3 ms·

I know NFT disgust should be an automatic reflex but chains of title are probably one of the few areas where a publicly accessible, append only, blockchain ledg

I know NFT disgust should be an automatic reflex but chains of title are probably one of the few areas where a publicly accessible, append only, blockchain ledger is a clear win for everyone involved.

owenmarshall 4y ago
Why? The entire purpose for a title is that you went to a large central entity with guns - the state - who will enforce your claim to property, with those guns if needed. There are precisely zero competitors who can issue an alternate title. The state definitely uses those guns to ensure that, because not doing so is an existential threat. There is one issuer, trusted universally. That problem is solved by a centrally maintained database, not a blockchain.
- Veliladon 4y ago
  There's not one issuer. In the US there are 51 issuers of car titles alone (50 states + DC) not counting the territories. Plus while the title is signed but not registered with the new owner it's a bearer instrument. People wash titles by taking them out of state because there's no effective way of binding both title histories together.
  - owenmarshall 4y ago
    > there's no effective way of binding both title histories together. The DOJ and the NMVTIS would beg to differ. https://vehiclehistory.bja.ojp.gov/nmvtis_states https://vehiclehistory.bja.ojp.gov/nmvtis_states
  - joecool1029 4y ago
    There is a national system for reporting title information that every state except Hawaii participates in: https://vehiclehistory.bja.ojp.gov/nmvtis_states https://vehiclehistory.bja.ojp.gov/nmvtis_states
  - wolrah 4y ago
    Again though what benefits are offered by adding the complexity of a blockchain compared to simply having a shared central database? Either way to be effective you have to get all the parties involved to agree on using one system, that's the actually hard part, the technical side is trivial by comparison. The only situations where blockchains really make sense are when you have multiple parties involved who don't trust each other AND are unwilling to agree to trust a third party. Digital currency is basically it. Almost anything else is better served by a traditional database or append-only log. Certificate Transparency provides a good example of a more efficient way to do things when you're willing to at least partially trust certain authorities.
    - AdamJacobMuller 4y ago
      The only benefit I could see is allowing title transfers without the involvement of the government. Do you think they will give up that kind of control? I agree a public CT-style log is far more appropriate.
    - syntheweave 4y ago
      That's the right argument for the wrong customer. The end user is supposed to trust the government to act as one - that's the point. But a government is always in a state of partial synchronization with respect to record-keeping. That's why there are delays and loopholes in the system. And the naive answer of "just trust the one database bro" immediately creates gridlock: for the number of services that government provides, all pragmatic answers instantly turn into Byzantine General negotiations: who hosts the servers, who gets access, what kind of access do they get, whomever has the most stringent uptime and disclosure requirements will push the others into adopting the same, and so on... On the other hand, if you say "it's append-only and distributed at the top level and every agency can put what they want on top of that, add new nodes at will, etc.", you have a blockchain. It redefines the desynchronization problem in a way that narrows the scope of the technical/political crossover. Negotiations still take place, unnecessary duplication and desynchronization still happens, but basic issues around where the global source of truth "is" are no longer a factor. I don't think the effort described is even a perfect match for the tech, but it's the kind of "wrong in the right direction" needed to move forward in studying the issue.
  - antifa 4y ago
    > In the US there are 51 issuers But only one of those can legally issue California plates.
m-p-3 4y ago
Nothing that can't be solved more efficiently by making the data publicly available via traditional means. What would the blockchain be the superior choice here, since the DMV would ultimately be the only one with "write" access to the whole thing? Why would any other nodes host that blockchain if they have nothing to gain from it?
- Veliladon 4y ago
  If there were a unitary source of car title, yes. But we have 51 issuers of car title in the US.
  - wewtyflakes 4y ago
    These issuers are not in an adversarial trust-free environment, also, it seems unlikely that California will convince 50 other issuers to use NFTs to issue titles.
low_tech_love 4y ago
And who pays for the proofs of work?
- f00man2 4y ago
  There is no proof of work, Tezos is proof of stake. If you mean who pays the gas then it could be the state but Tezos is super cheap.

owenmarshall 4y ago

Why? The entire purpose for a title is that you went to a large central entity with guns - the state - who will enforce your claim to property, with those guns if needed. There are precisely zero competitors who can issue an alternate title. The state definitely uses those guns to ensure that, because not doing so is an existential threat. There is one issuer, trusted universally. That problem is solved by a centrally maintained database, not a blockchain.

