This is a pretty shaky analysis:<p><i>Indeed, the biggest problems with their attack are social more than technological</i><p>Try making that argument with, say, an attack on PGP, or on TLS, or frankly any other cryptosystem people are not so emotionally attached to.<p><i>the honest nodes can always pick the ticket announced by the honest miner over the evil miner's ticket</i><p>Much like Internet users can <i>always</i> remove the root certificates of untrustworthy CAs. Again, apply this argument to other cryptosystems and see how far it gets you.<p><i>the Cornell attack would be easy for honest miners to detect</i><p>Maybe so, but then what? It is also easy to detect that a CA is giving out certificates for MITM attacks, but that does not even come remotely close to solving the problem. Suppose that the honest miners do what the author suggests and publicize their lists of trusted miners -- then we would have a block chain fork as people simple failed to ignore the supposedly dishonest miners, or worse, if the dishonest miners also published such lists (how would the average user distinguish this?). Why should I believe one mining pool's account of who is honest or dishonest over another, anyway?
I've been a fan of Bitcoin for a while but this latest brouhaha got me to start thinking seriously about how to help out the ecosystem. Think what I've started designing and building is promising, and doesn't really exist yet. But is badly needed.