On Tue, 2009-04-21 at 21:29 +0100, John5342 wrote: > > We just need a standard way of discovering what OpenID providers the > > user is using, instead of hardwiring Google/Yahoo/whatever. This is > > probably best accomplished by interaction with the browser, which will > > need an account manager... (which most already have a "remember > > passwords" system already...) > > Maybe i have completely misunderstood how openid works but from what i > gather the whole purpose of using a url as the id is that the url is > contacted to do the authentication. Therefore we already have a way of > discovering what openid provider the user is using (its embedded in > the way it works) and that means no hardwiring needed. And that's the problem. A user should not have to remember or type or cut and paste anything. The OpenID account is configured in the browser, and the browser handles the underlying URL bullcrap the user doesn't want to see. They simply go to a web site, see a "Log in with <the user's preferred OpenID provider>", click it and they're done. As usual, Open Source produces exceptional architecture but completely fails at user interface. I guess that's what happens when you let Brad Fitzpatrick design it.
Attachment:
signature.asc
Description: This is a digitally signed message part
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
