Comment by Yizahi | Hacker News Reader

Comment by Yizahi | Hacker News Reader

Parent Full thread

This is impossible. Basically you MUST have a centralised DB at the municipality storing all the data at before (because it is impossible to do so on chain, due to chain tech limitations). Plus you have a useless NFT chain on the side, where NFTs are primitive URLs to the centralised DB (CDB). Even more - they are one way links. So that CDB doesn't know anything about blockchain state and NFTs. Next problem - no person owns NFTs, they are owned by wallets. Therefore these wallet must be deanonimised and registered at the municipality and in the CDB, so that there is at least one place to show who owns a deed.

Now some potential scenarios:

1. A sells deed to B. So now they do in on chain and NFT is sold from wallet X to Y. All of that is registered in the CDB and confirmed with paper documents, contracts and stuff. Maaaybe that can work, but as we see NFTs are useless fifth leg in a horse here.

2. Inheritance. Deed is inherited by law by person C (with or wihtout wallet Z), but chain doesn't know about it and thinks wallet X still owns NFT. CDB has actual data and overrides "wallet X" line, so that chain now useless and outdated.

3. Law enforces deed transfer from A to B. Again, NFT is not sold and chain is not updated.

The list of situations can go on and on.

And last note - these so called "dapps", which are neither smart nor contracts, are the most stupidly delusional thing you can introduce into a valuable commodities market. Only people not familiar with the scale and rate of exploits caused by usage of these dumb and badly designed programs can suggest it. Or the one profiteering from this. Phishing is over the top and even experienced users at r/CC are regularly losing everything due to a single click mistakes. This is not how housing market worked and will work. Not how any market works actually.

No, it's not impossible. You don't need to store all state on chain. You will just need to store a hash of the off-chain data on chain. The supplemental off-chain data can be stored in standardized storage systems, which they themselves might be highly distributed and standardized. (e.g. IPFS)

> You don't need to store all state on chain. You will just need to store a hash of the off-chain data on chain. The supplemental off-chain data can be stored in standardized storage systems

It amazes me how people say this with straight face. "Just store all the data off-chain!". Then the chain isn't needed.

I'm not sure how you made that jump in logic. The chain is absolutely needed as it stores and establishes cryptographic proof of ownership. This is the entire premise behind validity rollups for layer 2 networks.

> I'm not sure how you made that jump in logic.

What jump? You literally said that all the data is stored somewhere else.

> stores and establishes cryptographic proof of ownership.

It stores some unenforceable token that is only valid as long as those who actually enforce ownership agree that it proves anything.

So:

- all data is stored elsewhere, on centralised servers

- proof of ownership is enforced by centralised entities (governments, courts, police, etc.)

Great application of blockchains.

> You literally said that all the data is stored somewhere else.

No, I didn't. The NFT would be stored on chain and inside the NFT would be hash of the data that is stored off chain. Anyone can then source the off chain data from any location and then compute the hash of that data and compare it with what is stored on chain.

> It stores some unenforceable token that is only valid as long as those who actually enforce ownership agree that it proves anything.

This is no different than any other system. What makes that Oracle database that your property assessor uses to store your parcel information in enforceable? Oh yeah, the government.

> > You literally said that all the data is stored somewhere else.

> No, I didn't... NFT would be hash of the data that is stored off chain.

Yes. You literally said all the data would be stored somewhere else.

> This is no different than any other system.

If this is no different than any other system, then it has no benefits of any other system.

> What makes that Oracle database that your property assessor uses to store your parcel information in enforceable? Oh yeah, the government.

Indeed, the government.

Who makes your fantasy tokens referring to physical objects enforceable? Oh right. No one.

> Yes. You literally said all the data would be stored somewhere else.

Meta data (ownership data) is stored on-chain. (supplemental data is stored off-chain) So again, no I did not say that all data is stored somewhere else. The meta data that is stored on-chain serves an important role.

> If this is no different than any other system, then it has no benefits of any other system.

We are talking about the enforceability. Governments can choose to embrace a blockchain as their system of record. If they do, then NFT deeds are as enforceable as traditional property deeds.

> Who makes your fantasy tokens referring to physical objects enforceable? Oh right. No one.

The government would in this particular case. Maybe in other cases it would be a company and in still other cases it could be an on-chain contract.

Ownership data is not stored on chain because it is technically impossible. Chain only knows which wallet owns which URL-hash, and nothing else. Which real life person owns which real life physical or digital off-chain artifact (like deed, or picture, or a key), chain doesn't know and can't know. This info is stored in the centralised DBs managed my centralised entities like a company or government.

