I completely agree about the app level. Any popular open-source client could quietly add a censorious cache layer and try to sneak it past the eyeballs of reviewers. And any "single client" users would be at risk of never hearing about this news if they've already installed the censorious update.

You're correct there that this is a threat to all software everywhere. That's the level playing field of "threat" that everyone tries to rise above. And in my mind, the shared game then is to try and add features and positive user experience above that line **without** adding new threats unique to one client or service.

The tradeoff is going to be different for everyone, and given the absolute shit state of the rest of the internet, and the potential for open protocols and censorship avoidance, **my** tolerance in that trade-off is very strict.

I've said in almost every note: Primal can completely dissolve even this suspicion, and I do honestly believe they will eventually (I've seen them do similar before!), by allowing the user to bypass the caching server if they desire. I don't actually think Primal (or you) are bad actors, but I do worry that powerful bad incentives are guaranteed to manifest themselves with success and the best way to beat them is to tie your own hands behind your back before they ever appear. Once a project is on the wrong side of that curve (bad incentives accelerate faster than honesty can hold at bay), the game is over and everyone loses (except those with the investment exit). So you have to head it off early. And community pressure can help with that, if done right.

I _want_ to be doing that right - so I'm also disappointed to hear that you think this is bad faith from me. I'd like to correct it if possible, for everyone's sake.