It is pretty cool to look at a #payjoin transaction with a block explorer and see that it completely breaks the "common input ownership heuristic". When I setup my #btcpay server I tested it with Wasabi which was one of the wallets that supported payjoin sending at the time (there are quite a few wallets that support payjoin send now). It works but of course with payjoin v1 you need to have a publicly accessible btcpay server to receive. The new payjoin v2 sounds like it's much more accessible for non-server runners to be receivers.