So $200 shoe purchases or general mall shopping is still up in the air?
$200 is still no problem for LN, $400-500 may be the limit I'd be comfortable with, or let's say channels with capacity up to $1000. Currently the fees are low enough to do these transactions onchain.
Ideally (for Bitcoin's current security model) the transaction fee would be close to 2-3 cents per vByte (currently about 30-50 sat/vByte), i.e. about $2.50-$10 per transaction (to today's USD purchasing power). This fee level would allow both cheap LN channel creation but also ensure a significant miner income based on transaction fees ($50-100k if block space is fully used). We don't need $100 fees for this.
I'm just wondering if our current PoW mechanism is as good as it gets or if there might be a better one that unlocks additional scalability without introducing MEV or forcing us into one direction of scaling.
The "PoW mechanism" itself isn't the problem, as it's not directly related to the "zero-emission" policy from 2140 on.
One could think about a PoW/PoS combination, like Vitalik's original "Slasher" or the ones very common in the altcoin world. But regarding their theoretical security it's not totally clear if these mechanisms are an improvement; for small altcoins however "practically" they seem to be the "salvation" from 51% attacks. But a trillion dollar network is another story. The good thing is that we have a "rabbit" to conduct PoS experiments, Ethereum, which is large enough to be attractive for hackers to discover PoS vulnerabilities
For example; has anyone considered subnetting? If blockchain is a "protocol" and engineered to be the TCP/IP of value... Then perhaps we should spend some more engineering cycles on the underlying protocol instead of taking a game development approach, bolting on features that IMHO creates more problems than they solve.
I don't understand what you mean here. L2s like Lightning, Ark or sidechains are for me exactly what "subnetting" in the case of TCP/IP would be.
It is not part of the core protocol, LN uses core features like HTLCs but they were not implemented with LN in mind. LN itself doesn't create problems at Core level, it may need to solve challenges at "subnet level" only. The same is true for sidechains.
The "fragmentation" can partly be hidden from the user. For example, there were ideas in 2017 to let the Bitcoin Core client decide if a transaction uses LN or on-chain, with the user not having to intervene. I don't think I would use this feature but for people not wanting to bother with technical details such a "hidden L2" may be attractive.
I've found a paper the other day explaining Proof of Entropy Minima that seems to have a lot of potential,
Well
that description at a first glance looks like pure technobabble, like we're accustomed to in the altcoin world. It doesn't explain how this solves the double spend problem. But yes, for a detailed discussion about this I invite you to create a thread in the altcoin forum.
