Hacker News new | ask | show | jobs
by bko 1905 days ago
It might be covered in one of the resources, but when people talk about NFTs, they just mean some token on Ethereum that the issuer of said coin say is "the" product. For instance, the NFT for "the first tweet" that sold is just someone created an Ethereum token and said "this one is the first tweet", presumably w/ some text in the token that states the tweet.

In that regard, its kind of centralized. There can be multiple NFTs that claim to be tweets. There's no naturally definitive "twitter" token, unless it's managed by Twitter itself. But you're relying on the issuer of the coin to say "I am the NFT coin" and everyone to agree.

Real world NFTs (e.g. art and collectibles) are unique in that there is no dispute about what the original is. The art was created by the artist or its a forgery. In that regard NFTs rely on some issuing entity that bestows upon itself an arbitrary form of ownership. I can make my own Twitter token and if I get someone to buy a coin for an absurd amount (insider dealing?), and convince reports to write about it, I can be "the twitter nft"

Or am I thinking of this incorrectly?

7 comments

It's a simulacrum, in the post-modern sense.

For those who are unfamiliar with the concept, a simulacrum is a copy of something that has itself become hyper-real and usurped the reality of the original. The canonical example is Disney World, which is a copy of Disneyland, which is itself a copy of many elements of Americana (Louisiana Bayou, the frontier West, Main Street USA, etc.) Disney World has eclipsed Disneyland in visitors, which itself has eclipsed visits to various small town main streets or the Louisiana bayous. You plan a family vacation to Disney World, you don't plan a trip to Bumfuck, Wyoming. Even though it's a copy of a copy (and a poor one at that), its reality has eclipsed that of the original, just because the original is hard to get to and kind of boring.

Similar with NFTs. Yes, you can make an NFT of the Mona Lisa and claim that you have the only one. Yes, you're a lying swindler and probably a bad person. But your copy is much easier to trade, move around, and store than the actual Mona Lisa. And just as there's only one original Mona Lisa, there's only one original NFT of the Mona Lisa, so your copy still has the same scarcity. As more capital comes onto the blockchain, your NFT is likely to acquire its own hyper-reality and become the Mona Lisa, and the original painting will be hidden away in a museum somewhere and disappear from trading.

> And just as there's only one original Mona Lisa, there's only one original NFT of the Mona Lisa

I apologize if I'm misunderstanding, but can't you create as many NFTs of the Mona Lisa as you want? What's to stop someone from just creating a second one, even the original owner of the Mona Lisa? You can't copy the Mona Lisa (at least not in such a way as to call it the original), but you can make another NFT of the real Mona Lisa and...as far as I can tell it's just as good.

I realize you can't copy the chain that's already descended from the first NFT, but why does that matter? You can still say this is a genuine NFT for the real Mona Lisa, and the problem is you can make as many of those as you want. Or at least that's how I understand it.

Copying something digital is very different from copying something molecular. A digital copy is as good as the original. A molecular copy is almost certainly possible to distinguish, and if you can't you may as well not have an original.

My understanding is that the NFT also contains a hash of the work that it represents, so there really is only one of a given digital work. You can create another NFT of a different digitization, but there's still distinguishing factors between them that would make one "better" than the other (higher resolution, maybe, or more ubiquitous file format). Also over time a particular digitizations' status of being the first gives it its own legitimacy, unless the environment changes in a way that makes competing ones significantly more convenient.

It's sort of like the forking problem for cryptocurrencies in general. Anyone can fork the Bitcoin code and say they made a new cryptocurrency, and many people have done exactly that (Bitcoin Cash, Bitcoin SV, Bitcoin ABC). In practice, the fork tends to die out, because much of a cryptocurrency's value is in who agrees that it's valuable, and there's a big first-mover advantage there.

An NFT is basically a number in a smart contract. Through the ERC721 (there are others) interface, you can ask the smart contract which address owns that particular number. So, "contract 0xdeadbeef... , who owns 123456 ? It's owned by 0xabcdef..."

This is totally fine, but keep in mind someone could deploy an identical contract to a different address, which also has a 123456 in it, which is also owned by 0xabcdef. You'd have to be careful when you're checking the ownership that you're checking everything. I could imagine an unscrupulous person saying "Look, I sent you 123456, it's in your address, see?" and then absconding.

> My understanding is that the NFT also contains a hash of the work that it represents, so there really is only one of a given digital work

I've never heard that before. Can someone confirm this is true? Is it solving for a hash of the original digitization of the work or something?

My understanding is that John can take a picture/jpg of the Mona Lisa, mint an ERC-20 token and sell his picture/jpg on a NFT marketplace. Jane can also take a picture/jpg of the Mona Lisa, mint an ERC-20 token and sell her picture/jpg on a NFT marketplace. I think the difference shows up when Jack Dorsey mints a screenshot of his first tweet and sells it, it's worth $2m because it's sold by Jack. Whereas I could screenshot Jack's first tweet, mint it and sell it for $0.01 because it's sold by me.

