You can do the same thing with blockchain NFTs, just mint another blockchain NFT with the same content. You can say "well, the NFT that was minted first is the true original" but the same is true of the "double sent" digital signature NFT, one will be dated more recently than the other, which affords one identical guarantees of authenticity to the blockchain NFT.

Well, no, only the minter can do the first step:

> "$doomrobo_public_key owns $k as of $date" signed with the minter's private key

If you mean the minter could sign multiple owners of $k, of course that’s true, but that’s like saying that a painter could paint the same thing twice or sell printed copies of their paintings. Of course they can, and if they claim that each one is unique they’re likely to find that people don’t value their stuff much.

It's not the same as selling printed copies or painting multiple painting. It's verifying that two different people 'own' the same painting but neither of them has guaranteed rights to it.

> will be dated more recently than the other

How does the receiver of such a transfer know that the transfer is final? There's no way for them to know that an "earlier" transfer of same item won't show up days later.

Right. This is, by the way, the sole innovation of Bitcoin. Describing a mechanism to establish a sequential order of transactions with finality.

Timestamping servers. Something like RFC 3161 or WebPKI's Certificate Trust logs.

That doesn't help you know the transfer is final, it just gives you a timestamp that the transfer occurred at which somebody reputable has attested to. You can't know that the owner didn't sign a transfer to some other buyer with an earlier timestamp.

There is some project. NFT Replicas - https://nftreplicas.net/ Nothing is stopping any one to mint another NFT with same content.

Indeed, but NFT enthusiasts would say that the replicas are not valid because they were not originally minted by the trusted author, of course, nothing is stopping the author from minting multiple NFTs of the same content so the point is moot.

Why would you want the minter to be prevented from minting multiple NFRs of the same content? The content is totally moot! It’s the ownership of the scarce token which the author claims uniquely represents the content that people are valuing. Obviously if the author is minting multiple tokens and claiming they all uniquely represent the same piece of content, most likely no one is going to value those tokens much. None of this strikes me as odd or troublesome. It’s how all collectibles have worked long before NFTs.

Minty's "Useless Man" could have been about NFTs.

https://www.youtube.com/watch?v=tqAOaNS1tqM

(May be NSFW depending on who you work for...)

Wouldn't the author be different if you just minted a copy? So if the original is from a trust/known entity (like an IRL company or suitably famous artist) this doesn't work. The list of trusted entities could be centralized, crowd-sourced, hand-managed, or anything else; let people decide who they trust to make those determinations and let them switch between them at will.

> Wouldn't the author be different if you just minted a copy?

The problem being described above is the possibility of the trusted author minting multiple digital signature NFTs for the same content. A reseller can't create a copy because they can't sign a message conferring ownership from the trusted source without their private key.

But at least you can check the blockchain for all the content the author has created.

The author could use a unique private key for each blockchain NFT if they wanted to avoid that.

The problem of how to associate a NFT with a physical object is a different one from the transfer/spending problem.

I think that is the "minting" problem.

An artist simply signing stuff does not even prove that a physical object corresponding to the token exists.

I like to think of other use cases, like real estate or stock options. If you talk about a plot of land, it seems feasible to find a unique identifier (you could even involve the government).

Edit: HN does not allow me any more replies right now, so I have to leave it at that. The reason why a "government signature on the deed" would achieve the same thing is because we trust on the government to keep a proper ledger. And the government asks for a lot of money for a transfer.

> I like to think of other use cases, like real estate or stock options. If you talk about a plot of land, it seems feasible to find a unique identifier (you could even involve the government).

A deed digitally signed by the government's private key satisfies this need without a blockchain.

"A deed digitally signed by the government's private key satisfies this need without a blockchain."

Only because the government keeps track of who owns what in a centralized database. Typically they also ask for pretty high transferal fees.

Huh? If you can’t verify that the public key is the author’s then you wouldn’t consider the message to be signed by the author.

Why wouldn't you be able to verify the public key?

You seem to be suggesting that a problem with NFTs is that the artist can mint multiple copies of their artwork, possibly using secret additional private keys.

The answer is that this is not a problem, because the whole point is knowing who the creator of the NFT is, and beeple trying to create additional copies using a secret wallet but not telling anyone will be treated by the market the same as me minting a copy of a beeple piece.

Thats not the same thing.

I just explained why it is. A one line response of "nuh uh" is not convincing.

Its an analogy and diverts from the original point. We were talking about how you were going to solve double spends without a chain.

You point about minting another is an interesting point but its not a double spend so we veer off the topic.

If we take your point then its the same as saying anyone can create a new crypto and call it Bitcoin. If I have an OG Bitcoin and you have a fake one then you can steel my real one. No you cant.

The reason is that people discern what Bitcoin is vs. other coins.

NFT cooy/paste is a problem but it is a distinct and different problem to double spend.

> We were talking about how you were going to solve double spends without a chain.

"Double spend" is a cryptocurrency problem, NFTs do not have this problem since unlike "currencies" the goal is to create a "non fungible" record of ownership for an off-chain asset; a chain of digital signatures satisfies this need, the double spend issue is not relevant to the stated use-case.

