Email Postage: Payment and ResponsibilityIf an ISP were to charge a fee for delivery of email to its users, what expectations would you have about the success of that delivery?
Brian Krebs of the Washington Post reported recently on a study by University researchers in California that dissects some of the economics of spam email. Although it presents a sobering picture of the gullibility of the average email recipient, it also suggests that the return on a spammer’s email blast is much lower than previous estimates. It takes about a billion messages to get a return of less than $10,000.
This has contributed to a renewed discussion of “email postage,” that is, the idea of charging a fee for every email message, similar to the USPS and other postal services charging for every physical letter. If it cost as little as 1/100 of a cent to deliver each of those billion messages, proponents say, all spammer profits would be wiped out and it would no longer be worth their while to flood the net with junk. At such a tiny rate, postage bills for ordinary users would total less than a few dollars, and usually only a few cents, per year, and could be absorbed into the fees already paid for internet access, so only senders of millions of messages would even notice.
Another advantage to a “sender pays” model of email is that it might suddenly become very expensive to have your personal PC taken over by malicious software and turned into a “spam bot.” A few days after Krebs reviewed the UC research, he also reported on the takedown of McColo Corporation, an ISP that appears to have been hosting the command controls for a large number of bots. Almost immediately following McColo’s loss of connectivity, sites that monitor the volume spam reported that the amount of junk email reaching their sensors had fallen by 30% to 75%. Clearly botnets are a huge contributor to the spam problem, and anything that helps to identify compromised PCs and encourages cleaning them up, is a good thing.
Unfortunately, spam reduction isn’t a sufficient incentive to put an email postage system in place. Ignoring the technical hurdles involved in organizing reliable payment for “stamps” and preventing forgery or re-use of stamp tokens, consider the shift in attitude that would inevitably accompany an imposition of fees by an ISP when accepting email.
In today’s email infrastructure, the majority of the cost is borne by the recipient’s mailbox provider. ISPs and other providers owe nothing to the senders of email; they charge the recipient, either directly or by sprinkling their user interface with advertising. Consequently the provider has no obligation, and no incentive aside from maintaining the goodwill of their users, to provide reliable delivery of email that originates outside their own network.
An email postage system would turn that on its head. When you put a stamp on a letter and turn it over to the postal service, you have both a belief and and expectation that the post office is going to do its job and carry that letter to the addressee. Would it be any different if you bought that stamp for an email message?
Accepting payment from the sender of an email obligates the ISP to make a best effort to get the message into the recipient’s mailbox. The ISP becomes liable to both the sender and the recipient if a system failure results in the loss of paid email. No more silent acceptance of email which is then discarded undelivered, and maybe no more junk folders unless each mailbox owner creates his own personal set of junk rules. A strong expectation that anyone willing to pay the full postage rate should be exempt from delaying tactics like greylisting and rate-limiting.
I don’t think ISPs are prepared to take on that responsibility. I think they’d rather stay right where they are, in complete control of what they allow on their networks and free from real obligations to provide reliable email delivery. Right now they’re the good guys, valiantly defending their users from the relentless siege of an army of baddies. Postage would be tantamount to taking bribes to let the baddies in; consider the (largely misguided) fracas that resulted when AOL announced a partnership with Goodmail. Does a tiny fraction of a cent per email outweigh the cost of implementing a postage system and marketing it to the users?
I’d like to be wrong about this. What are your thoughts? How would you react if you had to pay postage for your email?
