-Yes, have you considered the counter scenario where there are multiple chain analysis companies in a free market with different reputations for accuracy and honesty?
Yes. Do you pay multiple or one such company?
I don't run a coordinator.
-Yes, I accept this as a drawback. If a coordinator's data provider produces false positives, the coordinator will likely stop using that data provider.
For the coordinator to verify false positives, it means the chain analysis company must share their techniques and conclusions in transparency. Do you get such information from the paid company?
If the coordinator has verified false positives, it really doesn't matter what the chain analysis company's techniques are because they didn't work.
If they are merely reading their own copy of the blockchain, just like any other Bitcoiner is able to do, then they are just a data analytics company.
That's the definition of surveillance:
https://dictionary.cambridge.org/dictionary/english/surveillance. The fact that the ledger is public is irrelevant. I don't expect from regular Bitcoin users to analyze my data. However, I do expect myself to be surveilled from said companies.
If you don't expect your transaction data that you broadcast to everyone else's computer for permanent storage to not be analyzed by anyone, you're just naive. The solution to your problem of public transaction data exists already: WabiSabi coinjoins are purposely designed to make the data you broadcast useless to any analyzers.
-Yes, but I think it's easy to realize that this is outweighed by the surveillance that is made impossible as a result of the coinjoin.
If the chain analysis company has blacklisted certain inputs, doesn't this prevent not-yet-surveilled users from having their privacy preserved?
Did you mean "doesn't this prevent already surveilled users from having their privacy established?"