The genesis of this discussion is “NFTs are a scam and are retroactively obsoleted by digital signatures”. I basically said “how to avoid double spend in digital signatures”, someone replied “but you can clone NFTs!”. I said “yes but that is different”.

You are correct double spend is not a problem with NFT because they piggy back on crypto layers.

However double spend is a problem with just digital signatures that “retroactively obsoleted” NFTs.

Since there are multiple people commenting and steering the thread in different directions I feel this is being filibustered. So ill stop replying. Im exhausted!

> However double spend is a problem with just digital signatures that “retroactively obsoleted” NFTs.

As I already stated, this is not true. Nobody can stop an author from minting two different blockchain NFTs pointing to the same content, there's no difference between that and "double spending" a digital signature NFT. At the end of the day, the only thing the NFT does is prove "the author authorized this verbal statement of ownership", a digitally signed text file does exactly the same thing.

I think what you are missing is that for NFT adherents, they don't care about the original content, just the NFT.

btw, I own an NFT that represents all natural numbers, therefore all possible NFTs.

The only difference between a fungible token and a non-fungible token is its fungibility: everything else is the same. The stated use case for both is "I want to know for certain who owns something". A chain of digital signatures fails to satisfy that need. Please stop being so assertive about something you clearly don't know anything about: there are tons of useful complaints about how people are using NFTs, but this is not one of them.

A semantic distinction without an actual difference. At the end of the day all an NFT does is state "author X gives ownership of Y to Z", if you trust the public key of X a signed text file stating "author X gives ownership of Y to Z" gives you identical guarantees to a blockchain NFT stating the same thing. Nothing stops author X from minting a blockchain NFT that states "author X gives ownership of Y to Z" and also an additional blockchain NFT stating "author X gives ownership of Y to Q"... it's literally just a digitally signed string.

What blockchains provide is transaction ordering as a service. If Y is an NFT, what prevents that from happening is the code on the blockchain: the same kind of code (literally, on Ethereum) that prevents Y from being a fungible token for a cryptocurrency that they send to two people (aka "double spend").

If Y is not an NFT, but you know the original owner X, and you wish to create an NFT, the same kind of mechanisms work: the first time you attempt to assign ownership is the canonical one. Which you seemed to get up-thread, you just failed to "do the math" to see why digital signatures alone don't support that as you are so sure cryptocurrencies have no value.

If you send two digital signatures to two people claiming they both own something, the big issue is that the second person might not even know the first person exists. You need some kind of mechanism to invalidate the second one by exposing the first one in a trustless global ledger of events.... that's what blockchains provide.

If you only care about the semantics of disputes--which is potentially fair for claims over a physical object: it acts as a kind of "second factor"--you get a long way by just including the entire chain back to the root whenever you transfer ownership, allowing people to show prominence over other chains they obviously dominate...

...but any time there is a conflict--and not just on the original sale!--in your histories you can't compare timestamps to resolve the dispute as you can't trust them: digital signatures cannot show that one thing happened before the other thing as any timestamp is just data being signed, attached by the signer, and is meaningless.

To resolve these conflicts you need to use your distributed ledger to establish that no such signature has been signed previously: you need a way to authoritatively assign prominence. If you have a solution for this, what you have invented is a cryptocurrency and could be used as such (as there is no relevant difference between fungible and non-fungible tokens).

(And note that this is true even if you try to solve this by using some complex web-of trust of a ton of random third-parties or overlapping sets of parties that different people haphazardly might choose to trust instead of a linear blockchain... you are just talking about systems like Holochain or Stellar or Avalanche instead of systems like Bitcoin or Ethereum.)

> all an NFT does is state "author X gives ownership of Y to Z",

Incorrect. As @saurik says, but I'll just restate: an NFT also tells you the ordering of all such statements within one context (a single blockchain). So you can know whether this was the first such transfer by this owner or a subsequent one. This is the entire purpose of distributed databases, including blockchains.

Actually you can’t know it was by an ‘owner’ because nfts are not tied to the real world. You own nothing but the digital signature.

There are so many problems being skated over here about provenance and ownership and nfts provide nothing of value in that domain, they’re just a get rich quick scheme.

> because nfts are not tied to the real world. You own nothing but the digital signature.

The digital signature is thing that people want to own in the real world!

And yet all the talk is not about the signature, but the thing it refers to. Why is that distinction constantly elided? Is it because the digital signature has no value on its own?

It's because there is no other meaningful way to "own" the thing the signature refers to, if you apply a concept of ownership as you would to a physical piece of art.

Being in physical possession of the bytes to a JPEG is not what gives people the sense that they own this piece of art in the way they might own a painting, because the bytes are so cheaply copied.

Owning the intellectual property behind a JPEG is an entirely different kind of ownership; the value is in the commercial exploitation of those rights, and the buyer is now taking over from the creator in being responsible for said exploitation to get their money's worth. This is not what artists or art collectors want.

