> So you're basically proposing to make phones worse.
I'm proposing to make phone worse by suggesting my wireless broadband should not be behind NAT? How is upgrading this infrastructure worse than the patchwork disaster we have now? Can you give some examples?
> I don't see why.
I don't see why not.
IPv6 is lacking critical mass. The mobile space is fast-lived and semi-organized enough to serve as a vector to push fundamental things like these through into general adoption.
IPv6-only phones would be accessing mostly IPv4-only servers... through NAT64.
I don't think the concept of critical mass is applicable to IPv6. Even if, say, 10M phones switched over, I don't think that would increase anyone else's incentive to adopt IPv6. If you don't use transition technologies there's an insurmountable chicken-egg problem, but if you use transition technologies there's no real network effect.
> IPv6-only phones would be accessing mostly IPv4-only servers... through NAT64.
Not for long, that's the whole point. I postulate in order to make 80% of servers IPv6 addressable, it would only take a couple dozen companies to upgrade their servers. They'll do it when clients out there want it.
> Even if, say, 10M phones switched over, I don't think that would increase anyone else's incentive to adopt IPv6.
The number of mobile internet devices is increasing very sharply. And yes, accommodating this growing segment of users is going to force network upgrades globally. The chicken-egg problem is exactly the reason why I mentioned this idea. Mobile device manufacturers and operators have leverage, influence and money.
Comments
> So you're basically proposing to make phones worse.
I'm proposing to make phone worse by suggesting my wireless broadband should not be behind NAT? How is upgrading this infrastructure worse than the patchwork disaster we have now? Can you give some examples?
> I don't see why.
I don't see why not.
IPv6 is lacking critical mass. The mobile space is fast-lived and semi-organized enough to serve as a vector to push fundamental things like these through into general adoption.
IPv6-only phones would be accessing mostly IPv4-only servers... through NAT64.
I don't think the concept of critical mass is applicable to IPv6. Even if, say, 10M phones switched over, I don't think that would increase anyone else's incentive to adopt IPv6. If you don't use transition technologies there's an insurmountable chicken-egg problem, but if you use transition technologies there's no real network effect.
> IPv6-only phones would be accessing mostly IPv4-only servers... through NAT64.
Not for long, that's the whole point. I postulate in order to make 80% of servers IPv6 addressable, it would only take a couple dozen companies to upgrade their servers. They'll do it when clients out there want it.
> Even if, say, 10M phones switched over, I don't think that would increase anyone else's incentive to adopt IPv6.
The number of mobile internet devices is increasing very sharply. And yes, accommodating this growing segment of users is going to force network upgrades globally. The chicken-egg problem is exactly the reason why I mentioned this idea. Mobile device manufacturers and operators have leverage, influence and money.