All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Internet Protocol Version 8 (IPv8) proposal
https://datatracker.ietf.org/doc/draft-thain-ipv8/
Instead of forcing a dual-stack architecture like IPv6, the proposed Internet Protocol Version 8 (IPv8) introduces a 64-bit address space that is natively backward compatible with IPv4. Any IPv8 address with a zeroed routing prefix (0.0.0.0.n.n.n.n) is processed under standard IPv4 rules. Because IPv4 is a proper subset of IPv8, there is no "flag day" or forced migration required for existing networks. This architecture resolves address exhaustion by providing every ASN with over 4.2 billion host addresses, while structurally bounding the global BGP table to a single entry per ASN.
To solve management fragmentation, IPv8 introduces the "Zone Server," a platform that delivers every required network service (like DHCP8 and DNS8) in a single lease response. Security is built-in, authenticating every manageable device via universally cached OAuth2 JWT tokens. Furthermore, north-south traffic is strictly validated at egress: outbound packets must have a corresponding DNS8 lookup and be verified against a WHOIS8-registered active route. This protocol-level enforcement effectively eliminates malware C2 channels that rely on hardcoded IPs.
I'm just glad there's no more letters in ip addresses.
@yoursunny must be crying in a corner
Comments
Do you know that anyone can submit draft, right?
@yoursunny has already lunched IPv9 on their facilities in Anctartica ( summer host ) and on the moon
Looks like surveillance hell.
Is this AI slop on the IETF too?
Nah, looks like government affiliated party based on how fucking dystopian that entire presentation is technically.
Nah, based on its technical problems and slop like:
It's gotta be an LLM.
Nah, that was translation from finnish blog which analyzes the draft
https://markusjansson.blogspot.com/2026/04/ipv8-tappaa-koko-internetin-eika-ketaan.html
Sounds like "OK, ipv6 just doesn't get real traction and is despised by many, so let's create a reasonable IP solution -but- make surveillance the price!" to me ...
My response: IPv8 yes, but your surveillance and limitations, no!
Isn't this not LLM generated crapware?
You beat me to it.
Whether it is or isn't, there's a precisely 0% chance that this, or any new IP proposal, will get adopted or even seriously considered.
looks like a single person of a bermuda ISP, decided to write a whole new RFC and protocols suite(using LLM), instead of deploying IPv6 in his network.
ALL BECAUSE HE IS AFRAID OF HEX.
I wish the same had been true of IPv6.
The only real problem with IPv4 is just how poorly the space has been utilised.
I think if we had both BGP supporting up to /32 (and the arguments for how that requires too much memory don't make sense if routing a /48 for IPv6 is fine), opened up the approximately 30% of unused reserved space for NAT-like use, and adopted DNS extensions to support port lookup as well as IP, we'd have more than enough space. Arguably, if DNS standardised geo-DNS, we wouldn't even need BGP beyond /24 because there wouldn't be an advantage to anycast anything other than DNS.
I feel quite sad when companies own an entire /24 block just to anycast a single HTTPS and then we're told IPv6 is the only solution to this problem.
I prefer ipv4.1q which merges ipv4 with 802.11q allowing ipv4 to simply be used over and over again by adding an origin route tag.
Any IPv6 address with a 0.0.0.0.0.0.0.0.0.0.255.255 routing prefix (e.g. ::ffff:59.78.20.24) is processed under standard IPv4 rules.
Misread that as AFRAID OF SEX, and was kinda puzzled, what that has to do with anything
Makes a lot more sense actually
Yuck.
I know that IPv6 adoption has been rough, but the biggest issue has been people being lazy and not pushing for it/using it.
A backwards compatible IPv6 would've been nice, but I'm not sure how possible or practical it would've been.
That's not the problem. The real issue is that, because it's taken so long, there's now a developed fear—probably among suit-wearing corporate types—that the transition might halt some services, make things inaccessible to someone, cause them to lose customers, or that there are more edge cases than there actually are.
much better:
Not even that. The paralysis comes from a deadlock: the fact that there is no incentive to switch to dual stack because everyone supports IPv4, and there's no ability to get rid of IPv4 until everyone switches to dual stack.
I agree that this is part of the issue. Some organisations will adopt double stacking, some will operate on IPv6 only, and others will continue to operate solely on IPv4. However, is double stacking the solution or merely part of the problem? It may actually slow down the transition.
I run dual stack and I can connect to everybody. At this point arguing about the protocol is just for fun right?
Just use IPv6 !?
https://www.ietf.org/archive/id/draft-meow-mrrp-00.html
@lichade important news dont miss
Packet loss prevented