I accept that maybe Ethereum could be valuable (I don't really understand smart contracts well enough to say..) and if they move to proof-of-stake that will make it more sustainable. Likewise, the sheer anonymity of something like Monero might be valuable for the black market.

But everything else just seems like people seeking a "greater fool" to sell to.

Imagine if my sql db was free but charged me $10 for any write transaction. That would suck. When the price gets to an operational level, people will use it.

I think the price it will eventually get to is Napster/BitTorrent-level.

I dump a ton of crap on blockchains and just set the transaction fee low until someone takes it

Also check out gas tokens as a concept, you can prebuy gas when gas prices are low and subsidize your future time sensitive transactions when gas prices are high

Try setting 1 cent as your transaction fee and see if anyone picks that up.

This works great if you already prebought way back when, but is not suitable for any current use case I can think of. And certainly not usable by many.

There’s not many situations where $10 for a few bytes is needed, even with the cool feature of free storage forever.

For comparison, 1GB is 2 cents per year on AWS deep glacier. And I trust AWS to be around in a hundred years more than ethereum. Not to mention that storing 1GB costs a fortune in transaction fees.

As long as the system relies on externally originated information for proper functioning, we’re back to the issues of trust the technology was intended to resolve.

That is why digital currency actually still seems like the only appropriate application of the blockchain, although it may never see widespread adoption.

Instead, the job of producing non-deterministic inputs is offloaded to an off-chain source, an oracle, which still requires some degree of trust in an external entity.

My information might be out of date, though, and some implementations might have resolved this issue without destroying the integrity of the entire system. And I will concede that gambling seems to be the most popular use of the tech so far.

This is the basis of most distributed random number selection algorithm, which existed since a lot before blockchains.

In a smart contract, it could look like something like that:

1- each person which want to participate in the bet pick a random number, then send: their bet (0 or 1), the hash of the random number, and of course money to bet.

2- after some specified time, the first phase finish and no new person can bet. now everyone has to send their actual random number to the smart contract

3- after some specified time again, every random number gathered which had the correct hash are xored. If the popcount of the result is even, 0 win. otherwise 1 win. Everyone who send their random number and guessed correctly get a share of the total amount.

edit: note that I'm not overly familiar with smart contracts, so if this cannot work I would love to be explained why.

Of course there is more than one miner etc., but having a positive expected outcome is enough to be worth it to play.

All criticisms of NFTs could just as easily be applied to the way that all digital artists have always sold their art. Where were you, ten years ago? Why weren't you yelling at my roommate for being a scammer?

I felt the same, but now regard NFTs as the first meaningful application of blockchain.

First, let's talk about land registries. In developed countries, the state administers a land registry. There is a strict, human-administered API against this registry. It allows users to inspect the full history of land; cause property splits/mergers; execute land exchange; and there is capacity for extensions like water-rights and native title. There is formal governance around some of these operations - conveyancing; council approval.

NFTs have the potential to do something similar, initially for artwork.

Imagine if every time an artist created a painting, they issued a NFT against it, and got this authorised by an authority. That authority does not yet exist, but Christie's and Sotheby's could set up in this space with low effort. Each time a transfer of the artwork occurs, both the NFT and the physical artwork change ownership to the new owner.

This would have a snowball effect, and quickly become the new standard for provenance. Insurance companies would insist that works were marked on the exchange. A stolen artwork would be nearly worthless. It would be pointless to duplicate artwork, because the sold duplicate would effectively destroy the original.

Extension: firearms. Firearms are valuable goods, and would soon be forced onto this registry. Perhaps the state would encourage this transition by asking that firearms ranges mark the transition of guns to and from the range premises by marking this on the chain.

Extension: wine. There is a live problem in China, with rubbish wine being fraudulently marked as premium Australian varieties. Each bottle could be chipped in the cap with an on-chain identifier. In the wine shop, you could use your phone to scan the identifier, and witness custody of that wine back to the producer.

Evolution: complex goods. The next evolution would be to trace the grapes that went into the wine back to the vineyard that they came from.

Reputation becomes very important in this system. This would encourage a new type of auditing profession, who serve a similar role as conveyances, but for chain operations. Once you have that, you could mark carbon credits of goods on the chain, you could have eggs tagged as being free-range or organic, etc.

You're still trusting a third-party to release the data. Whether they release it via a blockchain or via a website, what's the difference? The trust model is the same.

You allude to a standard, which indeed is always valuable since it allows people to exchange freely with each other. In your other answers, you seem to take it for granted that using a blockchain means everything is interoperable. But a blockchain is no guarantee that any standard would prevail, especially in the long run. There are already tokens using different interfaces, or even using different blockchains. At this point, "using blockchain" is equivalent to having a public API, no more no less. That's obviously already possible without a blockchain.

So I don't see why you think the use-cases you mentioned benefit from blockchain at all.

There are other problems that need registries where there is no party with inventive and means to act as the state does above. I think the physical-art and wine examples fit that.

The key quality of blockchain is that it allows groups to form consensus. Even though they compete on some matters, all stakeholders want the platform to succeed. If it doesn't, their investment is worthless.

