I’ve had lots of conversations over the last few years about ways in which rising marginal cost could deal with grotesque abuses of online services. There are probably a dozen posts in this blog about that subject alone. Now the obvious example of a place where this kind of thing has been proposed has been e-mail – people have been talking about ways to get people to pay for e-mail “stamps” for years as a possible means of avoiding spam. Bill Gates has proposed another version of this scheme recently. His idea – ten-second pieces of computing time on the machine that sends the mail being given to some worthy cause (or to just solve some abstract puzzle). This would – apparently – be a gesture of good faith on the part of the sender that a spammer couldn’t possible match.
Now, my personal opinions about rising marginal costs have mainly been about how to deal with noise, distance and abuse in online communities. I once touched on the issue in connection with e-mail (only because e-mail was a suitable jumping off point) and ended up in an almighty fight with Cory Doctorow about it. Since that time, I’m still of the opinion that exponential graphs of effort or diminishing causality over space or increasing marginal costs (all features of the real-world) still have a role to play in how we solve gross abuses online. On the other hand I’ve seen no evidence that there’s a model that works particularly well with regards to e-mail. Certainly my experience of sending the fifty or sixty e-mails I send from my personal account a day (and the other fifty or sixty that I send at work) wouldn’t be radically improved by having my various computers churn through puzzles for twenty minutes a day.
With regards to the 1p-per-e-mail approach – I’m still of the opinion that a more successful version would be about the redistribution of money rather than the paying of it. What if the person you sent your e-mail to got the 1p you spent to send it to them and could then use that penny to send an e-mail in turn to whosoever they wanted. In those circumstances, most users (who get as much e-mail as they send) would be financially unaffected, the spammed would get a financial reward for all the rubbish they were forced to consume (there might even be a legitimate business model in collecting spam) and the spammers would end up paying much much more money than before.
This is not a new idea either, and nor do I think it’s a particularly practical one, but it does present some interesting opportunities to think about e-mail in very different (ludicrous?) ways – perhaps eventually even as a unit of currency that you write upon and distribute. After all noted currency is only an abstraction of value written on a rectangular piece of paper – why shouldn’t our future currency be based upon the transactions of plain text files…
3 replies on “On e-mail as a means of exchange…”
http://www.abstractdynamics.org/linkage/archives/001991.html
pay for email, with a twist…
How would this work for email discussion lists? A straight 1p per email recieved would mean it would cost four quid to post to a list with 400 subscribers. And for a 100 message a day list, lurkers would get a quid a day just for subscribing!
An alternative would be for the 1p to go to whoever’s running the list server, and for there to be no payment to list subscribers. I wonder how this could be implemented technically in a way that couldn’t be exploited by spammers.
Tom,
Interesting post and thank you for the opportunity to comment.
As an aside, think how boring the online world might become without abusive human behavior as represented by spam on the Internet to talk or rage about.
No conferences on spam or online anti-spam discussion groups.
Look at all the businesses which have sprung up to fight email abuse from filters to routers and everything in between.
The development time spent on designing new solutions to plug the security hole in sending email or otherwise deal with spam.
Publishers publishing, lawyers suing, legislators legislating, developers developing and the list goes on and on.
What would we do? And were would be without our Technorati or Feedster?
I am reminded of Rumpole of the Old Bailey as he raised a glass of his favorite claret at the pub to toast the criminal class.
Of course, I make these comments somewhat tongue in cheek.
Turning to your post, raising the marginal costs of sending abusive email is one potential solution and as you acknowledge has been discussed in the past.
Apart from the solution itself, another problem as pointed out by the previous commentator is distinguishing between “legitimate commercial email” – for example unsolicited direct marketing email sent with prior consent(to use the terminology of the EU directive) and the stuff sent without.
The underlying difficulty is the security hole in the mail transfer protocol.
Therefore solutions like the Sender Policy Framework hold promise. At the same time, using concepts as being now being implemented with the Turn Tide Router (a network solution, as opposed to a sender solution) to raise the marginal cost, while seemingly differentiating between legitimate commercial email as defined and the ugly stuff have potential.
However, even though both CAUCE and Microsoft like the Trusted Email Open Standard, to date, the internet access services seem unwilling to give up some element of control. Unfortunate.
Thank you again for the opportunity to post my comments on this otherwise heart wrenching topic, which can seemingly almost raise people from the dead and cause the House of Lords to snicker about spam, spam, spam.
Kind regards,
John Glube
Toronto, Canada