@gt94sss2 Indeed, agreed. I’d go further and say that many people have no clue which protocol they’re using and very many don’t even know that there are two internets.
But I think that various actors could cause a push for change. Perhaps a marketing campaign for ‘real ale’ ISPs to get a gold star which users are manipulated into looking out for. Apologies for mixed metaphors back there. There is a cost to software developers (like me) in developing, maintaining and testing code for two internets, and app firmware developers who want to switch to IPv6-only will be interested in putting there support behind some a ‘manipulation’ or marketing campaign very vaguely like the World IPv6 days (back in the two years around -what was it 2011 and 2012 ?), and the switchover to ubiquitous https that we mentioned.
It was clear even back thirty years ago that IPv4 had been a big failure, because I suspect that no one originally could imagine a world where one or multiple users per domestic household might have one or more computers, needing many billions of IP addresses. Even if in the 1970s and early 80s one knew about Moore’s Law, then developments such as the www, and miniature mobile computers with RF networking in them, such new applications / raisons d’ être for the internet were not obvious, not immediately predictable. So these drivers, answers to the question ‘why would you want that in your home, or even pocket?’ were altogether new and powerful and broke the internet out of its early role of connecting large machines only at universities and a couple of computing or comms-related corporates, a role which would only need a few tens of thousand IP addresses. Some sites were later even outside the USA! So the size of the 32-bit IP address was, as we know, so very wrong and this got acknowledged way too late, since the 1990s user explosion had already taken place and IPv6 was not even remotely ready, or not even born in time to get installed in that new massive user base.
With the interplanetary/ bundle network protocol, the designers were determined not to repeat their mistake of the late 70s and used variable-length addresses that are pretty much like email addresses, not n-bit numbers, where n is even greater than 128 perhaps. To address all the users on all the planets in the galaxy requires who knows how many bundle protocol addresses; astronomers could perhaps make a decent overestimate. But what happens if suddenly one day the requirements change to require extending the range to multiple galaxies, and a lot of them too?
