E-mail was the internet’s original killer app – the service that spurred the multitudes to go online. Now, it may turn into a very different kind of killer app – the one that kills the traditional internet. In an article in the New York Times (reprinted by CNet), Saul Hansell reports that email giants AOL and Yahoo “are about to start using a system that gives preferential treatment to messages from companies that pay from 1/4 of a cent to a penny each to have them delivered.” The tariff will, the companies argue, make spam easier to identify and manage. The companies also, as Hansell notes, “stand to earn millions of dollars a year from the system if it is widely adopted.”
The tiered system raises a much bigger issue, of course. By drawing a commercial distinction between how different types of digital content are treated, the system would seem to be the thin edge of a wedge that could utimately destroy the “network neutrality” that has defined the internet up to now. As Hansell explains, “the move to create what is essentially a preferred class of e-mail is a major change in the economics of the Internet.” If AOL proves the viability of creating a tiered pricing system for email, would that provide its parent, Time-Warner, with a model and a precedent for introducing a broader tiered system for delivering internet content through its big cable business?
This move shows how difficult it is becoming to see the internet in the black-and-white terms of the past. Whether it’s the censoring of digits or the pricing of digits, we’re now moving into a grey area, and that’s where the future of the net, as a commercial, social and technological force, will be determined.
This idea has been around for a while, and the article mentions Bonded Sender, but nobody has ever been able to make it work on a large scale. Mostly because nobody else wants to pay. It’ll be interesting to see if the rent-seeking works this time around. I’m dubious, based on the history.
This appears to be aimed specifically at making commercial email senders pay for a short-cut around spam filtering, and for that purpose it might work well. Unfortunately, there are many out there who are deluded and believe that a postage-based system can be generally applicable. Belief in technology that makes senders pay for email messages sent was one of the key ideas that caused Bill Gates himself two announce that the spam problem would be solved in two years. He made that announcement at Davos two years ago, by the way. My junk mail folder says that he was wrong. He does seem to have backed down a bit from his belief in the sender-pay solution since then, fortunately.
I say “fortunately” because the folks who believe in economic solutions to spam are all ignoring one fundamental thing that completely dooms these solutions: Spammers are criminals. They won’t pay. They’ll find ways to make other people pay for their spam. They already routinely take over other people’s computers. Until hundreds of millions of computers installed around the world are replaced with new hardware and software that is far more secure than what we have now, sender-pay solutions are a fool’s errand.
Hi Nick,
I’m a co-founder of Goodmail and an avid reader of your blog (http://dreymann.blogspot.com/2005/10/religion-920.html).
I am quite puzzled when people associate net neutrality with CertifiedEmail. Net neutrality is about allowing me, as a customer, to stream my videos from Google if I chose to, or download images from Flickr (if that’s what I’d rather do). Those are transactions initiated by me, the customer. I am already paying my telco/cable for internet access; I should be able to grab data from whatever corner of the internet I fancy.
Until the introduction of the first stamp in the UK in the 1830’s, postal systems were based on a recipient pay model. If you wanted to get a letter delivered to you, you had to pay the postman. Imagine the post office would require you to pay for all the unsolicited credit card offers you must get daily!
In all type of communications, “calling party pays” is the winning model. One could argue that the reason the US was a laggard in the mobile phone arena was the decision to charge users for incoming “air time”, rather than charging the callers. GSM took the approach of always charging the caller and making incoming calls free; mobile phones minutes in Europe and Asia soon dwarfed US usage.
We could rename “calling party pays” to “initiator pay”; this terminology change would address the debate about net neutrality.
Email is not a pull medium. Messages are pushed to consumers. There is no absolute mechanism to assess whether a message is wanted by its intended recipient or not. The mailbox belongs to the recipient. This is why anti-spam filters applied by ISPs, for years now, are legitimate – they’re trying to keep the bad stuff out of their customers’ mailboxes. Spammers have tried to cry “net neutrality” and make filters illegal — they lost. Consumers don’t want their messages and are happy when their providers block spam, viruses and phishing attempts. I believe courts have blessed this practice of filtering messages in several precedent setting rulings.
Problem is: no filter is perfect. Bad messages go through and “phish” innocent users. Good and wanted messages are blocked (just look at the caption for the NYT graphics: “Of legitimate commercial email, 20 percent is caught in spam filters and not delivered or sent to junk mail folders”).
The CertifiedEmail service guarantees good messages are not blocked and lets the recipient tell apart real bank statements from phishing attempts.
A good thing, isn’t it?
This is a premium service. Bulk senders are not forced to take advantage of it. If they want to assure delivery of their messages and to relay authenticity to their customers – they will. Senders that want to keep using the current probabilistic system certainly can.
Best,
Daniel
Daniel,
Thanks very much for your message. I agree that what you’re doing is different from what’s usually meant in discussions of potential infringements of net neutrality. And it’s a good business idea (if you can pull it off). But I think your “puzzlement” about why “people associate net neutrality with CertifiedEmail” is a rhetorical kind of puzzlement. I don’t think net neutrality just means freedom of access for the user. I think it means that the network, and those organizations that control it, are neutral about the content flowing through it. When you speak of your service being a “premium service,” you’re talking about drawing distinctions between different kinds of content flowing through the network. That doesn’t mean it’s not a valuable service; it just means that it requires something other than pure network neutrality to work. No?
Nick, as you agree that what Goodmail is doing is different from what’s usually meant in discussions of potential infringements of net neutrality, I am no longer puzzled (rhetorically or otherwise). In the email world distinctions are already drawn between “different kinds of content flowing through the network”. There is an overwhelming consensus against being neutral about spam and phishing: filters are using artificial intelligence and manually tweaked whitelists and blacklists in an attempt to route messages about “herbal Viagra” to the spam folder, and legitimate messages from pharmaceutical companies to the inbox. Our endeavor is to do the same in a more rational way and with much better results. So, the answer to your question is: yes, it does require something other than pure network neutrality to work, but in the email space this pure network neutrality is long gone.
Daniel presents an interesting argument, it sounds convincing, and I have to admit I heavily filter my email. I get about 150 legitimate email messages a day. I get much more spam than that. My current rate is about 300 messages a day black holed using rule based filters without me even looking at them and another 20 or so pushed into a junk mail box which I look at every so often. That’s a lot of spam, if I had to wade through it by hand I wouldn’t really have time to do anything else…
However Daniel’s statement that “…of legitimate commercial email, 20 percent is caught in spam filters and not delivered or sent to junk mail folders” misses the point, I get to decide what’s legitimate, not AOL.
The news that AOL is moving to a tiered structure, coupled with rumours that Google might be trying to establish its “own private internet” got me thinking about how things used to be, maybe the Internet is broken, and maybe the Google-net isn’t a bad thing after all?
This is the problem with putting intelligence *inside* the network: The people at the nodes have no control over it.
On my PowerBook I’m running my own postfix mail server, which both sends and receives my email. A blocked port 25 is a deal-breaker for me. This way I can send email wherever I am without having to worry about configurations.
The internet is just a dumb network which moves around IP packages… nothing more. Any attempt to make the network smarter will make it less the internet.