But only after enough time had passed that smooth and friends could have easily minted a small fortune off of the backs of their users.
In fact that is literally impossible because above blog post documents exactly who was doing the bulk of the mining (spending $250K on AWS, presumably at a profit) and it wasn't us. In fact I think he says in there that he had no connection, financial or otherwise, with the Monero team. It is one of the few cases where who exactly was doing most of the mining is out in the open, mostly a serendipitous occurrence since he happens to be a CS professor who likes to blog about his exploits. Most miners wouldn't.
He claims to at times to have his 60% of total network hashrate, and from what I can tell this was pretty far after the launch. Either way, someone figured out your trick and had to use a
significant amount of AWS just to keep up with you. How many instances were you and your friends using? Were you using the scam miner you pushed to the public in your ANN thread? Are we just supposed to believe you?
I was not mining on AWS. I mined some on my own hardware, using the code from github. I never got any optimized miners from NoodleDoodle or anyone else (in part because his test versions only ran on Windows, which I don't use at all, so I couldn't help him test). As for the 40%, there were many, many people publicly posting about mining at that time, and in addition publicly selling off some or all of their mined coins. I recall vaguely the names of some of the bigger public miners, but I would have to look back on the threads and find them. They were not core team members.
So a large portion of the hash rate is very transparently accounted for. Furthermore the fact that mining with the standard mining code was still profitable (even on AWS I recall) essentially guarantees that optimized miners were not in very widespread use, because of how difficulty adjustment works and mining economics.