> The author doesn't understand the primary goal. It's precisely to have global consensus that proceeds according to its logic - so if you own something it can't be taken away from you.
How is global consensus not centralisation though?
Humans are notorious for disagreeing on everything. Having one global arbiter of truth/consensus seems implausible at best.
Anything that could provide global consensus will need to be backed by something formed from a human brain. And our brains are not perfect. We keep updating code because we miss things. Who would maintain the system for global consensus? And how would they not just be our new rulers?
As a perfect example you mention ownership as a feature of global consensus. What do you own with an NFT? Who enforces that ownership? What stops someone pirating an NFT image?
Ultimately this screams of solving a people problem with a technical one. That never works. Not without strong co-operation of the people involved, and even then technology plays a very small role in the solution.
> The scalability argument is true in isolation - everyone agrees it's currently a problem, but has nothing to do with centralization or not.
It does in that a truly decentralised system inherently cannot suffer from the scaling problems cryptocurrencies have.
>How is global consensus not centralisation though?
In the same way reality itself is 'centralized'.
Global consensus is meant to add a digital layer to reality. Reality is objective by definition. What's subjective is trying to measure it - but this problem doesn't exist for digital systems.
>Who would maintain the system for global consensus?
The practical implementation has to ensure that
(1) consensus is maintained by a diverse set of potentially anonymous participants. Ideally, this would mean literally all humans, but that's of course impossible. The set should be large and diverse enough so that even powerful actors (especially governments) have no ability to force them to do something.
DPoS, PoA fail here from the start. PoW is vulnerable over medium term: it has infinite economies of scale which leads to centralization and taking it over externally is always possible (because 'stake' in PoW is external and potentially infinite - mining hardware).
(2) the most profitable course of action is for every validator to follow the protocol honestly. In ethereum's PoS that's achieved by slashing maliciously misbehaving validators.
PoW has an orders of magnitude weaker punishment here (just mining rewards for the time spent on mining a minority chain). Most DPoS systems rely on the assumption that a majority is always honest, even if it makes financial sense to not be - eg. Cardano's Ouroboros.
>Who would maintain the system for global consensus? And how would they not just be our new rulers?
In the early period updates are necessary as system evolves - but eventually it has to be frozen, with the possible exception of scaling parameters (like block size, number of shards) which could be determined by a vote from validators.
>What do you own with an NFT? Who enforces that ownership?
There's no difference to real life here. "Owning" a house actually means there exists a publicly verifiable promise from an entity capable of wielding some form pf violence to enforce that property right. Therefore, an NFT that's connected to any external asset can only a different form of such a record.
In the case of image NFTs, it's an improvement on physical art. The point of buying physical art is to either signal wealth, or to use it for money laundering. Image NFTs are infinitely superior for the first - because they are globally visible (as opposed to a physical piece of art) and fake NFTs are impossible (so you can't buy a cheap replica and pretend you spent a fortune on the original). They are much better for money laundering because they are inherently global (no transport issues) and evade identity checks in the banking system.
Exactly like physical art, if you just like it and want to look at it - you can save a copy. A replica of physical art is free in the same way - you can save a photo of it. If you want a physical replica, in both cases you can print it. An actual replica of physical art would be more expensive, but in many cases orders of magnitude cheaper than the original.
Therefore, nobody buys expensive existing art just because they like it, which leaves two reasons I mentioned.
>Ultimately this screams of solving a people problem with a technical one.
Yes. The entire history of human civilization consists of solving social problems with technology.
> In the same way reality itself is 'centralized'.
Global consensus is meant to add a digital layer to reality. Reality is objective by definition. What's subjective is trying to measure it - but this problem doesn't exist for digital systems.
This was always going to get philosophical fast, but I don't think reality is strictly objective. At least not the representation of it we humans create to understand it. Look at the constant evolution of our understanding of the universe (and those who inhabit it). If we were to describe that as objective truth, it would be unquestionable. Instead the scientific method asks us to consider it as the best answer we have currently, until new evidence emerges.
I agree measuring reality is subjective, but thats the only way we can interpret it. Our eyes even lie to us to make up for deficiencies, for example your blind spot.