Very happy to be corrected if this is incorrect.

This is also my understanding, and my point was (in this example) Jack Dorsey can just do it a second time, and a third time, and so on.

And it's not like Monet making more paintings, because each painting is unique. This is an identical digital copy from the same person, with no guarantee of its longevity.

It's not.
"Sod going to an art gallery to look the actual paint committed to canvas by the legendary artist over 500 years ago, we can queue up to look at this PNG copy and admire the uniqueness of the cryptographic token some anonymous programmer attached to it recently" said literally nobody ever.

Rollercoasters and Mickey Mouse make pastiches of common things more interesting; creating an inferior copy of a famous thing and trying to compensate for its inferiority and inauthenticity with a cryptographic string has the opposite effect.

What if you had a human model (or robot) who looked liked Mona Lisa, dressed like Mona Lisa and was certified as the only recognized copy of Mona Lisa and by purchasing the NFT you were the only one who could have s.e.x with her?
This has got to be a plot point from a William Gibson story.
How do you prevent two people from claiming the same nft? Wallet? What if two people have the key to the wallet? I'm not sure how that part works.
Yep, another way to put it is that any NFT is worthless unless it is tied to some network effect.

This can be merely social convention, e.g. an artist announcing that this is his Ethereum address, or an auction house acting as an intermediary. But it could also be a network effect based on software, e.g., a game, or something in between like the original CryptoKitties.

Yeah, that's what I figured. I get the idea of it being tied to a game, but you don't really get the benefit of block chain, except for making the token sellable without having to go through the proper market channels.

Overall I'm very bullish on crypto as digital gold, less so on Ethereum style apps and even less so on NFTs. It's an interesting litmus test. I see a lot of crypto enthusiasts speak highly about NFTs which I find odd. They generally have very well thought out and reasoned arguments for Bitcoin as a digital store of wealth, but then they go into NFTs and talk about scarcity. But the scarcity of Bitcoin is only valuable because it is the defacto digital currency, due to it's survival up to this point. It has dealt with governance issues, was built of strong cryptographic principals and most importantly stood the test of time (lindy effect). Then they speak about NFTs in a vague way as though they're naturally authoritative without pointing out the obvious: you don't actually own the digital asset you're buying. If it actually had some real concept of ownership (e.g. owning a tweet allows me to delete the tweet), then it would be a different story. But as it stands now I don't see the value.

The cheer leading from the crypto community is odd and makes me question their principals. I'd be happy to be proven wrong and have someone steelman the argument for NFTs without relying on "its a scarce asset and people want to own scarce assets". It's also very early days in NFTs, so ten years from now if the Twitter tokens are still selling for millions, I'd be more willing to agree as it would have Lindy effect by that point.

> I see a lot of crypto enthusiasts speak highly about NFTs which I find odd. They generally have very well thought out and reasoned arguments for Bitcoin as a digital store of wealth, but then they go into NFTs and talk about scarcity.

I've seen the opposite - most crypto enthusiasts I have spoken to don't seem to understand NFTs at all / why they are considered valuable by other people. I fit into that camp.

> Overall I'm very bullish on crypto as digital gold, less so on Ethereum style apps and even less so on NFTs.

I'm quite bullish on Ethereum style "defi" apps - once we finally have decent layer 2 options and software deployed on layer 2, I think interest rates on stablecoins will be a large incentive for folks to start onboarding. The big remaining piece then is convincing people to spend their stable coins at stores - I would expect some level of "rewards" paid by the store would be sufficient. Essentially, an app that you can deposit funds into, earn interest that is well above what your bank account pays, and earns rewards that are competitive (or ideally beat) with credit cards.

> Essentially, an app that you can deposit funds into, earn interest that is well above what your bank account pays, and earns rewards that are competitive (or ideally beat) with credit cards.

That would be fantastic! Give me 10% interest on my wallet, and 10% "cashback" for my purchases with crypto, and I'll switch my whole financial life to crypto.

10% cashback is probably pushing it unless the wallet is giving away "platform tokens" as a growth hack at the beginning. 2 to 3% is probably more realistic given what merchants pay credit card providers.

Long term, if merchants were smart, they'd make their own reward systems on top of this to encourage usage.

Why does defi imply higher interest rates?
You earn interest by "lending" your digital asset. People want to speculate against the asset (i.e. bet the price will go down). The only way to do that is borrow someone else's crypto (pay them an interest rate for the trouble), sell it, and eventually buy it back later at a lower price.

The interest rate is determined by the market, and since its pretty early on in defi world, the owners of the crypto demand a higher interest rate to make it worth their while. Also, the more volatile the asset the more valuable it is in terms of derivatives.

