How much technological progress can we really squeeze out of blockchain, though?
It’s my understanding that it is essential for cryptocurrencies, useful for certain niche applications, but otherwise just one tool among many suitable solutions for problems at hand.
Honestly, what is something blockchain could unequivocally do better than anything else for a large scale application that isn’t cryptocurrencies?
I think there's a lot of potential in "smart contracts", even though that phrase is kind of meaningless now a days.
A simple example of what can happen in the cryptocurrency world: a simple combination of Shamir's Secret Sharing [1] and the public/private key model leads to very easy to use escrow services.
For example, you can create a new wallet with 3-keys, and require 2-of-3 keys to transfer money out. Alice, Bob, and Escrow get the three keys. Anyone can transfer money INTO the wallet and check its value through the public key.
Under normal circumstances, Alice transfers money into the new wallet. Bob then gives Alice a real physical good. Finally, Alice and Bob provide two keys to allow the wallet to transfer money to Bob.
Escrow comes in if any disputes come up. Escrow provides a 2nd key to break any disputes. If Escrow sides with Alice, Escrow + Alice can together use the two keys to transfer the money back to Alice.
While if Escrow sides with Bob, they together can provide two keys to transfer the money to Bob.
True, escrow services do exist in the real world. And arguably, a "real" enforcement mechanism backed up by courts, police, and the government... are far more important than a silly crypto-game.
But the silly-crypto game is automatic, and way cheaper to invoke compared to a full court case. It should be seen as a way to provide escrow services at a cost far cheaper than one backed by courts and lawyers.
For example, "Escrow" can be an automatic email system that gives the 2nd key to Alice after 30 days (encrypted with Alice's public key, of course). Perhaps that's not sufficient for buying a car... but it'd be sufficient Escrow for your typical $20 video game purchase from Steam. As you can see, Bitcoin / Blockchain CAN automate things that used to take courts to invoke. Yes, its of lower quality, but almost any transaction can benefit from an Escrow service (even an Escrow service with default terms)
-------------
The main problem with this scheme (and probably why it isn't popular, even if its technologically possible...) is that BTC or Ethereum's price could change dramatically, even in just 30 days. Cryptocoins simply aren't stable enough for this kind of mechanism to be useful.
In this case, if both dispute, then Alice gets the money back after 30 days.
As I said earlier: this is insufficient for a serious car purchase. But it IS a sufficient mechanism for Steam Games (if "Bob" is Valve/Steam, then Bob can simply revoke access to that game).
Amazon and Ebay also work on this mechanism. Not on Blockchain... but these companies overwhelmingly side with the paying customer (in this case: Alice)... leaving the business owners / service providers at a disadvantage. But Ebay also has a customer-review system attached, so people can tattletale on each other if a particular customer becomes abusive.
---------
So in effect, you have a "cheap" default policy (in this case: Alice always gets the money back in 30 days by default) for the Escrow, but then you still have to create a review mechanism to handle the various cases of abuse.
If 30-days is insufficient, then 60-day or 180-day return policies exist as well. The point is, anything you can imagine can be the default escrow policy. You can even lock the money away forever by default (IE: FORCE Alice and Bob to work together)... if you really want to go there. Alternatively, the Escrow can have the policy of "We'll only email you the key once you win a court case", if they really want to integrate into the system of law / courts we already have.
That wouldn't be fully automated, but it'd be relatively effort from the perspective of the Escrow business.
Well, for one, Bob can see the BTC or Ethereum in the wallet due to the public blockchain. Bob can be assured that Alice has indeed sent the funds.
Furthermore, the Escrow service isn't actually holding any money. If you were to implement Escrow using today's banking technology, you have to give Escrow the money, and then Escrow gives the money to Bob (or Alice) pending the results of the automated service.
Ex: You can't implement an escrow service using ACH... not at least... without actually taking the money. But with Bitcoin or Ethereum, you can implement an escrow WITHOUT taking any money at all. Escrow can only give the money to Bob (or give it to Alice), provable through encryption.
In effect: Blockchain provides a way to divide the responsibility and prove it through cryptography. The Blockchain Escrow NEVER has access to the funds, throughout the entire transaction. And therefore, it is easier to bootstrap Escrow's reputation.
True, we can use well-trusted banks and courts to handle these transactions. Trusted Escrow services do exist. But my point is that Blockchain could make it cheaper and easier to do.
----------
I mean, provided that Bitcoin actually can stabilize, among other things. There are a lot of reasons to not do this with BTC or Ethereum, primarily because of the gross variance in price. But still, its a nifty concept.
Its not "change the world" level, but it is certainly a technological advancement in my eyes. Something that could allow new businesses (with lower-trust models) to accomplish things that previously took a lot of trust, and a lot of effort. (IE: Escrow services).
Also non blockchain Escrow can be quite cheap, for example transpact.com:
>Our flat charge means that your costs are normally limited to £2.99 per party for (GB) Pound transactions or €3.49 per party for Euro transactions – that really is it for transactions up to £10,000 / €15,000 !
That tends to be a problem with arguments about crypto/blockchain in general saving money - the alternatives can be cheap and buying/selling crypto costs a bit. It only really seems to come into it's own as a way to skirt the law when fiat transactions are banned by securities laws and similar.
It costs $25 to $35 to file a claim in small-claims court... and more if you go to the greater court system (with lawyers and such).
If you're worried about getting your $20 back, then its counter-productive to use the courts. As such, a cheaper automatic mechanism is just simply going to be more useful to any micro-transaction.
------------
Non-court mechanisms are the typical: A call to customer service for Amazon almost-always results in the money being returned. Same with Ebay. These are the mechanisms that can be automated.
Courts are needed to settle serious disputes. But the $5 battery pack that doesn't work right... or some other item that was damaged during shipping... that's something you might want to use the automatic escrow over.
The same incentive as selling on Amazon or Ebay, which overwhelmingly caters to the customer.
To attract more customers.
--------
Lets put it this way: its an automated, strongly provably, money-back guarantee. If you want your money back within 30 days, you can have it through this escrow service.
And I think a lot of companies would be willing to offer money-back guarantees on their services.
It’s my understanding that it is essential for cryptocurrencies, useful for certain niche applications, but otherwise just one tool among many suitable solutions for problems at hand.
Honestly, what is something blockchain could unequivocally do better than anything else for a large scale application that isn’t cryptocurrencies?