[lulin]

I love OpenID and use it as much as possible. The only problem I have with it is the URL-as-username approach it takes. When a site asks me for the URL, I don't use OpenID as I always forget it. If the site asks me to "log in with Google using OpenID" or something similar I will use it. I don't see how people say that OpenID is a solution in search of a problem: I DO have the problem that I don't want to create a new account for every site I use. The problem is there, and some uses of OpenID really do solve it.

[AndrewDucker]

The reason for the URL as username is that OpenID originated on Livejournal, where users have their own URL (i.e. mine is andrewducker.livejournal.com)

It therefore made sense to use URL endpoints as identifiers, as you could bounce people to their authorising server incredibly easily. Doing it via email address would be much harder (where would my email, [e-mail], be authorised by?).

It's caught on amongst people who have URLs (bloggers, journallers, etc.. It hasn't caught on amongst people who don't (everyone else).

[joh6nn]

seconded: login proliferation is a real problem that i face all the time. i've been on the web for very nearly 15 years at this point, and there's no numbering the accounts i've created in that time, let alone all the ones i actually still use. OpenID isn't be perfect, but most of the really egregious issues i've seen with it have revolved around sites not being willing to commit to it.