BitAlias I
Yanislav Georgiev Malahov

One important thing to keep in mind: a lot of developers implementing the protocol will, when an unknown alias is seen, traverse the chain in anti-chronological order until they’ve found it. That seems like a reasonable thing to do.
However, that is not what should be done. Because the protocol cannot prevent people from registering usernames that are already taken, it is very important to start searching the block chain at the point of one year ago and go in chronological order, ignoring claims of aliases that are already claimed.

Another thing: why two separate transactions? The rationale for this is not elaborated on, can you please do so?

Lastly, I assume that when two claims for the same alias are used in the same block, that the tx order will make out who gets it?
This kind of situations will definitely occur in the case a popular alias expires.

Looks like a promising idea, though.