99% of LN liquidity pools is controlled by CEX and other custodial liquidity providers.
It's that easy. All of them join with a money transmitter licence that means sharing data with chainanalytic companies.
In that regard Tor is vastly different from the LN. Both are theoretically passively surveillabkr, but due to the centralisation effects (liquidity for routing), LN loses a lot of the properties and while Tor is more of an egalitarian network opening channels is a social selection process.
Wow I didn't know it was that bad.
So basically even if many lightning nodes were sovereign, and therefore subsets of the network aren't easily mapped, then the large hubs could still "Know Your Channel". Is that the case?
But how could it ideally work if we assume all non-KYC lightning nodes over Tor only (or even an ideal mixnet) and users are running their own lightning (not trusting some 3rd party w/ their data)
It seems like this still suffers from channel closing attacks/ accidents revealing onchain the net balance.
That's the pseudo ideal network configuration.
Everybody running their own LN node. Which means a lot of Bitcoin locked up for enabling only small payments. Or you end up again with some Bitcoin OGs running big centralised liquidity pools, putting a target on them for doing money transmission without a licence.
The bigger the payment you want to make the more liquidity you need to route the transaction. There are ways to split transactions, but they come with their own failure rate.
LN theoretically means a great expansion of functionality. I wouldn't just give up on it. But it really isn't the solution many people propagated it to be in 2016/17.