Ownership data meaning the NFT itself and which can be proved by the holder using his/her private key. Off-chain data (such as personal information or supplemental documents) doesn't have to be stored in centralized databases. It could be stored in federated systems, or a local computer at the owner's residence, or a hosted server controlled by the owner. The chain can hold a hash of the identifying information and then an application can pull the data from wherever it is stored and referenced and check that is correct by recomputing the hash and comparing that to what is stored on chain.

> Off-chain data (such as personal information

So actual data that proves that this physical person is the owner of this physical asset.

> It could be stored in a, b, c

> then an application can pull the data from wherever it is stored

Ah yes. Because when it's stored in a non-centralised database of "local computer at owner's residence" or "hosted server controlled by the owner" that data is easily accessible and retrievable.

You can replicate it to multiple sources or use a decentralized storage solution if you are concerned about data availability.

> You can replicate it to multiple sources

Who is this "you" who is going to do that?

Remember we were talking about property deeds? Their availability is of great concern.

Before you answer that, go back over all your comments here and in other threads, collect all the limitations you've run into, all the workarounds you proposed, all the additional external systems you need to run just the most trivial of scenarios, and riddle me this: why would anyone want this?

"federated systems, or a local computer at the owner's residence, or a hosted server controlled by the owner."

That's just centralised in a different form. Or more precisely - not decentralised.

Sure, some app can interact with centralised DB holding all actual information like deeds or music or whatever. Why do you need NFTs for that? Centralised BD must me utilised to track ownership of actual data or items by real people and it won't work any other way due to the IP rights and complex ownership agreements. NFTs are useless and impotent for this task.

> That's just centralized in a different form. Or more precisely - not decentralized.

How many copies of the data in different locations do you need to be considered decentralized? In fact, I'd argue you just care about having enough copies of the off-chain data that you feel well insulated from data loss, so I'm not sure that the off-chain data needs to actually be decentralized. The goal is to have the ownership proof (on-chain NFT) be as decentralized and easily auditable as possible.

> Sure, some app can interact with centralised DB holding all actual information like deeds or music or whatever. Why do you need NFTs for that? Centralised BD must me utilised to track ownership of actual data or items by real people and it won't work any other way due to the IP rights and complex ownership agreements. NFTs are useless and impotent for this task.

I feel like many of these conversations in this thread are focused on the current state of blockchain adoption today. I don't know of any government today that has adopted a blockchain to track property records at scale, but I do think that will probably come eventually. As soon as that happens, then you can have rules and regulations drafted around all these tertiary concerns including IP rights and complex ownership rights, potentially some of those codified in on-chain contracts, but some would remain as processes for the government or the courts. Putting property deeds in an NFT on a blockchain does not solve all problems that arise with property ownership, but it could improve the record keeping process in some areas.

> but I do think that will probably come eventually.

For it to "come eventually" you need to answer all these questions that you ignore or swat aside as irrelevant, or easily solvable by magic.

> As soon as that happens, then you can have rules and regulations drafted around all these tertiary concerns including IP rights and complex ownership rights

No. These things have to be solved first. because complex ownership rights exist today, and not in some fantastical future after blockchain has been implemented at scale.

> potentially some of those codified in on-chain contracts

"On-chain contracts" of course make zero sense even for the most trivial of cases. Much less the complex ones.

Because what you propose is replicating human-readable texts with programs in esoteric programming languages that get routinely "hacked" because even multiple audits cannot find vulnerabilities in them. And for what?

> Putting property deeds in an NFT on a blockchain does not solve all problems that arise with property ownership, but it could improve the record keeping process in some areas.

It "may improve record keeping" but produce an insane amount of complexity, inane and insane workarounds, friction and problems everywhere else. So, will not be implemented until those issues are solved.

In addition to my other comment:

> In fact, I'd argue you just care about having enough copies of the off-chain data that you feel well insulated from data loss, so I'm not sure that the off-chain data needs to actually be decentralized.

Remember how it all started? "The [government] staff still does what they do today, but they no longer need to maintain their own database and IT application."

And now they need to maintain "enough copies of the off-chain data" anyway and deal with blockchain.

