Over on the always excellent hueniverse, Dirk Balfanz describes a workable way to use email addresses as OpenIDs.

This has been something of a religious war over the last few years, and I come down firmly on the side that says that email addresses win because they are good enough and people already know how to use them. Greenberg’s Rule #2: Convenience beats quality every time.

I spent the $12 to get an iName but I’m a big, dumb, nerd (=steven.greenberg surprisingly enough) and love this stuff. I was never under any illusion that my friends were going to start using them. There is simply no compelling feature that will make people switch.

The fact that OpenID URLs, iNames, or your pet technology of choice is “better” for some software reason isn’t even remotely relevant. Switching has a cost, and users aren’t going to learn something new unless you give them an extremely compelling reason. If users don’t know how to use these IDs, sites won’t support them. InfoCard has a chance — but only a chance — because it’s going to be baked into Windows.

Email addresses are far from perfect, but they get the job done and your users already have one. The path to identity portability might end with InfoCard or something similar but, like everything else, it begins with an email address.