At some point, some low-hanging companies are just going to start running IPv6 only applications and deal with the fall-out of some nn% of their customer not being compatible. Those customers will complain to their ISPs, which will provide incentive to get IPv6 deployed (should have been done 10 years ago honestly at any credible ISP). Getting the number closer to 100% will make it easier for new companies to deploy IPv6 only, until we get to the point that IPv4 is seen as some legacy thing that isn't relevant but used to be important in the dark ages of Web Deployments.
I'm very surprised at work and at home how often most of my Internet communications is running over IPv6. There are some surprising exceptions though [1].
I'd be willing to place a wager that there will be a $100mm company that is only IPv6 by 2030. And that the "Default IPv6 only on the Internet unless there is some weird edge case" will occur by 2040.
IPv4 as a protocol, though, will never, ever, go away - we'll see it widely deployed for the next 100 years.
[1]
$ dig news.ycombinator.com AAAA
; <<>> DiG 9.16.1-Ubuntu <<>> news.ycombinator.com AAAA
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 46616
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 1
;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 65494
;; QUESTION SECTION:
;news.ycombinator.com. IN AAAA
;; Query time: 176 msec
;; SERVER: 127.0.0.53#53(127.0.0.53)
;; WHEN: Thu Dec 09 15:21:09 EST 2021
;; MSG SIZE rcvd: 49
I'm very surprised at work and at home how often most of my Internet communications is running over IPv6. There are some surprising exceptions though [1].
I'd be willing to place a wager that there will be a $100mm company that is only IPv6 by 2030. And that the "Default IPv6 only on the Internet unless there is some weird edge case" will occur by 2040.
IPv4 as a protocol, though, will never, ever, go away - we'll see it widely deployed for the next 100 years.
[1]