[Zaskoda]

> Where to even begin with this? First of all, many local governments in the U.S. already have public databases of who owns what land. This is a problem that can be solved with a SQLite database, it doesn't need a blockchain.

I believe you're missing the opportunity that exists for automation. That database is heavily guarded by layers of human administration. I can't sell you my land without involving a bunch of humans who interpret contracts and push buttons. In a world where land ownership were recognized by tokens on a distributed ledger, buying and selling land no longer requires the overhead of all that administration.

[tdeck]

> That database is heavily guarded by layers of human administration. I can't sell you my land without involving a bunch of humans who interpret contracts and push buttons. In a world where land ownership were recognized by tokens on a distributed ledger, buying and selling land no longer requires the overhead of all that administration.

Those layers of human administration are called a "state". Without that, the concept of property is pretty much meaningless to begin with. I can claim to own any piece of land I want. The state pays people to show up with weapons if I'm using land that they've decided "belongs" to somebody else. It gets to set the terms of land ownership and to insert itself into every transaction because of the ability to use force, not because it has the most efficient record-keeping system.

What that means is that the fundamental premise of blockchain technology, that there's no need for an external authority, is completely at odds with the reality of how property works. If there's a smart contract that says I own parcel X, but the people with guns say I don't own it, who wins that argument? That means the state needs a way to insert itself into and modify the blockchain, but that kind of obviates the whole point of the blockchain. If the state can't alter it, it's not a meaningful source of truth.

That of course leaves the benefit of automation. But all of that automation could be done with a SQL database and a PHP web app. The reasons it isn't done that way are policy ones, not technical ones.

[robjan]

What happens when the owner of the NFT representing their land loses their private key or they die without passing the key to someone else? There are only two options: either the land exists ownerless forever or the blockchain is hardforked which would undermine the whole system.

Blockchain in general doesn't really work well when it has to interact with the physical world.

[gus_massa]

Let's spouse I have two properties, one big and one small, and I sell you the small one for $X. But I miskick and send you the big one. If you don't want to return it to me, can I go to a judge?

What about kidnaping? If someone kidnaps me and use a wrench to send my property, can it be returned?

[tdeck]

What about the very unlikely event that someone loses access to their private key or gets hacked? Does that mean their house now belongs to someone else? I suppose this speeds up the ability to lose one's home through cryptocurrency scams.

[altrum]

Author here; I agree, and this is something I've been wrestling for years. The lack of centralization means it's impossible to reverse hacks or scenarios you mentioned--I personally think _some_ centralization is crucial, but not sure to what extend (and if that basically just means we should just have private companies like we do today).