APRICOT 2026 keynotes: Agency, memory, and getting IPv6 done | APNIC Blog

The APRICOT 2026 keynotes set three complementary challenges for the Internet community — use automation without losing agency, preserve the operational record that explains how we got here, and finish the work of IPv6 where it still lags.

APNIC Blog
"Linkova pointed out that dual‑stack was never the finish line, as it increases operational surface, hides defects, and does nothing about IPv4 scarcity."

@tschaefer
The best statement of every Jen presentation:
"You are not operating IPv6 until you turn IPv4 off"

#IPv6mostly #IPv6only #IPv6 #enterprise

@goetz @tschaefer no dual stack love here?

@tschaefer i do ipv4 only for lack of complexity sake but i understand we are running out of addresses. maybe i just don't understand or appreciate the real appeal of ipv6 #arin #name rez #sdr #hosts #bind #unbound

Full confession: I run IPv4-only at home purely for simplicity. I get the address scarcity issue, but maybe I just don't see the real appeal yet.

The Core Appeal of IPv6: It's not just more addresses—it's about reclaiming end-to-end connectivity. No NAT, no port forwarding hassles. Every device gets a globally reachable IP.

For DNS/Naming (like #BIND/#Unbound): It eliminates the need for ugly NAT reflection/"hairpinning" to access internal services from your LAN. The host just works, internally and externally.

For SDR/Hosting: Imagine running an SDR server or a personal host—you can give it a static /64 prefix and actually reach it without wrestling with UPnP or carrier-grade NAT.

The Complexity Trade-off: You're right, dual-stack is complex. But the long-term play is IPv6-only with DNS64/NAT64 to handle legacy IPv4 traffic. That's actually less complexity than maintaining two parallel stacks forever.

#ARIN Reality: They're essentially out of IPv4. If you need more than a tiny allocation, v6 is the only path forward."