[5h]

You've set yourself up for a question: Please give an example, with a deeper-lever perspective, of a real-world use-case for (non-crypto) blockchain that isn't better solved with non-blockchain technology.

Edit to answer this question:

> What about this is intrinsically "dangerous nonsense"?

The article isn't about the tech itself, it's about the interfaces of the tech and the real world. The problems there are shameless shilling, deception & efficiency problems.

[z3c0]

And you've set yourself up for an answer;)

I think any adversarial system has something to gain from blockchain. When participants of a system are naturally pitted against one another due to being competitors (a la engaging in a zero-sum game), the solution is usually to agree to auditing by a third-party. However, this method is easily corruptible by replacing the auditors with biased parties[1]. This is not an uncommon problem. Try searching "biased auditors", and you'll get a wealth of literature and perspectives on the issue.

A lot of blockchain proponents are majorly against the idea of "private blockchains", but I'm not. I believe private blockchains hold a lot of promise as a way of auditing an adversarial system, in place of a corruptible third-party. Plus, auditors don't amount to much more than middle-men, and thus add an additional layer of complexity to the system, meaning more can go wrong. Redirecting the cost of auditors to maintaining a system hosted by the adversaries would result in a more lean system (by less nodes) with little chance of bias corrupting the audit process (because you can't inject former employees into a blockchain). Instead of the SEC having to oversee both the members of the network and the auditors to prevent trusts from forming, they would just need to monitor the blockchain updates.

What the technical implementation of this would look like would vary wildly by industry, and thus would require some brainstorming to ensure feasibility, but no such brainstorming would ever begin if the members of an adversarial system are convinced that blockchains are "dangerous nonsense".

Of course, such a technology would take a massive chunk from a very profitable industry, so I expect a lot of people (employed auditors) to be averse to the idea, but maybe those people need to do some soul-searching. When your job depends on the continued existence of a problem, you aren't really incentivized to diminish the problem. Rape-whistle companies sell less rape-whistles if people rape less.

[1] https://www.nasdaq.com/articles/ernst-young-auditors-to-pay-...

[FabHK]

1. How do you solve the oracle problem? Blockchain can only deal with things that are on the blockchain already. Any interface to the real world suffers the trust and bias issues you described. You're just shifting the problem ever so slightly. (That's why the only real application is crypto currencies so far, because those can be realised wholly within the blockchain.)

2. How do you incentivise expending the PoW work without cryptocurrencies?

3. Why not have one central/permissioned authority issue chains of blocks, and everyone that wants can verify them cheaply? That is millions of times more efficient than PoW.

[5h]

Short comment to say I appreciate the thoughtful response here, and that implications of public vs private blockchains in real world contexts is something I've not given enough thought to - thanks for the homework :)

[jedimind]

"it is the only known solution to the Byzantine Generals problem that doesn't involve a centralized authority"

Did you skip that one on purpose? How do non-blockchain technologies solve the the Byzantine General's problem without relying on a (corrupt) central authority? Address this before you make more demands.

[knorker]

How does blockchain solve the Byzantine General's problem without relying on a (corrupt) central authority?

Math or not, police & courts overrule them.

Actually, blockchain isn't even mathematically a solution to BGP, is it? You need to define your actors better, here.

A 51% attack means communication can be denied. And there's still the question of how does A know that B has seen that A has seen, that B has seen…

Am I missing something?

The fact that communication happens on the blockchain doesn't mean it's not communication.

I just read a couple of articles about blockchain and BGP, and I think they all misunderstood what BGP is.

I'm A, and see that B is "committed on the blockchain to attack, assuming A is too". I can then add my block that says "I see B commits to attack. I agree let's both do this. I'll attack if you attack".

Ok. So now B sees that. Do they attack?

If you think the answer here is clear, then you've misunderstood BGP.

[5h]

Good attempt at a deflection, I won't casually solve the very thing that is a great example of blockchain being a solution looking for a problem.

But, given the 'real world' context I tried to stress, do you think blockchain is a good solution to the variations of that problem in real world contexts?