User illiarian wrote a good answer for you, but I want to add one more point:

IP rights aren't just for property deeds, they are relevant for every single digital or physical creation made by humans. So let's take a picture as an example:

There are NFTs and associated CDBs which supposedly allow trading pictures TODAY. No need to wait for any government or laws or technical advances. Supposedly NFTs allow me to buy and sell pictures in a decentralised way right here and now. The problem is that they can't do it. There is technically nothing inside NFT which can deal with IP rights for the pictures. You can't transfer IP rights completely, you can't do it partially, you can't do it with multiple peers, and of course you can't do any combination of the above. So current NFT platforms either ignore it or do it old fashioned way with centralised accounts, Terms of Service and other centralised stuff tracked on the platform and not on chain.

No need to wait for some far future and government adoption, NFTs has already failed at the much more simpler task and have no paths forward from this failure.

> So again, no I did not say that all data is stored somewhere else.

This is splitting hairs. Actual data that is actually relevant to the property, is stored somewhere else. Which is ironic because elsewhere you stated: "but they no longer need to maintain their own database and IT application."

Oh look. They have to maintain their own database for the actual data

> The government would in this particular case.

Not in this particular case. But in all cases.

> Maybe in other cases it would be a company

So. You need central entities to verify and enforce rights to physical objects. You need central entities to store the actual data pertaining to objects and people.

So the point of blockchain in all this is...?

The blockchain is a piece of additive technology. It doesn't replace existing relational databases, nosql databases, object storage, etc. You can store data in any location you like (doesn't have to be a centralized location) and then verify it's authenticity later by recomputing the hash and comparing that to the hash stored on-chain in the NFT. Also, just because some data is stored off-chain does not mean you need to run your own infra. You could pay a hosting provider to operate it for you.

Public permissionless blockchains help in easing the verifiability of ownership records, transfer of ownership records, and auditability of transactions. In the digital goods you don't necessarily have to rely on a single third party like a government. Checkout Chia offer files (https://chialisp.com/offers/) for an example of this. You can post these offer files on Twitter, forums, text messages, emails, etc.

> The blockchain is a piece of additive technology.

To add it you need to show it produces more value than the problems it causes.

> It doesn't replace existing relational databases, nosql databases, object storage, etc.

Hm. Strange. In a comment elsewhere you literally claimed this, emphasis mine: "The staff still does what they do today, but they no longer need to maintain their own database and IT application."

> Also, just because some data is stored off-chain does not mean you need to run your own infra.

And the people will run the infra for you out of the goodness of their hearts? And of course the data that you store "somewhere elsewhere who knows where" will of course be easily available and accessible in perpetuity because blockchain, right?

Why is it that most NFTs in the world are now pointing to non-existent files?

> Public permissionless blockchains help in easing the verifiability of ownership records, transfer of ownership records, and auditability of transactions.

They don't. Moreover, they only work for the most trivial of cases, and make actual real-world scenarios hard or impossible. I've provide several of those, and you answer? "No one is claiming blockchain is helping in that situation and that's okay!"

> In the digital goods

We're not talking about digital goods.

Sorry, don't mean to be rude but this is the most useless discussion I've seen here.

I think it's fairly easy to understand (but you clearly hate the technology for some reason). The proof of ownership would be in the blockchain. NFT X represents X property, the address owning that NFT would technically own that NFT.

Also NFTs may also have descriptive data on them, so technically you could have a description marking that NFT as X property.

It's very much feasable, but from what I'm reading from you, you just hate the technology for the sake of it. Reminds me of Apple haters.

> he proof of ownership would be in the blockchain. NFT X represents X property, the address owning that NFT would technically own that NFT.

On top of that you need many more documents actually linking the actual physical person with that wallet. And additional documents dealing the actual property.

And of course "oh it's just simple record pointing to an address" is exactly why I keep saying that crypto proponents have this childlike inability to imagine anything beyond the most trivial of use cases.

Examples where "the address owning that NFT would technically own that NFT" falls apart and "useless discussion" is the one I'm having, not crypto-absolutists with their trivial cases and increasingly complex workarounds:

- ownership disputes: https://news.ycombinator.com/item?id=27212564

- ownership transfer: https://news.ycombinator.com/item?id=27217681

> you just hate the technology for the sake of it.

As wrote elsewhere: I've had this or similar conversation many, many times. Crypto proponents can only come up with the most trivial examples, and for anything more complex they rely on magic and increasingly complex and arcane constructions that fall under their own weight.

Edit:

Just in this discussion: we "don't need a centralised database", but we need a centralised database, or an external party to store our data that also needs to be available and accessible by anyone, but also reliable, maybe federated, or a local server, and it's also standardised, but maybe third parties will make it standard for governments, or maybe someone will provide infra to store data, or maybe it will be hash of data that may or may not be standardised, or...

