It's a digital certificate that says, "I own thing X". You can then sell/transfer that digital certificate to someone else.
For the NFT use case of which you speak, I think of it as a digital autograph. In the physical world, a famous author may sign 100 copies of a book that has a print run of 10,000,000; and those special copies may command a hefty price. A movie star may autograph a photo of himself. And so forth... N.b. that in all such cases, the scarcity is more or less artificial insofar as the marginal cost of autographing an inexpensive item is negligible.
Everybody understands how that works. Some people buy such things, some don’t—but everybody understands the concept.
The problem with these April Fools’ fNFTs is that they are a SCAM. Many of these fNFTs are not originally issued or authorized by the party whose name is associated with them. For example, there is at this time a Satoshi fNFT; and there are currently two different Last of the V8s fNFTs for sale, even though V8s has been inactive since November. Obviously, you are not obtaining any sort of “autograph” from those persons... Real NFT sites have similar problems with fraudulent offerings. Fortunately, the scam here only reaps fake BTC which, I presume, will disappear altogether at midnight UTC.
More generally, NFTs have many other use cases. For example, I think using NFTs to track real estate titles would be a great improvement over the current means of doing such things. If it were so, then when you bought a house, you would receive a blockchain transaction representing legal ownership of your house—just as today you receive a paper deed. The title could be passed on to another party in another tx, with a digital signature from your private key. Not your keys, not your house! For Bitcoiners, the concept should be easy to understand.
Simple electronic signature would solve their authentication 'problem'. It's legally recognised and well regulated.
Assuming that 'problem' even exists, which I doubt, because there would be thousand startups in that space by now.
Not so. Simple digital signatures do not solve the problem of atomically, transactionally passing a value from one party to another, with a transaction ordering that prevents double-spends (double-transfers). Satoshi invented the blockchain as we know it for the purpose of solving these problems; he applied the solution to currency, but it can also be otherwise applied.
If you have a simple digital signature on a deed passing ownership of a house, how do you prove:
- The time at which the deed was signed? (To at least the accuracy and precision represented by a blockheight.)
- That person making the signature did not first sign another deed transferring the house to another person?
Furthermore: How do you create an immutable record of the deed and its conveyance from one party to another—with an audit trail that can neither be forged, nor be altered after the fact?
Blockchains also provide a framework for helping to solve one of the biggest problems with digital signature systems such as PGP: Associating an identity with a key. (Obviously, blockchains do not magically solve this problem all by themselves...)
In short, the blockchain performs for a cryptographic digital signature many of the same functions for which pen-and-ink signatures require a system involving a notary public, a government recording clerk, etc.
Note: Legal terminology of “electronic signatures” is mostly insecure nonsense. Many jurisdictions accept as an “electronic signature” a form submission that has no cryptographic authentication whatsoever. Above, I refer to digital signatures in the cryptographic sense—which may or may not have any special status, depending on jurisdiction. Even a cryptographically secure digital signature does not in itself solve the problems that I have briefly sketched in the foregoing (or others I have not mentioned).
Obviously, laws and legal systems will need some time to catch up to the point where blockchains can be used to keep chains of title for real estate, etc.