On 2015-08-31 23:32, Peter R wrote:
On 2015-08-31, at 2:24 PM, Allen Piscitello via bitcoin-dev <bitcoin-dev@lists.linuxfoundation.org> wrote:
It is my opinion, then, that we should support multiple implementations of the Bitcoin protocol, working to reduce the network's dependency on Core.
That would be incredibly foolish given the history, where even heroic attempts at making consensus compatible re-implementations have ended rather poorly. bitcoin-ruby and btcd have collectively had numerous consensus failures, some only recently found by fuzzing the scripting environment. There are more failures than publicly disclosed, and almost any failure can be leveraged to split the network and steal money. Ethereum attempted to create four clients, written to a defined specification, and even with all the money in the world has managed to have numerous consensus failures due to misunderstanding or implementation.
I agree. What about decentralization in development? Gavin recently said that he wants to "get to the point where there will be multiple robust implementations of the core protocol."
Gavin clearly hasn't kept up with the ridiculousness of that task. _______________________________________________ bitcoin-dev mailing list bitcoin-dev@lists.linuxfoundation.org https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
