whether our technological level has reached the point where we can
do it now without sacrificing current security. Obviously, 10 years from now, a 2-4
MB blocksize hardfork is extremely doable and should preserve the current security
As someone who "supports" segwit on it's merits, and a 2Mb block increase for compromise, I have yet to see any compelling argument for this. Can someone point me to something I can read and digest that would actually accomplish that plus something that refutes that. I'm not talking about things that are filled with propaganda and ideology but solid arguments based on technical issues and facts etc.
But as "guardians of the protocol", their power is zilch, and hence as "decentralizing power element" their influence is zilch.
I've read a lot of your posts on this and you've failed to convince me except in the case where 100% of the miners are in agreement and in opposition to a very large majority of those running full nodes.
If, for example, 75% (hash) of the miners decide to change something radical about the protocol, the full nodes can choose to reject their blocks while accepting those from 25% of the miners. Yes, the entire system is disrupted for some period of time but it was going to be anyway.
I agree with you, but in that case, it are STILL NOT the full nodes that decide one way or another, but the users in the market *with their money*, not with their nodes. If the users sell in majority the coins on the 75% chain, and buy the coins on the 25% chain (remember that from the start, all users have equal amounts of both), what will happen ?
A) the exchanges listing both coins will make a lot of money on the transactions (so, yes, they will list both coins).
B) users will NEED TO TRANSACT on both chains. So if all nodes block one chain, they will find a way to connect directly to the miner nodes, and exchanges (see A) will do so too.
C) in the end, coins on the user-preferred chain will be worth much more than those on the user-rejected chain (offer and demand in the exchange happening in A).
D) miners will adapt their hash rate to the most profitable chain, until this finds an equilibrium: it will find an equilibrium when hash rates agree with market price. As long as an expensive chain has little hash rate, it becomes interesting for miners to start mining that chain, the forkers on the 25% chain will make a lot of money ; a cheap chain with a lot of hash rate (and difficulty) will not be advantageous to mine, the 75% miners will be mining at loss. So
miners will follow market price.That's exactly what happened with the ETH/ETC split (ok, the miners didn't initially go in majority to the small chain, true)
And whatever was the ratio of full nodes doesn't matter in this. Users voting in the market determined all.Now, in bitcoin, there's an extra difficulty, which is the very slow difficulty adjustment rule. Unless that is adapted in the fork too (could be), any split will bring initially the "low miner count" chain in difficulties, because they will only be able to mine a few blocks. But also the majority chain will slow down. This is what keeps unity between miners, and which wasn't the case on ethereum.