I'm not 100% sure of all of the reasons - I imagine some of it is defi being harder to access, some of it is the increased risk (uncertainty / it's reasonably new), some of it is removal of middlemen, and some of it is incentivized behavior as a growth hack (similar to Uber subsidizing rides). Curious for other peoples' thoughts on what drives the higher interest rates.
The confusion comes from how NFTs that represent images have become so popular right now (I believe it’s because it’s the easiest use case of an NFT and people want to make a quick buck).

NFTs themselves are just a cryptographic representation of something unique. It can be a token that says “you own this image”, but it can also be an in-game virtual item, a software license, a digital trading card, etc...

There is some level of centralization in many cases. For example, a trading card game will likely own the contract that can mint cards and own the art for the cards. Other services can read that contract to see what cards you own and use that data inside their own game logic (though probably not the images due to legal reasons).

This would already be useful for something like Magic the Gathering where unofficial online services exist to play by whatever rules you want. The last time I used one, you just put your deck list in so everyone used the rarest cards. They may have evolved now to check your Magic Online account, but you can’t do that as easily in Magic Arena.

Assuming the creators don’t limit your freedom in their smart contract, you can also buy, sell, or trade the NFT to whoever and for whatever you want unlike a centralized service which wouldn’t enable access to an open market outside of the game’s ecosystem.

Here’s the thing - if it relies on a central authority to say “this is the genuine article” why do you even need a blockchain? A PGP certificate or X.509 is arguably “a cryptographic representation of something unique”

EDIT even something like a domain name - which is arguably a better example as only one of such a thing can actually exist in that only one owner can derive the value from it.

An NFT on the blockchain can be verifired crytographically that it is the "original", it can't be duplicated, and you can transfer it to anyone if you have the private key and the gas.

A token off the blockchain controlled by a centralized group could transfer ownership without a key at all. They could issue the same one to more than one party and duplicate it. Someone else could claim to own the record and they could decide you don't anymore, they get to decide who it is sold to. Or prohibit you from transferring it at all.

The real world item that the token refers to is by definition “a token off the blockchain”.
What if two people have access to the private key? How do we prove who owns it?
I don't like NFTs, but the point of putting it on the blockchain is that it's globally available, verifiable, and immutable. Many versions of a PGP-signed message could exist and be passed around.
My point isn’t that the token itself is globally verifiable but that the asset it refers to actually is ...
Because then it’s not on the blockchain. Yes, I could sell you a txt record, saying you own this hash, which represents a jpg, but then it’s not blockchain.

Honestly, I don’t know.

Maybe it’s because the blockchain is immutable? Like, I could change that txt record because I’m evil or something. That makes sense to me, in terms of technology.

Someone is trying it your way: https://signed.work/

The problem is that these signatures cannot be traded anymore, since no one is tracking who owns which editions. I can sell my signature to you, and then sell it to someone else tomorrow. The artist intended to mint 100 "prints", but suddenly 1000 people claim to own one.

No but the point is, when you’re talking about physical artefacts you can in theory issue as many tokens as you like ... it’s still down to a central authority to certify.
That's the other reason I hate NFT. The only plausible use case is as a terrible DRM key.
Or, a better DRM key... at least you could transfer it without permission.
Not really. Access to the content referred to by the NFT happens at the whim of the platform. That gives them the ultimate leverage. In this case, transfer doesn't amount to much if the "object" is rendered useless.
Still no worse than some hash key that can be invalidated in the next version with a blacklist. At least a DRM key is something that NFTs are somewhat suited for.

For fuckery, there are still cracks.

> Real world NFTs (e.g. art and collectibles) are unique in that there is no dispute about what the original is. The art was created by the artist or its a forgery. In that regard NFTs rely on some issuing entity that bestows upon itself an arbitrary form of ownership. I can make my own Twitter token and if I get someone to buy a coin for an absurd amount (insider dealing?), and convince reports to write about it, I can be "the twitter nft"

I think of NFTs as a fancy system of determining provenance: X bought hash Y from seller Z. Whether other people assign meaning/value to that transaction is subject to the usual whims of human nature.

> am I thinking of this incorrectly?

You've got it, more or less. The "surely I must be missing something" reaction is actually indicative of the correct understanding: the emperor has no clothes.

NFT: No Fscking Thing

> there is no dispute about what the original is

Still though, the value of collectible goes up if you have extensive documentation proving it is the original, vs it just merely being the original, or worse, it being the original and someone else having extensive documentation convincingly making the case that their counterfeit is actually the original.

NFTs can't prove originality, only ownership. Since digital things are easily copied exactly, that's all you can do
For a very sui generis definition of ownership.
In reality, I think a lot of art these days isn't created by the artist, at least not mostly - it's created by their technicians or students or other people they employ and then just given their blessing, maybe with a few last finishing touches. The thing that makes it not a forgery is literally the artist's blessing; in principle you could even have a "forgery" that was created by the same person as the genuine article. This isn't widely advertised for obvious reasons but it isn't exactly a secret either, and it's not really a new thing - the actual creation of some of the classic, famous fine art paintings was a lot like this too and it makes verifying authenticity very interesting.