Nothing like a shower of gold to corrupt innocence. While Zeus took this nugget of wisdom quite literally in order to seduce DanaŽ, other blushing maidens guided by ethical governance, i.e. modern corporations, prefer to act themselves. Charging 1/4 of a cent per email delivery is a drop in their buckets but no doubt AOL and Yahoo! expect to gather a shower of gold.
For Chris Nuttal (*), this move announced in March '06 threatens to open a new Pandora's box. This is indeed a naked exploitation of quasi-monopolistic powers, kin to the campaign by telephone companies to introduce tiered-pricing for Internet data transmission. The issue is not, as proponents would have us believe, one of price efficiency. Rather no unregulated economic entity will offer two classes for what has become an essential service without making sure no one would ever want to receive second class service who is not already in desperate straights.
Often such a move is used to contain costs and speed up modern technology adoption by consumers, as when a bank replaces 5 tellers in one of its branches with a mix of 3 ATM's and 1 teller. When done too nakedly, this has a negative effect of course on the image of the company but what monopolies care for customer service? Ten days ago I went at the Gare du Nord in Paris to pay for a railroad ticket with a check, which automatic ticketing machines do not accept. I happened to notice the SNCF was not staffing all its windows even though its posters warned the time slot I was there had the longest lines.
In the case of email, poor customer service is an understatement. Remember Internet access companies justify their first class mail by guaranteeing delivery. Since they also filter all mails prior to delivery, one is forgiven if one thinks this nice way of boosting revenues looks rather like a protection racket and extortion scheme: pay now or unfortunate accidents may occur later.
Nevertheless the pretense used by AOL and Yahoo! has a basis in fact. There is indeed an issue of price efficiency which must be addressed as befits an essential service. That email is an essential service is proven by the proposal to tax it in Europe according to the report filed by Thomas Crampton (**), a modern equivalent of the salt tax of the Middle Ages, the gabelle. That free email as we know it is unsustainable is driven home by the prevalence of spam, i.e. undesirable mails, despite the best efforts of all (1).
One must admit that assessing a price on individual emails, whether it be a commercial charge or a tax, is a natural response. When Esther Dyson defends Goodmail (2), the system behind AOL's certified email service, she invokes "a market failure". Have I not said the same when investigating doctor practice profiling? Isn't spam a form of pollution stemming from the failure to set a fair price on email emission?
If we look beyond polemics, the debate is not about whether there is a problem but about the nature of the correct solution. Taking pollution studies as a template, one must first determine what is the abundant resource which is nevertheless depleted by untrammelled exploitation.
Is it Internet bandwidth?
As far as I can tell, spam is not the reason why telephone companies claim they must invest in better transmission services. They speak instead of video services. Truth be told, Internet has been such a resounding success that bandwidth is truly a free resource as far as text messages are concerned. We speak of marginal costs of course. But relevant fixed costs are adequately covered by the subscriptions paid by all users to "access Internet". Charging for email transmission would therefore be dangerous. Either legal market competition and illegal contraband would quickly undermine such a pricing scheme or monopolistic control of the infrastructure is powerful enough to sustain artificial rents. Given current conditions, I am far less sanguine than Esther Dyson about the power of competition when markets are dominated by a few giants such as AOL and Yahoo!. No doubt Google has not adopted Goodmail but it exacts an even higher price: permission to read all mails.
Is it email emission?
Hardly. The cost of sending emails is already born by the senders. The staggering number of undesirable emails, most of them ignored by receivers or filtered out during transmission, is proof enough that this cost is low. Of course its value is high: email based marketing generates enough sales despite the low ratio of orders received per number of email sent. AOL certified email scheme is nothing but a bully's attempt to extort part of the added value of the merchant to its own benefit.
Could it be perhaps email reception?
From the individual perspective, every undesirable email is spam, whether it comes from some obscure online shop, likely to go second class, or the most prestigious company, who will pay for first class. Laws are irrelevant from this perspective (see a forthcoming lecture). Indeed SNCF, the French railroad company, is responsible for at least 5% of the spam content of my French mailbox after filtering by my French service provider. Certified emails will be a small price to pay for such top tier suppliers to avoid getting caught in a well tuned filter.
- setting a price on sending mails first class does nothing to prevent spamming by second class spammers
- but certifying delivery to first class senders does guarantee more spam by spammers who can afford it
Making senders pay to guarantee email delivery is but a scheme to transfer the benefit of a "free" resource, receiver time, from senders to Internet service providers.
What then should be the solution?
It is based on a very simple principle: no email should be sent, let alone transmitted, without the explicit prior authorization of the intended receiver. I am not here speaking of the so-called opt-in approach as required by European law. This law is too easily flaunted by established companies through bundling, another form of extortion. For example, at the voyages-sncf.com site, no seat reservation can be confirmed if one does not opt-in to receive spam from SNCF.
The solution I have in mind requires each receiver to set up and post his or her own filter for all would be senders to download and process against their own profile each time they want to send an email to this receiver. Last week my 06/20/06 fillip included an architecture for so doing and one can look to (3) for more details
Granted this solution requires some effort from each individual receiver, but it is consistent that receivers bear the cost of protecting their precious time. Besides no one needs to bear this cost if one prefers the present situation.
For better precision, one should implement filtering by domain of interest. One could easily conceived that:
According to this solution all costs are paid by the receivers in need of protection. Such a solution may seem paradoxical since it requires the victims of spam to pay. But people do pay for locks. And once a receiver controls all would be senders, he or she is able to extract a fair value for being sent a message to read. In most cases this fair value will simply be access to useful opportunities as determined by the receiver. Notice that spam disappears ipso facto. I, for one, may want to be alerted to personalized promotions from SNCF as long as they satisfy my criteria and products to boost certain activities are indubitably desired by some older men.
- a generic filter would be distributed and posted for free as part of an Internet access service
- more specialized filters would be available for a price from domain specialists
- such specialized filters would be posted for a price
Finally Internet service companies would be able to charge senders for real added-value services such as bulk directory access to filters in a particular domain, as done currently by address database merchants.
- (*). AOL opens Pandora's inbox of protest with security move, by Chris Nuttall (Financial Times) - March ??, 2006
- (**) Idea for Electronic Message Tax Prompts Swift Outcry in Europe, by Thomas Crampton (The New York Times) - June 12, 2006
- (1) spam makes up 60 percent of all email according to Ferris Research, quoted from Saul Hansell (The New York Times) - February, 2006
- (2) You've Got Goodmail, by Esther Dyson (The New York Times) - March, 2006
- (3) see US Patent Application 2006/0053279 A1 by Coueignoux