The Economics of OpenID


I spoke at the Identity Solutions Symposium on the topic of Social and Economic Aspects of Identity (PDF of slides). This is a difficult topic because there is so much to say and so many issues that you could cover.

One of the things I didn't talk about that I wish I'd had time to cover was the developing economics around user-centric identity. With announcements like OpenID and CardSpace interoperability and AOL's support for OpenID only a few weeks old, I think that we're getting very close to the identity "big bang" that Kim Cameron talks about.

If you're building a Web-based app today, you could choose to not have an authentication system and instead just use OpenID. Anyone who doesn't have an AOL account could easily sign up at AOL or one of the other providers. You'd still have to associate the identifier with the properties you care about, but you wouldn't have to handle authentication at all. Big win--the authentication stuff is a necessary evil--something you'd just as soon not do if you could.

What's missing at this point is user experience. Most people building a Web site for business purposes wouldn't want the cognitive disconnect that happens when a user is redirected to login. Anyone who's studied Web site usability would tell you you're going to lose people with that step. Even more so if they have to sign up for an account somewhere else.

There's no reason that the OpenID user experience couldn't be improved to solve this problem, or at least mitigate it. You have to balance usability with concerns for security, but I think that even within those bounds, we could make real progress.