Veliladon 4y ago
There's not one issuer. In the US there are 51 issuers of car titles alone (50 states + DC) not counting the territories. Plus while the title is signed but not registered with the new owner it's a bearer instrument. People wash titles by taking them out of state because there's no effective way of binding both title histories together.
- owenmarshall 4y ago
  > there's no effective way of binding both title histories together. The DOJ and the NMVTIS would beg to differ. https://vehiclehistory.bja.ojp.gov/nmvtis_states https://vehiclehistory.bja.ojp.gov/nmvtis_states
- joecool1029 4y ago
  There is a national system for reporting title information that every state except Hawaii participates in: https://vehiclehistory.bja.ojp.gov/nmvtis_states https://vehiclehistory.bja.ojp.gov/nmvtis_states
- wolrah 4y ago
  Again though what benefits are offered by adding the complexity of a blockchain compared to simply having a shared central database? Either way to be effective you have to get all the parties involved to agree on using one system, that's the actually hard part, the technical side is trivial by comparison. The only situations where blockchains really make sense are when you have multiple parties involved who don't trust each other AND are unwilling to agree to trust a third party. Digital currency is basically it. Almost anything else is better served by a traditional database or append-only log. Certificate Transparency provides a good example of a more efficient way to do things when you're willing to at least partially trust certain authorities.
  - AdamJacobMuller 4y ago
    The only benefit I could see is allowing title transfers without the involvement of the government. Do you think they will give up that kind of control? I agree a public CT-style log is far more appropriate.
  - syntheweave 4y ago
    That's the right argument for the wrong customer. The end user is supposed to trust the government to act as one - that's the point. But a government is always in a state of partial synchronization with respect to record-keeping. That's why there are delays and loopholes in the system. And the naive answer of "just trust the one database bro" immediately creates gridlock: for the number of services that government provides, all pragmatic answers instantly turn into Byzantine General negotiations: who hosts the servers, who gets access, what kind of access do they get, whomever has the most stringent uptime and disclosure requirements will push the others into adopting the same, and so on... On the other hand, if you say "it's append-only and distributed at the top level and every agency can put what they want on top of that, add new nodes at will, etc.", you have a blockchain. It redefines the desynchronization problem in a way that narrows the scope of the technical/political crossover. Negotiations still take place, unnecessary duplication and desynchronization still happens, but basic issues around where the global source of truth "is" are no longer a factor. I don't think the effort described is even a perfect match for the tech, but it's the kind of "wrong in the right direction" needed to move forward in studying the issue.
- antifa 4y ago
  > In the US there are 51 issuers But only one of those can legally issue California plates.

m-p-3 4y ago

Nothing that can't be solved more efficiently by making the data publicly available via traditional means. What would the blockchain be the superior choice here, since the DMV would ultimately be the only one with "write" access to the whole thing? Why would any other nodes host that blockchain if they have nothing to gain from it?

Veliladon 4y ago
If there were a unitary source of car title, yes. But we have 51 issuers of car title in the US.
- wewtyflakes 4y ago
  These issuers are not in an adversarial trust-free environment, also, it seems unlikely that California will convince 50 other issuers to use NFTs to issue titles.

low_tech_love 4y ago

And who pays for the proofs of work?

f00man2 4y ago
There is no proof of work, Tezos is proof of stake. If you mean who pays the gas then it could be the state but Tezos is super cheap.