An update on the AOL pay-per-message issue. I have given a round of press interviews today and it looks like there is going to be a lot of coverage today and tomorrow. AOL have backed off on their plans to phase out the (free) whitelist by June 30, but it seems clear that their long-term direction remains pay-per-message. I was surprised, to say the least, to read the following in a recent article, in which I was quoted about the monopoly situation that AOL's decision would create: > [Eric Thomas] said the main problem is that AOL is only using one > technology supplier for its new certification plan, leaving Goodmail in a > position to become a monopoly. > > "Once the system is in place, nothing would prevent Goodmail from raising > prices to increase profits," Thomas said. "Higher certification prices > would lead to lower e-mail volumes and reduced operational costs for AOL, > so they would be unlikely to complain about any such increases." > > AOL's [spokesman] Graham said this was a hang-up Thomas and L-Soft would > have to deal with. Well, I am not ashamed of having a "hang-up" about monopolies. At this point, I have very little faith left in AOL's long-term plans for the existing whitelist. There has been a downpour of bad press, they have adjusted their position to control the damage, but industry concerns about the single-sourced nature of the certification are seen as a hang-up that we all need to get over. The long-term plan is still to make us pay ridiculously high fees for the privilege of not having our mail tossed in junk folders where subscribers will not see it. I think we need to react by creating an open standard for e-mail certification, something that could be implemented rapidly and that could support the creation of low-cost or even free (community-run) certification services. And the good news is that this can be done through a very simple extension of DomainKeys. Just picture the DomainKeys we have today, but with the public key stored at the certification company, rather than at the sending company. Same checksum algorithms, same overall structure with a few minor changes, and you get all the benefits of DomainKeys plus all the benefits of certification and reputation checking. That open platform would allow anyone to offer the service at any price, and it would work with any up to date mail software. Goodmail could continue to use its proprietary platform, although I suspect that it wouldn't have the same market penetration as the open design, but that is the drawback of proprietary designs. Nobody would be locked out, and everybody would win (except of course the monopolists). Eric