I think Gizmodo's is a much less rigorous article than the Wired one. It mostly consists of accusations and circumstantial evidence. Every early player in Bitcoin development has these in spades. In a world where Craig Steven Wright is not Satoshi, we'd see the exact same tenuous evidence.

In contrast, the Wired article stays close to the facts (GPG keys, email addresses, IPs, etc) and hedges its bets. Wright might be Satoshi. He might be a hoaxer. We don't know. It's not as satisfying to read as the Gizmodo pablum, but it's much more accurate.

I have looked at the GPG keys, and based on the hash algorithm preference order, I believe they were created after [edit: July 2009], by a gnupg version following this[1] commit, on a computer with its clock set to 2008.

1. http://git.gnupg.org/cgi-bin/gitweb.cgi?p=gnupg.git;a=commit... [edited]

If this is true, then it's pretty substantial evidence that this guy is a hoaxer.

But can you really tell when a key-pair was generated? This would make for a fascinating write-up if you could please spend the time to explain.

Which hash did he use?

The change you link to has the same SHA-256 as the first option both before and after the change.

Did he use something else?

I'm curious to know your thoughts on the scenario whereby another hash would be chosen for him by the software, and whether or not he may have chosen a specific hash?

This is potentially a good line of enquiry with more info