I don’t appreciate it. It broke so much of the original end-to-end promise of the Internet. Think of how many technologies would be vastly easier if two hosts could directly communicate, like audio and voice chats. I still miss the days when it was trivial to run a web server on your home computer if you wanted to. And now we have abominations like CGNAT that break lesser abominations like UPnP.
If we hadn’t had NAT for the last couple of decades, and someone invented it now, they’d be laughed off the stage. You see it as an enabling technology. I see it as a boat anchor that’s kept us on IPv4 way past the sell-by date.
NAT has done more than the promise of IPv6. As much as NAT sucks to have at least it's been mostly frictionless to transition to -- that can't be said about IPv6. The design of IPv6 is what has held it back for 25 years, and I don't think we'll ever see a proper transition to it, so I'd say embrace NAT.
Hard disagree. The very first NAT broke a lot of cool things that already existed. Again, think of all the serverless things we could’ve had, like truly end-to-end peer-to-peer messaging that wasn’t routed dispatched through a central routing platform. (We’d still need a locator broker, but that’s far different than piping all content through a third-party host). NAT ruined those entire concepts because the whole concept of “every host is a potential server” went out the window.
Without NAT, we’ve have already made the transition to IPv6 as the creaky old IPv4 wouldn’t have that critical bandaid that helped it limp along. IPv4+NAT can’t die quickly enough.