The key point here is underlying our shared reality is a shared truth. But neither of these are shared by everyone, and it's almost always too nuanced to pick one side and declare that the objective truth. Epistemology is centered around this challenge.
> (1) consensus is maintained by a diverse set of potentially anonymous participants. Ideally, this would mean literally all humans, but that's of course impossible. The set should be large and diverse enough so that even powerful actors (especially governments) have no ability to force them to do something.
An anonymous direct democracy? Even representative democracies are struggling to have a fully informed and engaged electorate. I love the ideal here, but practically I don't see it happening. In particular becuase governments have enough power to manufacture fake people to take over a system like that, particularly if it's anonymous.
> (2) the most profitable course of action is for every validator to follow the protocol honestly. In ethereum's PoS that's achieved by slashing maliciously misbehaving validators.
> PoW has an orders of magnitude weaker punishment here (just mining rewards for the time spent on mining a minority chain). Most DPoS systems rely on the assumption that a majority is always honest, even if it makes financial sense to not be - eg. Cardano's Ouroboros.
We have law enforcement and justice systems to account for when people do not act honestly. History proves time and again when the conditions are right, people will not act according the time's concept of honesty.
Blockchain was meant to solve the Byzantine Generals Problem right? That describes the need to communicate when not all actors are reliable.
Humans are not always reliable, sometimes maliciously so.
The bigger problem is that honesty is a question of morality, not logic. How can you encode that into a smart contract? Doesn't this mean we're swapping a judicial system for an unelected one?
> In the case of image NFTs, it's an improvement on physical art. The point of buying physical art is to either signal wealth, or to use it for money laundering. Image NFTs are infinitely superior for the first - because they are globally visible (as opposed to a physical piece of art) and fake NFTs are impossible (so you can't buy a cheap replica and pretend you spent a fortune on the original). They are much better for money laundering because they are inherently global (no transport issues) and evade identity checks in the banking system.
I mean I'm glad you said the quiet part out loud, because outside of money laundering it's hard to describe the NFT craze as anything but a bit suss.
Some of us buy physical art for the beauty and intrigue of it. Though not from Sotheby's.
NFT's as a record/marketplace for a physical asset backed by a means to legally enforce the ownership is something I can see happening.
But NFT's that record a log of ownership over a JPEG?
What's stopping me from making my own chain, and selling the same JPEG NFT there?
And then, how do we prevent the wash trading that's artificially inflating prices?
> Therefore, nobody buys expensive existing art just because they like it, which leaves two reasons I mentioned.
Ha agreed, though if they do they'd usually loan it to a museum. There are collectors who do it for the love of the art though, they exist.
> Yes. The entire history of human civilization consists of solving social problems with technology.
Fundamentally disagree here. At macro and micro scales social problems are solved with social solutions. I've seen this at enterprise levels and startups. If you try to tackle a social problem with a technical solution you alienate people and make the problem worse.
Technical solutions to social problems lack the humanity required for a meaningful fix.
A hilarious example is the type of solution that led to Goodhart's Law[0].
When a measure becomes a target, it ceases to be a good measure.
Probably the most famous example being[1]:
> In India while it was a British Colony, the Colonial Rulers wanted to reduce the number of snakes. So they offered to pay for dead snakes brought to them. Goodhart’s Law: Enterprising Indians began to breed and farm snakes to kill and get their reward, which was much easier than killing wild snakes, actually increasing the total snake population.
How is global consensus not centralisation though?
Humans are notorious for disagreeing on everything. Having one global arbiter of truth/consensus seems implausible at best.
Anything that could provide global consensus will need to be backed by something formed from a human brain. And our brains are not perfect. We keep updating code because we miss things. Who would maintain the system for global consensus? And how would they not just be our new rulers?
As a perfect example you mention ownership as a feature of global consensus. What do you own with an NFT? Who enforces that ownership? What stops someone pirating an NFT image?
Ultimately this screams of solving a people problem with a technical one. That never works. Not without strong co-operation of the people involved, and even then technology plays a very small role in the solution.
> The scalability argument is true in isolation - everyone agrees it's currently a problem, but has nothing to do with centralization or not.
It does in that a truly decentralised system inherently cannot suffer from the scaling problems cryptocurrencies have.