The problem I have grokking is the actual practical issue of digital ownership, which is that it is nonsensical.
In this context ownership merely defines who has the right to be the seller in a transaction. As far as I understand this, blockchains don't handle the issue of duplication, where it doesn't matter who has the rights to begin a transaction - anyone can still access the bits.
For something like currency which is entirely defined by its ability to be used in transactions then the application of blockchain to define ownership is obvious. For something like media, where ownership is not just transactions but also access, it isn't obvious to me at all.
In other words, blockchains and smart contracts handle authenticity of a piece of data when used in transactions. That's valuable for data that is defined by its ability to be transactional. It's not necessarily valuable for data that needs to be more than that.
I would also contend that it's not secure by all definitions.
In the context of blockchain tokens, “digital ownership” is equally as nonsensical as owning a fine art print with ink and a pencil signature on it. These artefacts, and the ideas of ownership around them, are social constructs.
The mental shift that most struggle with, is that the NFT ownership is not owning the digital image/file/media/bits (which can be copied on any machine, “Right click save as”). It’s probably closer to purchasing conceptual art (see Sol LeWitt), or similar to owning a certificate of authenticity handed to you by the artist. Or, perhaps somewhat analogous to “owning” digital property such as a domain name.
I agree this tech does not magically make everything more secure. But, for example, you can query the “current owner” of a million dollar CryptoPunk - as long as the blockchain survives, the database record will be secure & immutable, and cannot be changed except from the token holder’s own wallet address.
Reading the Solidity code & querying the contract myself helped me understand the value here is not in the digital media file; it’s in the ledger record (or state).
I think you miss the forest for the trees in my comment, which is attacking the semantics of "ownership" of a digital asset, which in blockchains corresponds to "the ability to give ownership to someone else" - or the only kind of ownership that a blockchain provides is the ability for one person to transfer the ownership.
For some assets like currency this is sufficient - currency is only valuable by virtue of the fact it can be transferred.
For art it is not sufficient, since ownership must also cover the ability to be transferred and the ability to view the artwork.
While all of it is a social construct, the semantics of those constructs are very different. My argument here is that blockchains and smart contracts are not sufficient to represent ownership beyond its transfer, which makes it useless as a mechanism to define ownership for assets that need to be more than transferred.
Perhaps this is a narrow view of ownership that will be expanded by NFTs (or, perhaps not - let’s see in a couple years).
I don’t see it all that different than purchasing and therefore owning a conceptual Sol LeWitt artwork, ie: a few written instructions on a scrap of paper, that (eg: if it is lost) can be simply re-written to restore the work.
Your requirements (ability to transfer and view an artwork) are met by many on-chain projects, eg: see Autoglyphs.
In this context ownership merely defines who has the right to be the seller in a transaction. As far as I understand this, blockchains don't handle the issue of duplication, where it doesn't matter who has the rights to begin a transaction - anyone can still access the bits.
For something like currency which is entirely defined by its ability to be used in transactions then the application of blockchain to define ownership is obvious. For something like media, where ownership is not just transactions but also access, it isn't obvious to me at all.
In other words, blockchains and smart contracts handle authenticity of a piece of data when used in transactions. That's valuable for data that is defined by its ability to be transactional. It's not necessarily valuable for data that needs to be more than that.
I would also contend that it's not secure by all definitions.