I hope you meant that, Eric, because I’m about to quote you! In a ZDnet post titled "A tipping point?" he comments on Technorati’s adoption of OpenID, and almost casually, he says:
Note: "internet identity" is now called "URL-based identity," or even more broadly and less accurately "user-centric identity".
This is quite remarkable. As far as I recall (and Google seems to recall), the very term "URL-based identity" is only about a year old. There have been and are and probably will be for a long time, many other approaches to identity. And Eric, one of the definitive opinion leaders in this industry, is now equating URL-based identity with the internet-scale identity.
I should quote more of his reasoning:
Way back in the early mists of identity time, I was speaking with Bryan Field-Elliott (then CTO of Ping Identity) about the earliest drafts of the Liberty Alliance protocols, and whether or not they could be used for what we then called “internet identity.” … Bryan told me that while SAML or Liberty *could* be used for “internet identity” (theoretically), they never would be. They never would be because web developers are their own breed — they don’t gather at hotels, “spec out” requirements, and engage architects to build an elegant solution. Instead, web developers stumble upon something that excites them, pull in disparate pieces, kludge something together, get a big guy or two to buy in, and start using it.
Bryan was, of course, right.
I got to take exception to the "kludge something together" because OpenID so far isn’t, and I’m hopefully it will stay that way, too. But describes the dynamics very accurately, and apparently, it’s working!
We’ve come a long way since that late-night joke between my wife and me about solving the identity problem by "giving everybody a URL and be done with it"…