News for hackers. Stuff that matters.
Re:Empirical explanation for Craig Wright's selfish mining B... by barbegal on Thursday April 12, 2018 @01:01PM (#16822834)
Craig Wright would be correct if not for the information that the dishonest miners found the "next" block. If anyone could have found the next block and the dishonest miners simply found a block at t=0 then the expected time for the honest miners to find a block would be at t=5 minutes.
Re:Empirical explanation for Craig Wright's selfish mining B... by __blockcipher__ on Thursday April 12, 2018 @01:41PM (#16823178)
Nope, this is totally wrong.

To put it bluntly, please stop spreading this garbage. CSW completely misunderstood memorylessness (which, just to state the obvious, is solid bayesian evidence that he is not Satoshi).

Regardless of the knowledge of the participants, it will always take an EV of 10 minutes when alpha = 1, or an EV = 15 given alpha = 2/3 in this example.

Re:Empirical explanation for Craig Wright's selfish mining B... by barbegal on Thursday April 12, 2018 @01:45PM (#16823204)
I completely agree but in this formulation of the problem the honest miners started at t=-10 so may have found the block before the dishonest miners in the timeframe from t=-10 to t=0.
Re:Empirical explanation for Craig Wright's selfish mining B... by __blockcipher__ on Friday April 13, 2018 @07:09PM (#16834711)
They did not, that is (perhaps implicitly) part of the problem description. It never says the honest miner finds block N, it asks when they are expected to implying it hadn't happen up to that point. Also, the dishonest miner found the "next block" but did not broadcast it at t=0, which is essentially irrelevant information because it doesn't matter whether the honest miners know about the hidden block or not, either way it will take them an EV of 15 minutes at any given point in time regardless of knowledge
Re:Empirical explanation for Craig Wright's selfish mining B... by madavidj on Thursday April 12, 2018 @01:36PM (#16823140)
That is still incorrect. Assuming a poisson process, the expected time to find the next block from T+0 is T+15 no matter what information was given or not given.

edit: and assuming 2/3 hashing power