otroan@xxxxxxxxxxxxx wrote on 05/11/2021 09:58:
Perhaps a better lithmus test would be: "Would an end-user notice if IPv6 was removed from his/her/it's host?"
this would not be the only important test. Another would be: if ipv6 were removed from the entire user base, would the operator notice, and if it were bad enough, would the end user notice - e.g. pricing, video-provider dropping out due to NAT session concurrency problems, CGN hardware/software total-cost-of-ownership, etc.
ipv4 will not go away until there is a compelling reason to abandon it in favour of ipv6 or another future protocol. One reason to sit up and take notice is that ipv4 addresses currently cost $50/each, so if you need lots of ipv4 addresses, they are now a commodity which may make a difference to your bottom line.
Conversely there are plenty of organisations in some parts of the world that have more than enough ipv4 address space for their needs.
Dislodging organically evolved protocols is always difficult and often impossible.
Nick