Hence, Blockchain can act as a synthetic state for enforcing standards and governance of registries. This may go as far as having elections, rulings. The registry data itself could live off-chain. [Sure, there are details to be sorted out here. Maybe the blockchain entity would control all APIs, to allow it to enforce standards against the registries. Those APIs may not necessarily be on the blockchain.]

Due to incentives, I don't think the trust model is the same as a standalone registry. One of the many challenges of dealing with data providers at the moment is they can routinely have low data quality and still get paid. Whereas a state would intervene if a land registry fell below standard.

"In your other answers, you seem to take it for granted that using a blockchain means everything is interoperable"

This is a real problem. I think the foundation problem to solve is how to get somebody to fill a role equivalent to the state in the land registry example. Once you have that, you have a path to solving issues such as this.

Thanks for poking at this.

> The key quality of blockchain is that it allows groups to form consensus. Even though they compete on some matters, all stakeholders want the platform to succeed. If it doesn't, their investment is worthless.

Isn't that a company (or other kind of association) with shareholders? That's not specific to a blockchain, is it?

> There are other problems that need registries where there is no party with incentive and means to act as the state does above.

Either there are common financial incentives for stakeholders to agree on a registry (you can sell for a higher price if consumers know what they're buying quality products), and it will happen eventually. Or there are none, and it won't. In both cases, having a blockchain does not bring anything. Wine is I think a bad example given that there is already labels such as https://en.wikipedia.org/wiki/Appellation_d%27origine_contr%.... This specific label seems to have its own law, but as soon as you have something resembling trademark law you can have such a label, even if it's designed by a private association of producers. It has its own problems, but I can't think of any that are solved by the use of a blockchain.

To summarize, I do agree with you that we should trace production, quality, and better registries can bring a lot. But there is no blockchain in this discussion, so I don't see how you connect the two.

Just think about the few comments you wrote in this thread, and what you wish would exist in terms of registries, "synthetic state", and so on. How does removing "blockchain" from this equation make any of what you wrote impossible or even slightly harder?

Yes. There are jurisdictions where there is no prospect of this, but the answer to your question is yes.

"Isn't that a company"

Yes.

"as soon as you have something resembling trademark law you can have such a label, even if it's designed by a private association of producers"

Appellation d'origine contrôlée does not show chain of custody, so it's easily copied. But it may be possible to project a registry into low-rule-of-law regions without rule of law.

"Just think about the few comments you wrote in this thread, and what you wish would exist in terms of registries, "synthetic state", and so on. How does removing "blockchain" from this equation What I am proposing is significantly different. make any of what you wrote impossible or even slightly harder?"

Will do.

An attraction of a non-blockchain solution - it is so much easier to write robust code: put an API around a central server, write good acceptance tests.

So I think the right play would be to not speculate and wait for a government to use blockchain/NFT for registries.

The land registry is a good idea, but is already done without blockchain/NFt so I’m not sure the value of layering this on. For these things to work, it’s better to have a common operator under a monopoly (government) and fees be nominal.

This is how deeds work in the US today, but there’s some clerk filing sheets in a courthouse and processing transactions. This can be improved by databases, etc but NFT doesn’t provide much benefit over a parcel number. As we automate this process, we would not have a market for selling NFTs, we would just need a robot instead of a clerk.

Organic: it would be unrealistic to attempt to solve the wine problem as a one-off registry. But once the infrastructure exists, it becomes trivial to solve the wine problem.

You point out that many of the individual scenarios could be solved by a conventional registry. This is good criticism, thanks.

Maybe the role of the chain is not to be a log of everything, but rather to be a system of governance.

Participating registries need to be onboarded to the chain, and need to abide by the rules to stay onboarded.

Governance consists of humans (proof-of-stake parties) who serve as regulators and magistrates for participating registries. They ensure that the APIs are in good order, and have the ability to excommunicate systems that are not playing by the system. In this model, the chain serves the same role as a nation state, but it does not have land.

The artwork problem is still useful, though, because it is a problem that allows you to bootstrap the platform. Analogy: Amazon was initially about books, but Amazon was never really about books.

Or a single trusted entity is more likely to organically arise than an NFT market solution.

The issue that these sound decent in theory, but in practice they suck because of costs. And for sustainability, I’d rather deal with the government to run it.

If they would arise organically, they would have already. The problem is that the incentives is for jerks trying to make money than people to collaborate.

The only area where I can see something like NFTs working is pure digital goods which can only be enjoyed by the keyholder decrypting them. (This sort of DRM also exists in other forms, but the risk of the vendor deactivating the online validation system when they no longer wish to maintain an old game is a real one)

Proof-of-stake participants are strongly motivated to ensure the integrity and good reputation of the platform.

By analogy, this is a bit like the Netherlands Water boards. They have power and motivation to do whatever it takes to ensure the dikes are in good order. The population are motivated to support them in this. Everyone suffers if the water gets past the dikes: it destroys the platform.

