I'm torn on the idea of "everything on Bitcoin"...

On the one hand, consolidating to the hardest money makes sense. 

But... everything becomes *dependent* on Bitcoin and could be overturned and abandoned with a hard fork. So if, say, there is a Bit-Monero drivechain, and a state power influences enough miners to hard fork OR just not merge-mine the Bit-Monero, the users of that chain are screwed. I don't see anyone from "real" Monero moving to Bit-Monero at all. 

And Ethereum-- fuggetaboutit. No one is leaving Ethereum for Bit-Ethereum.

So maybe all this does is sound the number of shitcoins in existence.

Then again, it might be worth it if just one if those chains would allow billions of people to onboard to Bitcoin quickly.

There is also this implication that "bitcoin isn't good enough" as it is, which maxis don't appreciate. There seems to be a loud group of ossifists saying it's fine if Bitcoin is only and always a store of value (or "property" as Saylor is now saying) and not currency.
 

 What is going to happen will happen, but I don't think we should put all our eggs in one basket. More options is better. Let the crypto ecosystem be diverse and flourish. The market will sort out the gems in the long run. 

 I'm trying to wrap my head around the fungibility. A perfect peg means the value on the side chain is exactly 1-to-1.  But there is *less security* on any side chain, because there is a non-zero chance of drivechain being "undone". So people value side chain BTC *less* than they do main chain BTC.

Of course the purpose of the drivechain is to enable more functionality on side chains. So the decrease in security may come with a benefit of privacy, functionality, or scalability.

 

 Paul makes an interesting assertion which really makes me question my metrics for evaluating layer 2 proposals.

Now I had been assuming that one of the most important characteristics of a layer 2 is unilateral exit to layer 1.

But Paul argues that if it's impractical or impossible for everyone on the planet to hold a utxo, then there is no need for them to exit the layer 2 independently.

This blows my mind a bit because, yeah, if people have a UTXO of poverty-level sats, then the fees in this "hyperbitcoinized world" will make the UTXOs into dust on the main chain, even if they hold value on layer 2.

But maybe they can exit to lightning or ecash somehow. 

 

 https://achimwarner.medium.com/thoughts-on-drivechain-i-miners-can-do-things-about-which-we-will-argue-whether-it-is-actually-a5c3c022dbd2 

 "Any future FAANG type corporation or bank is going to be making orders of magnitude more revenue than all Bitcoin miners who are working on tiny margins. So if Mark Zuckerberg get his sidechain and then wants to make sure that Elon never gets a sidechain, all he has to do is take a fraction of his budget and go buy out a small mining company, enough to get a 10–12% veto share. Now anybody who wants a sidechain has to make their case that the sidechain is going to be beneficial to Mark Zuckerberg, who can reject it at will, or demand concessions.

Miners have never had this sort of voting power before, but Drivechain will give them explicit power as gatekeepers."

This is scary AF 

 Paul asked @Shinobi why the miners cannot currently conspire with Amazon (or Mark Zuckerberg or whoever), and Shinobi has not answered this question. I also asked Shinobi multiple times.

Amazon can bribe miners right now, and it gets easier to do so as miners lose revenue every four years. If we want to stop them from conspiring, we have to incentivize them. 

 They are incentivized with higher fees. 

So what can a billionaire do to the Bitcoin network? Buy half the mining power. Can they censor a transaction? No. The power of the network is that only a single miner needs to create a valid block for a transaction to not be censored. Can they reverse a transaction? No. Can they "spam"  the network with 1 sat dickbutts? Sure. So the other miners who instead build blocks with high-fee transactions will see their revenue skyrocket as competition for block space increases. This incentivizes new miners to enter the space and capture those high fees. Meanwhile the attacker is wasting money AND increasing revenue to "normal" miners. 

With Drivechains, miners are not required to merge mine, and they aren't obligated to any particular chain. They can stop mining or a competitor could come in and immediately take over 90 percent of the hashing. It's possible that the incentives will be for each mining pool to stick to one drivechain, allowing each have a virtual monopoly. Basically, the point of failure has become smaller and cheaper. No single government regulates *all* Bitcoin mining, but it would be trivial to manipulate a single pool. 

I think the broad idea is that because drivechains look so kind-of nebulous and ephemeral in how they "attach" and"detach" from various miners at whim, people wouldn't be afraid of being rugged so much as the whole drivechain just disappearing or potentially being dominated by a censorship-friendly pool. The end result is no one wants to use it FOR MONEY, so all drivechains and up getting used for is scams and spam. 

I think miner centralization is a very serious issue, but rewarding them in this way doesn't stop centralization, and doesn't scale monetary use of Bitcoin. 
 

 I've spent a couple days reading about drive chain, including reading the bips and listening to about 6 hours of Paul talking about it.

I'm always open to changing my position, and I have done so, from being optimistic to being very cautious. 

Drive chain side chains could be at best a way to enable innovation directly on Bitcoin, rather than independent chains or on Ethereum.

However, I think other proposals allow this in a safer manner.

The neutral case is nobody cares about these side chains because they are subject to the whims, vulnerabilities, and incentives of Bitcoin miners.

The worst case is far from the benign outcome Paul portrays.

But since my interest is really in *scaling Bitcoin as money*, my final judgement is simply that drive chains *cannot* be a solution to scale.

I am grateful to Paul for his work, and am strongly considering his proposition that unilateral exit to layer 1 is not a necessary characteristic of a layer 2.  However, if that is the case, then the layer 2 must provide *strong* censorship resistance and privacy.