Esther Dyson Is Wrong — Most Email Will Remain Free

Ideas for email senders to be charged for the privilege crop up from time to time. The idea is that email senders should pay per message, much as postal mail senders do. The latest call for such a scheme is Esther Dyson's op-ed piece in Friday's New York Times. We think this idea is misguided.

I like Esther Dyson. I first met her in the early '90s, and found her thoughtful, insightful, and straight-talking. Dyson begins with a refreshingly accurate, measured description of Goodmail and its partnerships with AOL and other email providers. She goes on to belittle those who aim to boycott Goodmail and its partners. Those who keep up with our newsletters or blog will know that we fully agree with her on this point.

However, Dyson goes on to say, "Pretty soon sending most e-mail will cost money, but I think that's only right." We disagree -- it won't and it isn't. While we see a role for companies such as Goodmail, we don't agree that "most" email will cost money in the future, and it's not "only right" because the fact that email is free is part of its appeal.

Her argument stands or falls on the assertion that today's spam filters aren't working -- Dyson asserts that, "The senders of 'bad' mail are getting better and better at defeating them." However, it's clear to us that, although the smarter spammers are making their messages trickier to filter, the filters are also getting better.

All in all, today's state of the art in spam control solutions is far ahead of where it was, say, two years ago. Improved spam filters being available to more people -- plus laws that allow the citizenry to penalize spammers -- will cause the scourge of email spam to wither and die.

... Richi Jennings

One Comment

  1. Posted March 20, 2006 at 2:41 AM | Permalink


    I was surprised to read your comments … “Esther is wrong” … as you seem to be arguing that just because the spam filters are getting better that in some way this is going to fix the problem.

    I believe that we MUST address the cause of this issue NOT the symptom. Stopping SPAM at your corporate gateway does not really help – for the SPAM has already consumed ridiculously high amounts of internet bandwidth BEFORE it hit your gateway.

    We must STOP this at the source of the problem. And thus, in some ways Esther is on the right track. Charging for email (as proposed by Gates in 2004) is one way of reducing the CAUSE of the problem. However, authenticating every email address and having a receiver “opt-in” option would be even better.

    Keep up all your other good work !


  2. Posted March 20, 2006 at 5:10 AM | Permalink

    The thing that people — even smart people like Gates and Dyson — miss about pay-to-send schemes for email is that spammers are criminals who already sell fraudulent products and who routinely steal access to other people’s computers in order to send their spam. They were criminals before CAN-SPAM and they are still criminals afterward. What, therefore, is to stop them from using other people’s money to send their spam? Surely their conscience won’t, so we’ll need to rely on security measures implemented in software. We already know, however, that malware writers are successfully targeting their victims’ bank account login and password information. Given the poor state of security on tens — maybe hundreds — of millions of broadband-connected PCs, it’s highly unlikely that it will be any more difficult for the bad guys to sniff their way through the security mechanisms that control email payment accounts. Pay-to-send will therefore be a fool’s errand until there is a worldwide rollout of truly secure desktops.

  3. anonymouse
    Posted March 20, 2006 at 6:42 AM | Permalink

    Dyson’s comments are part of the new guard moneymakers who are creating a two-tiered internet — your arguments against groups that detest Dyson’s Goodmail are in vain: this is exactly the kind of thing we are expecting from here and others of her ilk (read: telcos)

  4. Posted March 20, 2006 at 8:20 AM | Permalink

    Whether or not Esther is right about charging for email, it is nonetheless the case that Goodmail’s approach of signing messages with a token will not help with the resource consumption problem posed by Spam. Only approaches that work at the network layer — and to a lesser extent those that work at the envelope layer — will be truly effective “reputation” based approaches to removing the infrastructure cost burden of spam.

    Also, it’s very important to remember that Goodmail is only focused on the very large ISPs — that is the only place their value proposition makes lots of money, since it’s the largest ISPs who have the lion’s share of “eyeballs” which the marketers are trying to reach. For enterprises and smaller ISPs, pay-to-mail will not enter the equation for many years to come — if ever.

  5. Posted March 21, 2006 at 5:25 AM | Permalink

    Here’s the fundamental issue I see with Goodmail:

    Does the fact the sender is paying make an email solicited? No.

    An unsolicited email is STILL unsolicited. Why would users see this as any different to conventional spam?

Post a comment

You must be logged in to post a comment. To comment, first join our community.