After all the scams of the last few years, I think that in 2023, folks will start approaching these projects more cautiously, only investing when the thing actually delivers some value.

I don’t know what a “valuable” (in terms of what it does, not token value) really means in web3 yet, though it seems like the building blocks are there.

I really want to see a web3/crypto project that has value outside of pure speculation, when we will see that (if ever) is what I’m not sure about

You have a misplaced faith in people. There is absolutely nothing of societal value in Web3, and the fact that the author of the article can claim otherwise with a straight face shows that enough people haven't understood this yet.

Whether you put the source of truth in a centralized database or a distributed ledger controlled by the crowd, the ultimate arbiter is the rule of law. As soon as you accept this, the only correct engineering solution to any problem of tracking ownership is the one that is the fastest and lowest cost (per transaction, to evolve the spec for new requirements including compliance with new laws, to roll back failures, etc.), and that rules out distributed blockchains.

But with "web3" so many projects had insane usage but only because the promise was get-rich-quick. The product itself had no value.

The usual signal was distorted.

90% of helium access points will be installed right next to other helium access points in places that are crowded with early adopters. If there are a lot of them they will interfere with each other in some places, but almost everywhere there will be no service.

If there wasn't network planning as in a cellular network you would need very strong market based mechanisms such as people getting paid 100x as much to put up access points in areas without coverage, being charged to join the network if they are in a place that has too many access points, etc. Even cellular networks have awful economics if you demand 100% coverage, they look cheaper than fiber when you have 70% coverage but somewhere in the 90%'s they become expensive again. Big Telecom wishes it could get telematics hardware in all automobiles to add hundreds of millions of subs but it does not have the desire to make the investment to get sufficient coverage for those things to really work.