Pure digital goods are easily duplicated with no loss of utility. For example, you can copy a piece of music and lose nothing from that. The value I see relates to scarce goods.

I'm not so sure it'll work well for items for consumption. There are several issues

- At any point you can swap out the content of the wine (if you can re-cap it), and sell the genuine wine to someone who trusts the person doing the fraud. - You're somewhat dependent on many people actually scanning the bottle. You can duplicate a bottle, and if someone doesn't scan it, it's not detected. If you've managed to insert yourself in the supply chain and only care about milking it for profit until you're detected and then running away with the money, you might get away with it for a while. - I know of at least one liquor from China that marks each bottle with an ID that you can check. It achieves much of the same benefit without blockchain if the supply chain isn't very long and complex

I also don't think PoW or PoS makes any sense with these use-cases where you track items or handle logistics. You probably want some organization setting standards for auditors, and to handle necessary changes to the blockchain algorithm and such. So might as well be proof-of-authority.

If someone cares about having a piece of art badly enough to steal it or create a fake do they care at all about what an NFT says about who owns it?

I think we live in a world where holding the real thing in your real hands is all that matters.

In regions with the rule of law, there are things more important than physical possession - ownership, contract.

1) In contracts, you can add a clause that mandates a dispute-resolution process that must take place before either side goes to real courts. These could be codified into the blockchain contracts. If that process is well run, I assume the court will have no reason to over-rule it.

2) In blockchain matters, is there a distinct party to file against? Where there is, do you know who they are? And are they in the same jurisdiction as you?

It's weird how tech inventions came from publicly funded research, yet it's being adored by libertarians and anti-government people.

At the end of the day "ownership" mostly has any meaning within the legal system. NFTs seem to create something akin to a shadow copyright but which makes no sense. Oh well, maybe I am just a simpleton.

No, no, no. It's worse than you think. An NFT is more like a plastic cup with Pikachu printed on it. You don't own the rights to Pikachu. You just bought a cup with his face on it. With NFTs, there is no contract to own the actual "thing". Just simply the piece of "merch" (the NFT) with the likeness associated with it. Except, a plastic cup still has value to drink from or hold your toothbrushes.

E.g. on some art NFT platforms, a legal license is attached that says that if you hold the NFT for an artwork, you get permission to display that just as you would have permission to display lets say a piece of software art you bought from them. The NFT just is a representation of who holds that license right now, instead of a software license represented by a piece of paper being sold, or an entry in a license database somewhere.

Digital artists have struggled with how to sell a digital thing that can be copied losslessly for a while. Expensive digital art is often sold as some physical thing, e.g. a print, an installation with some screens showing the artwork, or a movie on DVDs. The limited number of the physical things provides a scarcity that apparently has value to a lot of people, leading to higher sales prices, but is also somewhat frustrating if you want the digital thing to be the art. In contrast, when they try to sell just files, that's basically not happening. People will spend hundreds on a print, but not $5 on the digital file the print was created from. And for some, it now turns out that NFTs also provide that scarcity, in the purely digital domain. Sure, everyone can go and download the file, but if you bought the NFT you have a record that proves that you actually bought the thing from the artist. That appears to have value to some people, and they pay the same kind of money as they would for a print etc.

If you don't have an explicit legal thing, it comes down to social convention that this (virtual) thing you are holding has some value to some people. Some collectibles are somewhat of an analogue: Stamps. signed photographs. pokemon cards. Skins in Counter-Strike. All those have no obvious reason why they are worth a lot a money (or not), except that enough people want them and are willing to spend to get them. If a piece of paper with some dudes signature somehow has value to some people, why can't a bag of bits with a digital signature have value to others?

And of course there's lots of hype and too much money floating around that makes all this more extreme, leads to people throwing money at it in the hope to make a profit, and makes the amounts paid for the headline-making things just seem entirely stupid. But that's roughly the forces at work, if we agree to the specific value judgements or not (very similar to collectables markets again, those seem equally weird to me in the extremes).

Nobody will ask you in 20 years what it was like to own an NFT, because they can all have the exact same experience by just looking at the blockchain on their device.

Humans ultimately value stories and experiences, and anything that gets work out of the way so we can spend more time having them.

But yes there are ways to link NFTs to products. Louis Vuitton and Nike are linking NFTs to their products with things like built-in NFC tags and QR codes plus other probably more secretive things.

Well, the high-end art market is kinda scammy, is it not [0]?

I think it's pointless to pay millions of dollars for an original painting when a decent copy can be had for a fraction of the cost. And it would look just as nice.

---

[0]: https://qz.com/103091/high-end-art-is-one-of-the-most-manipu...

100% the same thing

If there were destructive scans of a Picasso where the only thing digital was the NFT, then yes.

It would also be kind of neat in a terror kind of way to actually shred the Picasso and issue an NFT.

The problem is that I still have copies that look just like the Picasso. Infinite.

If NFTs start being linked with rights, then that might work. Like the copyright ownership is conveyed to the owner of the NFT.

It’s kind of like buying a numbered print of a painting and trusting the artist to not make more.