The whole point is to artificially recreate the traditional experience of ownership, but for immaterial/digital goods. People talk about owning the work (or an edition of it), not the signature, because that is the whole point of the exercise.

Immaterial artworks such as video art, performances etc. are already being sold in the art market; what you get is a piece of paper stating that you own the work. It's exactly the same as an NFT.

It may seem utterly useless for people who don't care about owning editions of immaterial works of art, but there is nothing scammy about it. It is done entirely in earnest.

Of course there are other ways to meaningfully own the digital artwork, you listed some in your response:

You created the image. You own the copyright to the image (which has legal weight). You may restrict distribution or reproduction in those situations.

These are all more meaningfully ownership than owning a digital receipt on the blockchain pointing to a url on a 3rd party site, which is more akin to buying a star or a piece of the moon from somebody who sets up a website and purports to be selling these ineffable things.

I disagree with you about the point of NFTs, the point of them IMO is to provide cover for ponzi schemes and convince marks parting with their cash that they are buying into a get rich quick scheme in which they'll get rich and someone else will be left holding the bag when they sell their receipt for digital art to someone else.

The references to art markets, asset tracking, provenance, new technology etc etc are just indirection used to distract from the fundamental transaction - give me money (including really high transaction fees for the marketplace) for this worthless receipt which says you own* this digital lion pic, unique amongst thousands of other similar lion pics because it has a cigar or an eye patch. No wonder companies like Nike, Games companies and football clubs and are rushing to take part too and sell fans NFTs, they can't lose!

* For certain values of own not conferring any rights over the image whatsoever.

> Of course there are other ways to meaningfully own the digital artwork, you listed some in your response:

By my count you listed just one - owning the copyright. But as I said, while owning intellectual property is certainly meaningful (and can be valuable), it is not art collecting. There is a reason why there exists no art market where people are trading copyright claims.

As a collector, I don't want to be in the business of managing an intellectual property licensing enterprise. Yet, I want the artwork I own to become a well-known, celebrated cultural object; the artist also wants this, and thankfully - incentivised and empowered by still owning the copyright - will try to manage their career to achieve this.

We can all agree that crypto lends itself to rampant and unrestricted speculation / gambling, which attracts all kinds of people looking for a cash grab or get-rich scheme. Those speculative bubbles have been going on for years, and in all likelihood will continue. It's an issue that artists and collectors in this space are struggling to deal with; people don't like that new releases are sold out in minutes, because flippers and their bots think they can make a profit (this is not dissimilar to event ticketing issues).

However, artists who have experimented with blockchain 10 years ago, were not looking for marks to scam. While pics of lions will go up and down in price, and people who bought even respectable art at the height of the bubble will be looking at a loss in resale value, this market will not disappear; it provides a product that artists and collectors find useful.

I am not of the opinion that this will be as world-changing and all-encompassing as web3 proponents proclaim. Art collection is a niche in the real world, and it will be a slightly larger niche in the digital realm.

Creating something is distinct from owning the copyright (though many jurisdictions automatically assign to creators unless otherwise assigned). For example illustrators for hire immediately assign copyright to someone else, so these are distinct.

The art market is a racket entirely divorced from value, if you want to emulate that, good luck to you, but know what you're getting into, the comparison with the manipulated and celebrity led current art market is entirely unflattering to NFTs as it shows them up for what they are - a get rich quick scheme.

I think this market will disappear when the cryptocurrency bubble bursts, which seems imminent now it has mainstream attention again and with the coming withdrawal of stimulus.

I think you are confusing the image of an NFT with the token itself. That two tokens point to the same JPG (for example) doesn't make them the same NFT. They are two distinct tokens.

You just invent a thing called notaries. You'll have to go back in time to do it since they already exist but that's besides the point. Anyways, a notary puts their reputation/legal liability on the line and counter signs your transaction. Since they've been granted license by a real legal entity with guns and jails they commit to verifying a transaction occurred at X time.

Monies for the transaction can be put into escrow with a third party (you'll need to go back in time to invent that too). When the notary notifies the escrow holder of the signatures they can release the monies.

When you have a legal system to enforce contracts and people put their own money or legal liability on the line two parties don't need to trust one another to conduct transactions. People buy expensive shit all the time. They don't need to trust each other with the money or valuables because they involve third parties that proffer up (staking if you will) liability to let the two parties conduct transactions.

The blockchain model replaces the staked liability with "expensive" proof of whatever. The only problem is monied entities can these proof systems. They can afford to run a majority of miners, stake the most (or spam stake offerings), or whatever otherwise supposedly expensive resource. They're only expensive resources if you naively assume all mining will be done by randos in their bedrooms.

Blockchains are just a way to do something almost no one needs in a slow expensive way.

dish and washing machines are slow and expensive (use a lot of electricity). But in turn you save some time.

Dish and washing machines are not slow. They cut out whole portions of the task of washing dishes or clothes.

You can require edition number / total copies in the original signed claim. Beyond that, you take it to the courts, which is the same thing you would need to do if the same resource was reminded on blockchain. The blockchain may only make it easier to detect a double mint.