And with all that it still doesn't solve or help any of the complex issues that arise from property ownership.

Again, you're calling me "crypto-absolutist" shows your clear hate for the technology only for the sake of it.

This would also make very easy to prove ownership. You have access to the address that owns said property, it's yours. It's as simple as that.

The database you'd need locally would be pretty much to store useful information and to track data on the blockchain.

I don't see any country using this anytime soon, or at all, but it is possible.

PS: I do understand you hate NFTs representing pictures of monkeys being worth millions of $, I also find it stupid, that doesn't give me the right to hate the entire technology that makes it possible.

> Again, you're calling me "crypto-absolutist" shows your clear hate for the technology only for the sake of it.

No, it shows that I define people talking about this tech by how much of reality they ignore.

> This would also make very easy to prove ownership. You have access to the address that owns said property, it's yours. It's as simple as that.

Who is this "you"? And who do you prove this ownership to? How all the other use cases like death and loss of access to the address are resolved? How are court orders are resolved? And so on, I've provided links where I listed various use cases in similar discussions.

See? Your "easy" ignores so many complexities.

> The database you'd need locally would be pretty much to store useful information and to track data on the blockchain.

So, you need to store useful information somewhere anyway. What exactly is the value of blockchain in this case?

> that doesn't give me the right to hate the entire technology that makes it possible.

Your emotional outbursts don't affect me. It's not hate when every single thing that you say is possible is only possible for the most trivial cases, makes other cases hard or impossible, requires complex workaround for complex issues, and in general doesn't improve an iota over existing situation.

> The NFT would be stored on chain and inside the NFT would be hash of the data that is stored off chain

Ok, but what's the benefit of doing that?

> as it stores and establishes cryptographic proof of ownership.

But if the entity that is charged with keeping track of who owns what needs to keep a central database anyway, what value does keeping the ownership records elsewhere bring? I see doing that as accomplishing nothing but increasing the number of things that can go wrong.

When it comes to things like real estate, the "proof of ownership" is completely without meaning unless you also have the extra data beyond that. Things like where the land is, what the boundaries of the property are, etc.

The hash literally changes nothing except for maybe verifying integrity of the data, and that is debatable. Add hash in the examples above and literally nothing will change. You can't just throw technojargon term at everything and see where it sticks.

The hash is critical as it is what enables you to verify that the off-chain data is correct. What exactly did you think was technojargon? I'll be happy to elaborate.

Why not just check the off-chain database and sidestep the hash stuff in the first place?

If the off-chain data is "incorrect", what are you going to do? Can anybody mint those NFTs? What's the point then? The government is the only one who can kick you out of that house irl.

Since the on-chain hash is part of an immutable object that no one can change, then anyone can use it to verify that the off-chain data is valid at anytime. (No one needs to go ask someone else whether it is valid they can check for themselves) If the off-chain data was found to be invalid, then a correct version of that off-chain data would need to be provided or a process would have to be developed that would handle exceptions where the document is no longer available and now a new NFT would need to be minted.

The minting process could be managed by the government or some entity using an on-chain contract or it could just be a manual process that the government does in conjunction with banks. The enforcement of the ownership records would still be held with the government. The point of having property records as NFT on-chain data is it enables easy to verify ownership of property records, transparency/auditability of property records, and easy transfers of property records to new owners.

> then a correct version of that off-chain data would need to be provided

Correct, which means that there has to be an ultimate source of truth that everyone agrees on that is independent of the blockchain. Which means there has to be an authoritative central agency of some sort to handle that. Which means that the blockchain has not, in fact, accomplished the stated goal.

The purpose of having property deeds represented as NFTs on a blockchain is not to eliminate the need for a governing body, but to make the ownership transparent, trivially verifiable, and easy to transfer. I've outlined this many times in this thread at this point. In the event, that a valid copy of off-chain data could not be recovered, then the governing body could use a longer and more manual process to identify the rightful owner and mint a new NFT.

Perhaps the difficulty is that how deeds are handled right now already makes ownership transparent and easily verifiable, and not much more difficult to transfer ownership of (because transferring the "token" of ownership isn't where the friction is).

So those don't look like things that NFTs make better in this space.

That's I wrote. Sure, it enables data integrity (maybe, but let's be charitable). Now what? How does the fact that token knows about external data prevents that data from being manipulated if needed? Or do you propose banning inheritances, lawsuits and other actions which apply to the real world artifacts just because blockchain can't deal with that? :) That won't happen.

