One example of things that have changed is zero knowledge roll ups. When I first used ethereum, transacting with it was free and fairly quick. The QR code approach meant that it was easier, faster, cheaper and safer to transfer money with it than with my bank account. That changed dramatically for the worse once the systems started getting popular (and traditional banks upped their game too). With L2 systems like zksync, we're back to cheap and fast, and we don't have to give up the central security properties.

Another example of something that has changed for ethereum is Proof Of Stake. Some years ago, I worked for a large financial organisation, and we had a use case that would have saved a lot of internal work to move onto a public blockchain. Given the organisations eco aspirations, using a POW based system was never going to be acceptable. Now that ethereum is on Proof Of Stake, it's actually conceivable that these kinds of applications could be deployed by large financial institutions.

> we had a use case that would have saved a lot of internal work to move onto a public blockchain

If anything, large financial institutions have by now realized that they're not interested in public blockchains[0], one notable example as of late being the Australian Securities Exchange[1].

These technical improvements are a bit as if a salesman reduced the price of a product, on top of free delivery. That's nice, sure, but that doesn't mean you need the product any more than before. There are more fundamental reasons why smart contracts are not widely used, and they're not addressed by these technical improvements.

[0]: as defined in https://www.schneier.com/essays/archives/2019/02/theres_no_g... [1]: https://www.reuters.com/technology/australian-securities-exc...

My example was a situation where we would benefit from a public blockchain solution in terms of more efficient administration, but were unable to use it because of ecological concerns. In that situation we already needed the product, and the improvement removes a barrier.

> a situation where we would benefit from a public blockchain solution in terms of more efficient administration

That's what I'm doubting. Ecological concerns etc. are a problem, but in my experience public blockchain solutions fail earlier, at the stage where one asks "would there be any benefit" before asking what the costs might be.

Then we have different experiences.

The argument that use cases aren't developing is quite brittle when you dig into it a bit.

Take the use case of global stablecoin transfers. The ability to send cash point-to-point anywhere in the world is obviously valuable and novel.

But for global stablecoin transfers to be generally useful 1) everybody needs crypto wallets, 2) crypto wallets need to be increasingly safe and easy to use, 3) transaction fees need to be sufficiently low, 4) stablecoin on/off-ramps to integration with the traditional financial system must be widely available, and 5) stablecoins in a variety of popular global currencies must be available and liquid.

The industry has made progress in all five of the above areas in the past few years, and so good progress has been made on the greater goal of generally available point-to-point global cash transfers. Saying crypto "has no use cases" is lazy, uncurious, and intellectually bankrupt.

> Take the use case of global stablecoin transfers. The ability to send cash point-to-point anywhere in the world is obviously valuable and novel.

You're not sending "cash" though, you're most likely sending USDT or some other centralized stablecoin backed by a shady company ("stablecoin" is a very misleading term in this context given the counterparty risk). That's one point about trust being moved instead of removed, you need to trust the issuer to some extent. To this day I still haven't heard of a centralized stablecoin being fully audited, you merely get attestations, meaning that you can't treat them as "cash", you have a big risk just holding them.

Transferring / converting currencies from most countries to most other countries is easy and way cheaper via centralized means, e.g. https://en.wikipedia.org/wiki/Wise_(company) has been around for 11 years and was probably not first, and the fees are lower than what a blockchain transaction on BTC/ETH would cost. This is not even an apple to apple comparison, since in most cases you still need to convert your USDT back to USD to actually use them, incurring more fees in the process.

What ETH does allow you to do which you can't do with Wise and others is sending USDT without censorship. That's probably the best way to bypass US regulations if you want to send USD where the US government doesn't want you to. And some people suggest that this is how things should be, which is an interesting point. Whether you agree or disagree about the morals of it, it'll remain marginal compared with the amounts transferred via centralized entities for reasons listed above. In particular, this is incompatible with your 4.:

> 4) stablecoin on/off-ramps to integration with the traditional financial system must be widely available, and 5) stablecoins in a variety of popular global currencies must be available and liquid.

As soon as you integrate with the traditional financial system, you have the same legal requirements. But as explained above, skirting regulations is actually the only one thing that you could do more easily with Ethereum. Once you've accepted to deal with the law, you're better off using centralized platforms all the way. If I have EUR and want to send you USD legally, why would I go through USDT or whatever blockchain token and pay higher fees?

In this space, there's still very little in terms of products with actual impact, to the point that it's hard to tell what will be left once (if ever) the noise from the blockchain-related scams disappears.

I was debating these points 5 years ago already. The fundamental problems I'm mentioning now were already there (the decentralization of a blockchain-based system breaking at its boundaries with the real-world, the oracle problem). The hype has evolved, but retains the same fundamental flaws.

USDC is not nearly as shady. Nothing forcing you to use USDT.

Friends in Turkey were turning their savings into USDC and DAI to save themselves from their country’s near 100% inflation.

Crypto critiques need to take off the first-world Goggles.

Maybe your friends, but most people were buying the ordinary US dollars.

And store it in a local bank with the economy on the verge of collapse and the government being what it is? You have no idea wtf you're talking about sir.

No, just keep the cash.

Worst advice.

The second worst at least.

It’s not that easy to buy dollars in large quantities. At least in my country, there is a cap on how much foreign currency you can buy.

The problem is that Ethereum continues to be used almost exclusively for speculation of ethereum, followed by creation of literally hundreds of thousands of ERC20 cryptocurrencies, followed by anonymous lotteries/gambling (and for a brief time NFTs which themselves were a form of speculation, although that's mostly collapsed). At what point will that change? How much more do they have to do before Ethereum is mainly used for real-world applications? Ethereum is a solution shoe-horned into a trustless blockchain to an already solved problem.

While I'd certainly agree that Ethereum has problems, the problems you list aren't Eth problems. They are legal issues in some geographic jurisdiction.

Can you elaborate? Isn't the whole point of Ethereum's creation for practical applications, which they've continued to fail to have adoption for?

> and as each year passes with still no meaningful democratization of this tech, it's becoming doubtful that these limitations are the real issue.

I think it's pretty easy to see that each transaction costing a few bucks (or tens of bucks at it's peak) is a huge issue if you want more adoption.

Gas prices (determine costs of transactions) haven’t been too high lately, though there can still be spikes when transaction volume surges. Gas prices will become less sensitive to such surges once the mainnet is sharded (plus L2s) because the chain will be able to handle a much higher max volume of transactions at any given time.

https://www.gasprice.io/

A lot of the past few years has been spent on research.

The move to PoS is now complete and rollups are under active development.

For a lot of use cases scaling is a prerequisite and that's coming soon, likely 2-3 years of progressive improvements.

Transactions on rollups are as low as 2 cents and EIP 4488 is expected to bring that down about 100x.

You should really take a look at the roadmap: https://notes.ethereum.org/@domothy/roadmap

(This comment was originally a reply to https://news.ycombinator.com/item?id=33878419. Hence the "your conclusion".)