Thursday, November 8, 2018

On forks, Nakamoto Consensus and hash wars

So basically I wanted to dump some things on y'all that I've been thinking about.

First of all, I assume most of us agree that hashrate follows price in a stable environment. We've seen that quite clearly with BTC vs BCH.

That said, during these highly contentious periods where forks are possible, people talk about hashrate as a power in and of itself. The reason being that it doesn't matter how many users want to fork off, if there is no hashrate to back them, they can't do a proper fork (unless PoW algo is changed). Because it'd be trivial for 99% hashrate to smother the 1% (produced by thousands of full nodes mining on their GPU) to the point where the minority chain becomes unusable.

However, there's a distinct difference between having a majority hashrate and having a majority hashrate that is hostile towards the minority. Short term, it's beneficial not to attack because attacks cost money (wasted energy mining empty blocks). It's only when miners perceive a chain as a long term threat, they might consider attacking it.

Basically, large majority hashrate can force the ecosystem to stay together* if it's in their interest. Naturally, this relies on them expecting users aren't totally unhappy with the rules of the majority hash rate. Ie. they'd rather stay on the majority chain that fork to a new PoW algo.

So basically, as long as the demands of the supermajority of hashrate aren't unreasonable, they can prevent the ecosystem from splintering.

When it comes to SV vs ABC, that's how it see it. There's little to be done if SV magically manages to have enough hashrate to propel their own chain and kill the ABC one.

But if we apparently must simply obey hashrate (some people have been calling this "Nakamoto Consensus"), why are we on BCH and not BTC? BTC has 10x more hashrate than BCH and could easily crush us, right?Well, I actually already posted the answer:

There's a distinct difference between having a majority hashrate and having a majority hashrate that is hostile towards the minority.

My theory is that the percentage of BTC miners who are so hostile to BCH that they'd burn short term gains in order to get rid of it, aren't that big. For one, we have one of the biggest miners, Bitmain, openly demonstrating they ideologically back Bitcoin Cash (by acquiring a shitload of it).

So despite the majority of Bitmain's hash is on BTC, they don't follow the BTC vision.

It's not about where the hashrate currently is, it's about where it would be in the event of a hash war.

So let's say BTC hashrate looks like this:

- 33% Hostile towards BCH

- 42% Neutral (or not willing to waste money for attacking)

- 25% Pro BCH

The difference between 33% and 25% is only 8%, which (when translated to BCH figures) is not enough to be a majority. So if both attacking and defending factions would come and fight in the "glorious BTC v BCH hash wars", nothing would really happen.

Well, apart from BTC losing over half its hashrate**

Now I have no clue what these percentages really look like. But I doubt BTC miners who are hostile toward BCH know either. Which means an attack would be insanely risky for them and the BTC chain as a whole.

So back to Nakamoto Consensus. If that refers to "whoever has more hash decides", then BCH being alive doesn't invalidate that because BCH is actually backed by a significant portion of hash.

The fact this hash isn't actually mining on BCH right now doesn't matter that much.

So anyway, that's my thoughts on the matter. Let me hear what you think!

*of course a PoW algo change is always a possibility to get rid of Hashrate attacks, but it basically removes your entire mining inftrastructure. It's a huuuge trade-off that few people would consider -- especially in the Bitcoin space.

**which is a huge problem for BTC if their blocks become twice as slow because that effectively halves their TX throughput due to the blocksize limit. On BCH that'd not be the case.


No comments:

Post a Comment