That alone makes it worth watching imo.
Not really. There are many things, which could be introduced in a no-fork way. However, it doesn't automatically make it worth watching, because guess what: if something requires no forks, and introduces new rules, which are not enforced by any old nodes, then it may be possible to make a transaction, which would be valid on Bitcoin, but invalid on some subnetwork. What then?
There is a reason, why for example Taproot is a soft-fork. In theory, you could make it a no-fork. By just committing to a signature, any additional code can be executed. So what? Old nodes wouldn't validate them anyway. And without having majority of miners on your side, stealing any "tapscript-over-public-key" coins would be trivial.
In exactly the same way, quantum signatures can be introduced without any soft-forks. And again: it doesn't matter, if new rules would be ignored by most miners, and any coin could be moved, by using a valid ECDSA signature, without any quantum data. Which is why a soft-fork is needed here anyway. And the same with "Shielded Bitcoin", or other "metaprotocols": they are unsafe in their current no-fork form.
So theyre trying to bring much stronger privacy to BTC without a soft fork or separate chain.
Technically, it is doable, just not in a way they are trying to make it. A better example of a working no-fork is
2P-ECDSA. Then, all old rules are preserved, but you can actually execute multisig on a single public key, without introducing Schnorr signatures, or other things like that.
Another good example is CoinJoin, where existing protocol is fully followed, but it is hard to guess, who owns what. Which means there are some solutions, where you can improve privacy in a no-fork way. But note one thing: in 2P-ECDSA, CoinJoin, and other similar tools, you don't have any rules, which are not checked by old nodes. They still validate a lot of things, and you don't have for example any Bitcoin signatures, which would be fake inside a CoinJoin network, or would steal coins from 2P-ECDSA, without knowing someone else's private key.
But if this actually works well, I do wonder what happens to the privacy coin narrative.
Nothing, because native privacy will be still more private, than any optional one. Even if you create a perfectly private scheme for Bitcoin, it would be limited to the small group of users, who will decide to use it. All other users will be unaffected, and will still for example reuse addresses, and compromise not only their own privacy, but also decrease the anonymity set for others.
Would you use shielded BTC yourself?
Probably not. Maybe if there will be enough poorly-protected test coins, it could be worth sweeping, to teach some people a lesson, why some schemes are unsafe. But well, today, even test coins are no longer worthless, so it is harder and harder to do it ethically, without stealing any coins.
And do you think Bitcoin should push further into privacy, or is that better left to separate coins?
It should have many features, which won't be there, because there is no consensus for that. For example, decentralized sidechains could be activated long time ago, and people could experiment with new things, without changing the base layer in any way. However, because we don't have them, they were replaced with centrally controlled federations, where you have to trust some multisig, controlled by developers, and they can be extremely fragile.
In the same way, we should have Merged Mining, implemented and supported by default, so that people could solo mine Bitcoin, and receive coins from other chains on a regular basis, while having a chance to hit a real BTC block once per lifetime or so. But instead, that idea is heavily criticized, and instead we have completely separated, and disconnected chains, where it is more profitable to mine and sell them on the spot, than mine Bitcoin directly.
There are many good ideas, which won't be there. Yet another example are "weak blocks". Maybe testnets wouldn't be even needed, if we would have them, but again: if there is no consensus, then some things simply won't happen. It is easier and much more profitable to release an altcoin, pump and dump it, and get BTCs out of it, than actually improve BTC from the start, and reach any consensus there.
