The dev team is necessarily centralised. When centralised is the only ideology that works, I'm in favour.
Pick your centralising influence: amorphous Core team or Mike Hearn's self described "benevolent dictatorship". Benevolent in respect of who, exactly?
Pick your centralising influence: amorphous Core team or Mike Hearn's self described "benevolent dictatorship". Benevolent in respect of who, exactly?
Can you explain why you believe a model with several competing implementations of the protocol wouldn't work? In fact, with the emergence of BitcoinXT it looks like it is beginning to work. The consensus critical code would of course need to be compatible between implementations; however, if a change was desired by the community (e.g., increasing the block size limit), each implementation could attempt to solve it "their own way" and then the community would decide the winner by switching to the implementation they favoured. Then, to retain some portion of their previous user base, the losing implementations would adopt the same change to the consensus code to prevent their clients from forking off from the longest proof-of-work chain when the change goes into effect (at some future time similar to BIP101).
Is this a way to give the community more power in exercising their free choice?
Of course it is. Competing implementations is the only way to make it a market choice. In some years we will wonder why this was ever not obvious.
I agree. In fact, I see this as the Bitcoin Consensus mechanism playing out exactly like it should. The Bitcoin white paper says that nodes express their acceptance of a block by building on top of it; the longest chain becomes the correct chain. Giving people the choice between competing implementations just makes it easier for them to express their acceptance of, for example, blocks larger than 1 MB.
In hindsight, it will be seen as ironic that so many fought for centralized development in order to keep Bitcoin decentralized.
