> don't know if there is any real way we can go back to the P2P paradigm without destroying NAT I think it's time for NAT to go for residential ipv6 and the n

> don't know if there is any real way we can go back to the P2P paradigm without destroying NAT I think it's time for NAT to go for residential ipv6 and the numbers I've seen show that TURN isn't required for most connections. Unfortunately universities and businesses will probably never remove NAT as there is limited incentive to do so. I think the best we can do is have somewhat decentralized networks with limited yet trusted centralized authorities (the need for discoverability will always remain, even when using otherwise decentralized networks like SSB). This could be IPFS with bootstrap nodes for their DHT, or as I have been using to circumvent NAT when latency is unimportant, Tor directory authority to host ephemeral, local onion services.