April 2021
Bruce Schneier has good comments in his post Blockchain and Trust which partly inspire or inform my comments.
I'm very honestly puzzled about the enthusiasm surprisingly many people have about blockchain technologies. I understand the technical aspects of it very well, and have for example, designed cryptographic protocols in my life (ones that have actually been used for practical purposes even).
Much of the enthusiasm comes from people who are distinctly less technical than I am, and I know to them many of the details are understood somewhat vaguely or metaphorically. However, it's not only that. A bit over a year ago, about the last thing I did before quarantine was attend the MIT Bitcoin and Cryptoeconomics Expo. Many of the speakers there were far more immersed in technical details than I will ever be, and often they presented various mathematical proofs of certain properties of novel variations on blockchain methods.
One thing I did NOT leave that conference with was the slightest idea about WHY any of these people cared about this stuff. I mean, yes, buying drugs on the internet is great (or even a mattress or a book, which I've bought with Bitcoin just because I could). And I personally own some fractional amount of BTC and slightly over 1 ETH, which have increased in price in the last few months by about a month of my income (I'm happy to have the money... if I sell before it crashes; but not a fundamentally life changing amount... obviously, I wish I had bought much more a few months ago).
I have a friend who is really quite brilliant in a social-sciences field, who is one of these enthusiasts (but not technical). When we first met a bit over a year ago, we chatted about their enthusiasm for creating a system to protect anonymity and combat censorship (and a few other attributes of the hypothetical protocol/software)... in other words, create something with exactly all the same properties/virtues as Freenet, but 20 years later and SOMEHOW "using blockchain."
Another (online) friend, waxed philosophical about blockchains and NFTs recently, purporting various good uses of them. One such purported use was to make a property deed an NFT. This is very much the same pattern I mentioned above of describing something that is actually worthwhile, then tacking the phrase "with blockchain" onto the end of it for no obvious reason.
There is NOTHING about NFTs that make them relevant to a property deed. A deed does not belong to anyone, it states that the named physical thing does. Everyone in the world can print out a copy of the deed to my house, and every copy is exactly as authentic as the one I have.
A deed—or a contract in general—can benefit from non-deniability. If I write a contract in which I stated I gift my house to so-and-so, those words are of value inasmuch as I cannot disclaim them without my disclaimer being refuted.
In the physical world, mechanisms like ink signatures, and witnesses, and notaries, and court judgements entered into record, act to reduce deniability. Cryptographic signatures serve the same function.
Yes, cryptographic signatures are one of about a half dozen cryptography primitives used in blockchain, but most of their use has nothing to do with blockchain, and hence nothing to do with NFTs.
In a concrete example, my final will is published at a world readable URL. That publication includes a digital signature, using a private key matching the GPG key widely available from key servers or at other public URLs. No blockchains or NFTs were used, nor would any be relevant. The simple existence of the document IS the (presumptive) non-deniability. I also included as part of the text the NYT headline of the day it was written, which strongly proves it was written no earlier than the purported date. Decently strong evidence of the "no later than" is as well contained in the logs of the widely used site where I published.
Let's look at the failure mode of my will. Or likewise if I signed a document saying "I give my house to so-and-so." A disputant could claim "David's private key was leaked and used falsely." The claim would have to be adjudicated, of course. But an NFT adds exactly zero additional protection.
The "first tweet" that Jack Dorsey assigned in exchange for millions of dollars as an NFT also requires a digital signature from Dorsey. A disputant could equally claim that Dorsey's private key was compromised, which might be adjudicated as true or false. The relevant evidence would be exactly the same sort as for my will, that is NOT an NFT.
The difference is only the possibility of pushing around those bytes on the blockchain ledger. If not for this weird status economy (which is IDENTICAL to Duchamp readymades), Dorsey could simply sign the rather vacuous statement "My first tweet belongs to so-and-so." That would do neither more nor less to affect "ownership" (for example, copyright) but it could not be sold, just copied verbatim with signature.