Sharing the same IP address as other devices doesn't improve usability, it actively hampers it by breaking the end-to-end principle. If you care about using peer-to-peer applications (voice calls, online gaming, torrents, Bitcoin, Nostr, etc.) without requiring the use of middlemen/relays that are publicly reachable over IP (thereby introducing other issues such as natural centralisation pressures and additional latency), then you care about maintaining the end-to-end principle. If you don't care about P2P apps, that's your prerogative, but just know that such an architecture is significantly responsible for the current state of the internet economy (among other things, of course). It can also actively harm usability by lumping you in with bad actors that happen to be sharing the same IP address as you. What do you do when someone on the other side of your neighbourhood is using the same IP address as you and does something that gets that IP address blocked by a service that you want to use? Sharing an IP address also doesn't give you any extra privacy, because (1) NAT on a per-household basis still identifies the household, (2) where CGNAT is used, NAT mappings on a per-neighbourhood basis are still logged by your ISP, and (3) the actual endpoints you're communicating with over the internet can fingerprint you using other means anyway.

Wait, who said something about usability? Nor I'm saying than NAT by itself is the way to achieve full privacy, but it's just a good thing to have, privacy-wise. By the way, IPv4 and NAT are not responsible about the internet moving from decentralised protocols to centralised platforms. 'User convenience' and tech companies are much more responsible for that outcome, imo.

Regarding P2P filesharing: * NAT has not stopped it to happen. * Moreover, sharing an IP address has protected a P2P user from legal consequences (I'm Spain, at least):

Abogacía Española

Around with the IP and with the IP

Consejo General