Defending against SSH brute force attacks

I think what the author fails to realize is that this isn't a targetted attack by a single group, but rather a mass of unfocused attempts to catch low-hanging fruit anywhere on the internet they can find it. Not to say that the hardening efforts aren't worth it, but that there's a difference between a bunch of automated attacks against "common" username and password combinations and a focused brute-force attack.

I just disable password-auth, install fail2ban, and call it a day.

I have found the suite of "projects" from R-fx Networks to be fairly effective and easy to deploy:

http://www.rfxn.com/projects/

How can this stuff get voted to the top every week?

100 different ways to write "Install a firewall and use fail2ban/denyhosts. Oh and make sure you sshd config is sane."