Widespread usage of self-signed certificates might have the unfortunate effect of training people to click "allow invalid certificate" when they shouldn't (such as what happened with Vista's UAC).

I would also argue that unless your users are technically inclined, a self-signed certificate is as good as (or worse, due to the scary browser warnings) as an unencrypted connection.

It would be workable if browsers stopped loudly complaining about self-signed certificates and gracefully fell back to using something like TOFU/POP [1] or Convergence [2].

[1] http://www.xmlgrrl.com/blog/2010/07/06/tofu-online-trust-and... [2] http://convergence.io/

Ah, true. I wonder if it wouldn't be best if, by default, browsers would only complain when a site switched to an untrusted cert it hadn't used before (i.e. if Facebook.com is suddenly self-signed, that's a problem, and if a site switches from one self-signed cert to another, that's a problem as well). This would prevent most MITMs while still making self-signed certs useable. A site with a self-signed cert is better than one using no security at all, so it shouldn't get more flack.