@SysMan and core dev are of the mind that constant tinkering and optimization is needed for Bitcoin. i disagree. imo, in the >5y since i've been in Bitcoin, code is merely to enforce the ideology of the coder.
Can't agree more with that.
Apart from the Blockstream takeover attempt, imho we have a huge problem with the way some CS people look at Bitcoin:
- Some are seeing Bitcoin as 'incomplete' and see it as their duty to finish it.
- Some are seeing Bitcoin as their personal toy to play with to get their CS degree / to write a Phd thesis.
- Some are still confused why Bitcoin is working (cough cough, Greg Maxwell) and are trying to create the "better" Bitcoin which doesn't need to use all these "unpredictable probabilistics".
- Some CS'ers are afraid of forever growing data (the blockchain) and they see it as their duty to find a system that is more elegant.
- They see it as a merely CS project.
Bitcoin is so great b/c it is relatively 'simple'. It is fucking elegant and ingenious (excuse my French) in it's simplicity and the straightforwardness. And it's definitely not an only CS problem.
Of course these people don't like on chain scaling. It is too easy. We need something more complex, and while we are at it, we need to introduce 27 layers of abstraction into the code.
Combine these ways of thinking with the feeling of "you're one out of a million" who can work on Bitcoins code and you get the arrogant idiots we now have in charge of bitcoin.
If Core forks to 2 MB and doesn't get stopped we are as doomed as we are without 2 MB.
We will still get unnecessary complicated hacks and idiot miners accepting them. If this happens I will still support a PoW fork. Kick the idiots out of the cockpit. I don't see where this (edit: keeping Core) could be an option at all.
Gavin's KISS is exactly the right approach for Bitcoin. To be honest, I don't know an engineering application where that shouldn't be your first priority.
I don't want to see Adam Back, Greg Maxwell, Luke Jr. and Peter Todd near any relevant Bitcoin implementation. Never.