Paul uses the analogy of marriage and courting to say that my comment about the new OpenID extension is essentially trying to shortcut a well known process. While I may have snuck the ring on my wife's finger while she wasn't looking, we did go through the normal courting process.
However, I would say the analogy doesn't apply here as the OpenID AQE effort is just getting started while Liberty's ID-FF was done in 2001 and SAML 2.0 was done in 2005. Not looking at those specs at this time is more like saying that my wife and I went back to the courting state when we decided to have our second child and had to wait to be married yet again before we could try. Luckily that wasn't the case as I'm not sure I would have made it through the process again (she had learned too much about me by then).
Again, I think that OpenID, or any other group working on some concept, should look at what's available and see how it can be profiled and/or adapted to the work at hand. The people involved in SAML are very open to helping figure out how their work can be adapted to meet the needs of any group, even to the point of working on new profiles (as evidenced by Jeff Hodges' & Scott Cantor's Simple SignOn profile).