it seems someone decided to prove you really can just publish any nonsense protocol draft with the IETF https://www.ietf.org/archive/id/draft-meow-mrrp-00.html

[marlies@tacobelllabs.net]

@0xabad1dea I’m in a ridiculous amount of group chats, Discords, and IRC channels in which this was brought up yesterday

[mndflayr@metalhead.club]

@0xabad1dea

Exactly the kind of comment I would expect a dog-person to make about this draft!

MEOW!!!

[ghostinthenet@hachyderm.io]

@0xabad1dea Well, this explains that late April Fool’s day draft. #IPv8

[evoscale@c.im]

@0xabad1dea But is this just a... ruff draft?..

[toadjaune@hostux.social]

@Infoseepage @0xabad1dea the MEOW designers definitely desire and planned for this interoperability, however, for some reason, existing IPoAC implementations tend to be unhappy with the interaction and deny further service !

[kasperd@westergaard.social]

Before I viewed the link in your post I thought you were talking about that "IPv8" draft. I have some experience writing networking code, and based on my experience I will say that the meow draft has a slightly better chance of resulting in a useful implementation than the "IPv8" draft.

I didn't read enough of the "IPv8" draft to make it to the parts you are referring to. I didn't need to read that far to figure out that it was worthless.

The meow draft I can see is based on TCP. But I think before it can be implemented IANA would have to allocate a protocol number. And the meow draft seems to fall short of the requirements for such an allocation.

[fuzzyfuzzyfungus@cyberplace.social]

@Infoseepage @0xabad1dea "Come see the violence inherent in the OSI 7 layer model!"

[kasperd@westergaard.social]

When I visited that link I was reminded of: https://www.youtube.com/watch?v=yL_-1d9OSdk

[nila@infosec.exchange]

@0xabad1dea I wonder how interesting of an attack vector that could be in real life scenarios. How about having an LLM generate 500 drafts with non-obvious misinformation and watch crawlers crawl it, models being trained with the data and then people citing the resulting misinformation?

Or, more wildly: what's the authN process for working groups? I wonder how far a swarm of agent sock puppets could get in the RFC process...

[shwell@mastodon.au]

@0xabad1dea Bluesky?

[hugo@someone.elses.computer]

@0xabad1dea @slothrop ? Am just saying it d be a good April’s Fools thing to do. Don’t see replies on my fedi client.

[0xabad1dea@infosec.exchange]

@hugo @slothrop I’m saying it’s a pointed political statement and not a random joke https://infosec.exchange/@0xabad1dea/116419619962656186

[itnomad@ruhr.social]

@0xabad1dea I, for one, would rather welcome our feline overlords.
(Edit: remove the URL)

[zackrobbin@mstdn.social]

@0xabad1dea

[ohir@social.vivaldi.net]

@cinnamon @0xabad1dea
> about time to have a standard for some kinds of communication

There is one

Toki Pona (official site)

(tokipona.org)

[ddgulledge@social.linux.pizza]

@0xabad1dea I hadn't heard about it before, but assuming it was an April Fool's Day RFC, I just searched for the ones that were published on Apr 1, 2026 and it wasn't there. A little more searching found it, and it seems to have been intended as one, but was still in a draft state. The delay in the joke feels like meta-humor and works well considering the context of IPv4 address space exhaustion and the slow implementation of IPv6.

[jelu@mastodon.social]

@0xabad1dea I hope people learn a little bit about the IETF draft process also: https://www.ietf.org/participate/ids/

Yes, anyone can submit a draft. It does not mean anything beside showing how strong what-ever they just took is...

[ratsnakegames@mastodon.social]

@nila @0xabad1dea the auth process is "if the IETF isn't interested, it doesn't proceed to the RFC stage", apparently.

You can't just post a standard, but you CAN just post a draft/proposal, which is what both this and "IPv8" are.

[ratsnakegames@mastodon.social]

@nila @0xabad1dea as long as crawlers correctly interpret the draft as a draft and journos / wiki editors etc do their due diligence (which, i am told, they always do), there is no attack vector

[nila@infosec.exchange]

@ratsnakegames @0xabad1dea Do you think an agent swarm led by an APT could social engineer its way into "the IETF"? Setting up web pages, blogs, arXiv submissions etc. for a network of sock puppets definitely seems within reach of such actors.

(And, come to think of it, any other standards body...)