It doesn't prevent the manipulation, but it does assist in detecting it. If someone manipulates the off-chain data, then when that off-chain data is hashed and re-verified against the hash stored in the NFT on-chain then there will be a mismatch in the hash values.

Let's suppose that the off-chain data is a PDF containing all the information for the property deed. If the hash values don't match then it means the PDF should be disregarded. If the off-chain data was kept in multiple locations, then one could potentially still retrieve a good copy of the PDF whose hash matches and then can be trusted.

Now, just because a blockchain can be used as the system of record for property deeds does not mean one wouldn't need to still have laws and other organizations enforcing that system of record. The blockchain is just a tool that makes some aspects of the record keeping easier and opens up some interesting possibilities as I've outlined in other comments of this thread.

We don't need detecting unwanted manipulation (clarification, we do, but that is a small and rare problem), we need to deal with expected manipulation. Let's say I was phished and some tokenbro across the globe now owns the NFT with a deed to my house, initially this info also got propagated into the centralised DB with stores actual data with deeds and related stuff. I go to police, then to the court and they reaffirm my ownership of the house. Then they restore correct information in the CDB. Now everything is all right, excep that blockchain is now outdated and show a tokenbro as an owner of NFT pointing to deed for my house, and hash is of course not matching now. Now what? Code is not law (surprise), and blockchain doesn't override laws. It is technically useless for this task.

It's possible to intentionally alter data that a hash has been computed from in such a way that it still computes to the same hash. Hashing alone is a weak way of ensuring that data has not been altered.

It is extremely hard to break cryptographic hash functions to the point it is considered impossible. If you could do that, then you could be breaking into all sorts of systems. Cryptographic one-way hash functions are actually the preferred way to secure passwords.

It isn't trivial, but it's not as hard as you make out. In the case of real estate which is often worth a great deal of money, plenty of people will be willing to put the effort in to do this.

> If you could do that, then you could be breaking into all sorts of systems.

Absolutely. And this is one of the methods that is used to do just that. I'm not talking about a theoretical security issue here, this is a weakness that has been leveraged in the real world for a long time.

> Cryptographic one-way hash functions are actually the preferred way to secure passwords.

Yes, but they're also not considered bulletproof. They're a bit of a compromise effort. That's why the leaking of password files is considered a security problem even when the passwords are salted hashes.

And, like with password hashes, it's not actually necessary to break the hash in order to alter the hashed record while maintaining the same hash. There are mathematical shortcuts to doing this, but you can even just brute force it if you have enough computing power or time.

> Absolutely. And this is one of the methods that is used to do just that. I'm not talking about a theoretical security issue here, this is a weakness that has been leveraged in the real world for a long time.

Oh really? Care to share some examples of SHA-256 and SHA-3 collisions?

Not sure what would count as examples here, honestly. The ones I know of were what I encountered when I was working in cryptographic security, and I don't think any of those incidents were made publicly available.

That said, there are readily available tools that use things like rainbow tables to "crack" SHA-256 salted password hashes. By "crack", I mean to come up with a password that hashes to the same value.

These tools are in successful use every day.

Yeah, I didn't think so. :) I know of no single incident of SHA-256 or SHA-3 having collisions successfully generated.

Cracking SHA-256 with rainbow tables is a fundamentally different exercise as you are relying on someone having selected a weak password that you can then generate a hash for. The weakness is not in SHA-256, but in the weak user selected password.

> I know of no single incident of SHA-256 or SHA-3 having collisions successfully generated.

That's fine, I don't expect you to believe me without evidence. But I have seen this happen more than once in my work.

> you are relying on someone having selected a weak password

Stated another way, you are relying on having some idea of what the original data looked like, so you can reduce the search space. Absolutely correct.

However, if you're hashing public records like real estate, where you literally have the clear text, that's a much simpler problem than cracking passwords. All you need to do is alter the document in the way that you want, then find what other changes are needed to create a collision with the original hash. This is not very computationally intensive compared to password cracking.

Add in that the amount of money on the line with real estate can be high enough that it would make it worth throwing serious resources at it -- more than the average password cracker could even begin to summon -- and my confidence in the security of the hashes is greatly reduced.

Here's an interesting general overview of the problem: https://medium.com/asecuritysite-when-bob-met-alice/can-i-cr...

Reply on news.ycombinator.com