True, this post contains speculation on some matters, but if a professor at Cornell whose work focuses around distributed systems who authored the selfish mining attack isn't an "expert," then I don't know who is.
It was all speculation, and it is inconsistent in its inputs -- on the one hand it disbelieves what MtGox claims, but on the other it takes historic claims of security (that keys were in cold storage) as proof of other conclusions.
Speculation that is based upon cherry picked belief and disbelief might yield the conclusions that one wants, but it adds little to the information pool.
And regarding claims that keys were in cold storage: Maybe, but one thing I think we're all learning from these amateur hour platforms is that a lot of claims aren't backed by reality. They have "Top men" working on their security. "Top men".
...
> We believe that there is a high probability that these bitcoins were stolen as a result of an abuse of this bug
I find the insistence that this catastrophe was brought on by a protocol bug very suspicious. A recent post on HN by an expert elaborated on why transaction malleability wasn't likely to have been the source of this exploit [1]. Meanwhile Mt. Gox is insisting that they were robbed by a bug in the protocol.
This could be the position advised by Mt. Gox's legal representation, in a sort of "that's my story and I'm sticking to it" sort of way. If this line is actually true, then either the transaction malleability issue is much much bigger than we anticipated, or the bitcoin protocol has some vulnerability that we don't know about.
That's scarier than a paltry exchange collapse.
[1] http://hackingdistributed.com/2014/03/01/what-did-not-happen...