the recieving end of this
crap in the meantime.  

Try if you can to think of it this way: spam is a good thing because
it draws our attention to underlying vulnerabilities in internet mail
transport protocols.  It's better that we are incentivized to fix
these problems (and Spamford/cyberpromo is doing us a service by
providing this incentive), than it is to leave it to governments to
"fix" the problems by dragging laws into it, which will likely result
in "Internet drivers licenses" and other undesirable effects.

Cypherpunks, I think should be involved in providing crypto solutions
to allow metering for services, using anonymous ecash.  If we don't do
it, someone else will, and it won't be anonymous.

Adam
-- 
Have *you* exported RSA today? --> http://www.dcs.ex.ac.uk/~aba/rsa/

print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<>
)]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`