Original URL: http://www.theregister.co.uk/2010/05/09/email_and_trust/
Email 2.0: Trying to catch up with the web
Reports of the death of email greatly exaggerated
INET Remember the days when you clicked on a link in an email without worrying about hitting a porn site or letting loose a virus?
Well, these days could be back with us soon if the engineers at INET SF have their way. “As we moved forwards to Web 2.0, email moved backwards to 0.7,” explains Daniel Dreymann of Goodmail Systems, pointing out that email systems these days pull out just about anything that isn’t plain text. “We have to move forward to Email 2.0.”
So how do you do that? It’s all about trust or, more accurately, restoring trust. Thanks to worms and false declarations of love and naked photos of Anna Kornikova, we no longer trust the information that comes into our inboxes.
And yet at the same time, 97 percent of households use email; 54 percent of them believe it is more useful that snail mail; 52 percent more useful than the phone; 41 percent think it is a better way to receive bills and statements; and 64 percent prefer it as the way for companies to communicate with them.
Technically, we have no good way of checking email. It comes with no guarantees whatsoever, explains Jim Galvin of Afilias. We just assume that it came from who it says it did because it says it did.
The problem, adds Dave Crocker of Brandenburg InternetWorking, is that we have trained ourselves to look for bad behavior, and so we have ended up being ineffective at looking for good behavior. “The trust side of Internet world is not just flip-side of abuse,” Crocker argues. “They are two different things.”
Despite the fact that spam filters these days are 98 percent effective, there is so much of it that the two percent continues to cause huge problems. And the status quo can’t go on, the panelists warned: the filtering systems work on IP addresses, and as the Internet moves to the next-generation IPv6 network, those addresses will become far too numerous to be effective any longer.
So what is the technical solution to all of this? A whole bunch of acronyms, of course. DKIM, ADSP, SPF, and DNSSEC. What they amount to is a way to “trust” emails using domain names, rather than IP addresses.
DKIM adds a signature that validates that an email that appears to come from example.com actually comes from example.com.
ADSP allows you, as the domain owner, to tell others how exactly you are signing your emails. Do you sign all your emails, or just some? Are you so rigid with signing that if you receive one without a signature you should just delete it?
And so on.
Once you have added all these acronyms to your email system, it becomes a matter of “domain reputation.” If you send lots of emails over time, and none of them contain worms or virus or naked pictures, then people can start to trust emails from that domain.
Of course, DKIM is far from a brand new technology. It’s been around for several years, but as Jim Fenton of Cisco explains, "we’re currently in a chicken-and-egg situation.” Fenton believes, however, that the trust technologies are rapidly reaching a tipping point and will start to enter the mainstream.
And as for the currently popular view that Facebook and Twitter mean an end to old-fashioned email, the engineers beg to differ. “Yes, using a walled garden as a messaging medium solves a lot of the problems,” explains Dave Crocker. “The problem is that as the walled garden gets bigger, the problems that show up in email show up there as well. If people wish to communicate very widely, we really need technologies like DKIM.” ®