Paul Madsen points out this Chris Messina post about a study of OpenID usage and awareness among Mechanical Turk users. Paul makes some interesting distinctions about SAML being envisioned to be invisible to the end-user while OpenID was invisioned to be a “branded”.
Personally I believe that OpenID adoption will happen en-mass not when it is branded by OpenID, but co-branded primarily by a small set of large identity providers. A lot more people are “aware” they have a Yahoo account or LiveID account than an OpenID ID.
But what I find absurd is all of the statistical entrails reading that is happening to determine what the OpenID adoption rates are when a couple of large identity providers could simply just tell us. Why don’t the big OpenID identity providers simply publish OpenID authentication stats on a monthly basis?
The providers must have those stats internally. The fact that they are not published says a lot more than any studies about brand awareness.