> This even allows them to have fully-Unicode usernames; username phishing is less of a problem when users expect duplicate usernames, and none of your systems

> This even allows them to have fully-Unicode usernames; username phishing is less of a problem when users expect duplicate usernames, and none of your systems depend on username uniqueness. It's surprising but providing an extra degree of freedom makes the system more robust. Username phishing is a real problem on Twitter because people expect unique names associated with each person they interact with but if usernames can not be assumed to be unique then using the name as a heuristic for identity is no longer a viable shortcut so people have to develop other ways of making sure they're talking to who they think they're talking to. On a related note, keybase (keybase.io) proofs never made sense to me until I started thinking about how I would prove to people that I am indeed who I say I am. Keybase provides a cryptographic basis for trust, which is much better than what most social media systems currently support with their verification mechanisms. I personally trust cryptographic signatures over whatever verification mechanism Twitter is using to provide blue check marks to verified accounts.