IPv6 is not a solution when there still is a lot of people whose networks don't support it. Making the network publicly routable shouldn't be any more complex than any other ISP which seem to do this just fine.
When there's enough time for you to conceive a new child, send them through school, then university, then most of a PhD program…
If your network equipment supplier doesn't support the tech, demand better of them rather than blaming the ISP that's specifically intended for people who weren't being properly served until now anyway.
And Chinese is hundreds of years old and I still don't know it. People don't invest into supporting things the instant they release.
>If your network equipment supplier...
In this hypothetical my network supplier would, but someone who is trying to connect isn't. IPv6 isn't even at 50% support in the US. That means if you want your friend group to all connect to your server there is a very good chance that some of them will not be able to. What are you supposed to tell them? That they should complain to their ISP?
> People don't invest into supporting things the instant they release.
Virtually every other aspect of the tech world, including networking, has managed to update multiple times in the intervening years.
We don't (mostly) use dialup; we switched from 2G to 3G to 4G and now to 5G mobile telephony since then; HTTP has been mostly deprecated for HTTPS and the cryptography standards for the latter have changed; etc.
> What are you supposed to tell them? That they should complain to their ISP?
Yes.
I mean, what, is their ISP also using Windows 98 on a 486 chip running at 75 MHz with 8 meg of RAM and a 500 meg hard drive, and software patches being passed around on floppy disks? Because that's what the tech world looked like when we should've started refusing to buy products and services that were IPv4-only.
Today?
Today, this kind of nonsense has gone on far too long. It was obvious even during my degree ~20 years ago that IPv4-to-6 transition should have already been finished by the time I even learned about it. As there are now far more devices on the internet than IPv4 addresses, I would go further and say that normal ISPs should be banned by law from using IPv4 — if we're going to do that for which type of USB connector personal electronics use, why not this?
> These solves real problems for consumers. IPv6 doesn't.
Apart from the problem of "we literally can't have all of you online at the same time in v4, y'all have too many devices".
Not that normal people should have to care about CNAT vs IPv6, but you act like the correct solution is an unacceptable horror story and the sticky plaster is panacea.
>"we literally can't have all of you online at the same time in v4, y'all have too many devices".
CGNAT solves that problem. The issue is more that all of your devices can't act as servers on the internet. Most people on the internet don't have a desire to run a server. So really the 4 billion cap of IPv4 is more about 4 billion different servers than 4 billion client devices.
>but you act like the correct solution is an unacceptable horror story
No, I am acting like the correct solution doesn't have any incentives to be implemented.
Wait, I don't know tons about networking but my impression was rhat CGNAT basically does not allow your users to be publically addressable? Have I misunderstood what CGNAT does, or it just that they should've used a publically addressable architecture in the first place?
CGNAT has multiple users share a single IP address which means that there needs to be some logic to know which user an incoming packet should go to which is why they are normally not publicly addressable. Not all ISPs use CGNAT though.
>just that they should've used a publically addressable architecture in the first place?
I would be okay if there was a way to get assigned a dedicated, though not necessarily a static one, IP address that was not at a massive margin.
