This is very interesting. Funny how difficult it is for a decentralized system to ensure it remains decentralized throughout the whole vertical of the system.
Don't confuse control of the bitcoin core github repo, or even control of Bitcoin Core as controlling the bitcoin protocol.
The code running on my node is all that matters. If they release something I do not agree with, I can reject it and continue running the existing software on my node, or run some other implementation (such as Bitcoin Knots, which is a fork of Bitcoin Core but currently supports is compatible by following the same protocol rules). That is why Bitcoin Core development is so careful to ensure they only make changes welcomed by the near unanimous consent from the "economic nodes".
reply
This brings up another question: If enough node operators and miners use prebuilt OSes where they don't verify the signatures of the bitcoin core software included, could it be possible for the developers of those OSes to inject code and then steer Bitcoin in a different direction? I guess this would be a sort of 51% attack. Although I guess depending on the code injected, it would create a discrepancy between some operators and others which would eventually raise enough eyebrows to cause further investigation into the matter.
reply
Luke Dashjr did exactly this with Gentoo, adding address blacklisting to bitcoind.
reply
Whaa? Do you have any links or resources that goes into more detail on this? I'm curious.
reply
This was quite a while ago and I'm going off memory, but I found this - https://bugs.gentoo.org/531634
He did so with the intention of blacklisting certain "spammers" (his words) which included SatoshiDICE which at the time was extremely popular. Very slippery slope imo.
reply
thank you for digging this up! This part of Bitcoin's history is close to being forgotten!
reply