[hyko]

The consensus protocol doesn’t help in removing either of these trust requirements?

If I own the mechanism of root node discovery, I can mount an effective Sybil attack against a new network participant.

The history of Bitcoin is full of hard and soft forks and changes to the rules. That’s why your original Bitcoin client wouldn’t even let you join the Bitcoin network now. The people in control of the network are in a position of immense power with minimal oversight. So on what basis should I trust them with my life savings? What recourse do I have if they ruin my life?

[xur17]

> That’s why your original Bitcoin client wouldn’t even let you join the Bitcoin network now.

You can run whatever client you want.

> If I own the mechanism of root node discovery, I can mount an effective Sybil attack against a new network participant.

But you wouldn't be able to replicate a chain of blocks with the same order of magnitude of work behind them. And then, even if you could, all you'd be able to do is change history, with the depth of that history dependent upon how much mining power you had available. You wouldn't be able to spend other people's money or give yourself more.

[hyko]

You can run whatever client you want...if it conforms to the design of the protocol.

you wouldn’t be able to replicate a chain of blocks with the same order of magnitude of work behind them...I grant you that, but does it matter? If the client can’t find the real network, how would they know what to expect? You and I would know, because we have a copy of (I hope) the real Bitcoin blockchain...but how do you acquire that knowledge without trusting an entity first?

all you’d be able to do is change history...wouldn’t that mean you could lay claim to whatever you wanted in the eyes of this poor client? Or just harness their hashing power.