The problem with "making it computationally expensive," is that it's
pure thermodynamic entropy, and in an era where computers account for a
large slice of overall energy consumption, it translates to increased
energy consumption that produces nothing.
By analogy, like trying to reduce frivolous automobile usage by forcing
cars to drive around in circles before they can go to their destination,
as compared to just raising the tax on gasoline.
Further, as the recently exposed Bitcoin botnet schemes have
demonstrated, bad actors can simply export that entropy load to innocent
parties, who for the most part remain blissfully unaware that their
machines are busy grinding away for the benefit of the scammers.
Whereas charging a fee per outgoing email, even at the level of 1/100 of
a cent, also helps pay for the communications infrastructure, which is
part of an overall strategy of breaking free from the
"advertising-driven" (surveillance-driven) model of the internet.
Driving around in circles produces nothing. Increased gasoline taxes
fund public services.
This also provides a means of alerting users when their machines have
been hijacked, via an odometer-like display in their email app. If you
see your odometer incrementing, while you're not sending email, someone
else got into your machine. And the fact that there's a cost, even if
it's only ten cents per thousand emails generated on your machine, is
also a direct incentive for people to secure their machines and
investigate when they start seeing their cost-odometer move while they
aren't sending email.
BTW, the name for the cost-per-email proposal, via Jaron Lanier, is "a
homeopathic dose of postage." (That does not imply belief in
homeopathy, only that it fits as an analogy.)
On 13-06-11-Tue 7:55 PM, William Budington wrote:
On 06/11/2013 06:32 PM, GtwoG PublicOhOne wrote:
There is a way to stop spam dead in its tracks,
which is to charge a
teeny-tiny fee for every outgoing email. This would destroy the
economics / business model that allows spam to thrive on "free" email.
(As Richard Stallman says, "free speech isn't the same thing as free
beer.") However that won't help if scumbags break into networks and
steal user accounts.
Back in 1997, Adam Back came up with a similar concept.
making sending spam financially expensive, make it computationally
expensive. He proposed a system whereby the email would have to compute
a hash with certain characteristics in order to be valid. He called
this a Proof of Work, because your machine had to work to generate a
valid hash that the receiving server could verify instantaneously.
Generating such a hash for an individual email is relatively easy, but
you'd have to generate it for each email you want to send. This quickly
becomes computationally (and thus financially) expensive, when you want
to generate thousands of emails at once.
Eventually his Proof-of-Work system became the key element of the
The reason why it didn't take off back then is that it's not just the
spammers who are sending bulk emails. With mailing lists, every email
you send is multiplied by the number of people on the list. So those
costs add up pretty quickly, even for people who aren't sending out
sudo-discuss mailing list