> This is why I think the really valuable and underserved use case of the blockchain is decentralized identity. You can prove you are who you say, you’ve studied where you claim, you’ve worked at the places on your resume, and do this in ways that cannot be subverted or lost. This would be invaluable for refugees who often struggle for months or years with proving they are who they are.

That's a very interesting use case, but it's hard for me to see exactly how this can be made to work.

Suppose you study at the National University of Unstabilia, which is located in a disaster-prone and conflict-riven environment. You complete your B.A. there, and you get the NUU to record this fact on a public blockchain.

A few years later, things are really bad in Unstabilia, so you move to Belgium. After you arrive there, you tell someone (maybe a prospective employer?) "hey, I'm Joeri, I'm a refugee from Unstabilia, and I have a B.A. degree!". For some reason this person is skeptical, so you say "it's OK, just look up the blockchain record with the following hash!".

Sure enough, the public blockchain contains an entry reflecting that someone named Joeri did, indeed, earn a B.A. at NUU a few years back. This is great, because maybe

* Unstabilia City was mostly destroyed in an earthquake, making it hard to contact people there, and many of the people who would have known you during your studies have likely died or become refugees themselves; and

* Lately, the new NUU administration really hates your ethnic group, so much so that it prefers to deny that people of your ethnicity were just recently widely represented among its student body; and

* Many of NUU's records were previously lost in a fire; and

* Before that, someone reputedly hacked NUU's computer systems and stole all of their records, and probably all of their cryptographic keys.

But thanks to the blockchain records, your new Belgian friends can still confirm that you actually studied at NUU, right?

But, how do they know that that record is really from NUU? How do they know that NUU really exists? How do they know what its signing keys were, and how long they remained under the university administration's control? How do they know whether it's a legitimate university? And, maybe most significantly, how do they know that you're the same Joeri who earned that degree back in the day, as opposed to some other Joeri? Are these records including some kind of digitally signed biometrics?

> But, how do they know that that record is really from NUU? How do they know that NUU really exists? How do they know what its signing keys were, and how long they remained under the university administration's control? How do they know whether it's a legitimate university? And, maybe most significantly, how do they know that you're the same Joeri who earned that degree back in the day, as opposed to some other Joeri? Are these records including some kind of digitally signed biometrics?

Asking blockchain to solve those problems is a bit ridiculous. Those are problems that need to be solved in any system, and are solved enough in many today. For starters, its not hard to archive your signing keys somewhere safe and public, especially on the blockchain - the group of Universities and employers who care about that validity will have some central organization in identifying that archive.

So now remove the blockchain entirely and what value was lost?

This is what the article demonstrates. All the value is in the trusted authorities issuing things, not the transaction record on a blockchain.

Trust is important and trustless transactions with pseudo anonymous entities are not worth much.

The issue being discussed is putting college degrees on the blockchain such that viewers can be sure they are genuine and robustly hosted without tampering - no revocation.

The blockchain solves the last two, but if your conception of them is as a magical technology that can solve every issue by virtue of hosting data then you're going to be a dissapointed simpleton.

Your core issue is that colleges are a centralized institution which decide who gets rewarded - that's what it boils down to when you say "all the value" is in trusted authorities issuing things. For starters that's a ridiculous assumption that trust is still necessary for value, but more importantly stating that blockchains are useless because they cannot replace colleges is disingenuous.

My problem with blockchains as the proposed solution here is that they solve none of the hard problems, introduce some new problems, (and no they are not an irrevocable record (as if that were even desirable), look up the DAO Hack or Bitcoin Cash fork and they certainly aren't proven to be permanent or reliable) and removing them would make the solution simpler and cheaper - the essential problem here is trust, not recording and sharing data.

You have not demonstrated any added value, and the straw-man insults sprinkled with spelling mistakes do not help persuade.

Yes I am straw-manning when you've built your original critique off a single niche use case (I believe the progenitor even used the phrase 'what if') and cite failing projects at least attempting to innovate as evidence of the uselessness of a technology which has achieved its original goal and continues on.

Bitcoin is a dismal currency.

It has thus failed at its original goal (a useful currency to rival state backed currencies).

If they can verify that, how much does the blockchain add then?

Well the example clearly stated an issue of redundancy. Things which can be done off chain which as little trust should be done off chain - that doesn't mean a distributed file storage protocol which runs off some chain and uses economic and cryptographic incentives isn't the solution.

I have an idea! Why not create a _second_ blockchain which verifies the identity of NUU? :)

None of that requires a blockchain. The same level of investment in digitization to get all of this info on to a blockchain could be used to publish it to the cloud, secured and authenticated by cryptography.

The blockchain gets you exchange, with completely transparent meditation, in the form of the smart contract / script code.

> In regions with unstable governments you cannot rely on the government to keep saying that your house is your house.

You can say it's your house all you want, but if the new regime sends soldiers to evict you, no amount of evidence that it belongs to you is going to help you.

I don't really see how that works.

Do you propose that "all" authorities provide digital certificates, in preparation for the region becoming unstable? If yes, paper certificates already exist, and seem to go missing -- why would it be harder for digital certificates to go missing? Or for the thing that ties one person to their digital certificates?

Or do you propose that authorities in unstable regions provide digital certificates? If yes, how can you trust them, given the unstable nature?

I value thinking about these things, but somehow I still struggle to see where the proposed extra value comes in. Maybe I'm thinking too much in extremes, and the value breaks down in extreme cases.