Poa.st / Chudbuds.lol General Discussion Thread - !! Poa.st and Bae.st have been compromised, all direct messages have been leaked. !!

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
Posted this to the 4chan thread as well but Graf has considered buying 4chan from hiroshimoot. Going as far as to email him a 1 million dollar offer. (Which is surprising considering the occasional funding issues he has)
This is insanely optimistic.

1745021568781.webp
1745021615325.webp

Besides this he seems to intend to federate it via ActivityPub. I'm not entirely certain about how scalable the protocol is, but for the volume it seems insane. Much like attempting to buy it in the first place.

He's already spent money and cleared it with the server host.
1745021686055.webp

My understanding of ActivityPub is that it slaughters your database, and that federating requires a network setup that doesn't allow for great defense against attacks. It's why the Kiwifarms instance died after Cloudflare dropped us; the rotating reverse-proxy that keeps the site up doesn't work with federation. Poas.st has only been lightly attacked, relatively to either KF or 4chan.

Beyond the technical issues, I don't know what 4chan (or any anonymous imageboard) gains from federation.

The experiences are completely different. 4chan's catalog is incompatible with fediverse software, which is server and timeline based. If you're on poast, you get a Following view, a Poast-specific view, and an Entire Network view. These are unsorted, chronologically arranged, endless streams. How would 4chan even show up? An endless stream of posts that you have to dig through expanded threads to figure out the full context?

But nobody wants to see EVERY board on 4chan, they hang out in a small number of chosen boards. /pol/ posters don't want to see /lgbt/ posts in their timeline, /a/ doesn't care about /lit/, nobody wants /b/, etc. A fediverse user on Poast doesn't want all those boards in their Network view, even if there's something interesting they'll just wind up muting the entire 4chan server.

The only way to make it bearable is to make each board show up as its own "server" on fedi so you can mute them, but that requires even more hacking, probably to fedi software as well. And it'll start out spamming every federating server and forcing people to mute a whole new batch of "servers" like they did in the old CP flood days. Not a good way to gain popularity.

In reverse, 4chan could start consuming federated posts in a way that a server = a board, but then you get flooded with boards. Plus every single post will be "namefagging" because all posting on the fediverse is under unique accounts, even the posts coming over the Nostr bridge are uniquely identified by wallets. A channer commenting on a fedi post will show up as an anonymous response on the fedi side, which is sure to piss off fedi users even more.

And that's probably the biggest mismatch: the culture. Fedi is pseudonymous, account-based micro-blogging, that stays up forever due to decentralization. 4chan is anonymous shitposting that's supposed to disappear. In today's MATI Null correctly pointed out that the ephemeral design of imageboards meant their users treated them as chat rooms, not forums or blogs. Jump in, chat about something, tomorrow that chat is gone and you start a new one. You never know who you're talking to, and that's part of the appeal; it's also why the boards would go nuts over identifiable posters, because they were rare and unique enough to be noticeable.

The user who wants that doesn't want to transform into a name-fagger whose posts live forever. The friendship-building, follower-oriented, account-signup micro-bloggers don't want the same kind of engagement as a channer. You can mash the two together in one feed, but that culture clash is not going away.

Poast is infamously protective of its own culture, and brags about chasing away mastodon, Gab, Truth Social, Kiwis, Jim fans, TRS/NJP, Gab refugees (again), and multiple waves of Twitter refugees. Throw them together with channers, and they'll just gatekeep them into another separation.

Ultimately, why bother? If you're a 4chan user and you want to view Poast posts, just make an account. Same if you're a Poast user who wants to visit /pol/ or whatever. Merging them might be technically possible, but ultimately pointless. Neither side is gatekeeping a desired experience from the other, and merging the two doesn't make either side happier.
 
Activity Pub has its problems, but there are certain things you can do to help mitigate denial of service attacks. It's not quite as easy as it would be if you just hosted a website and stuck it behind Cloudflare, as there may be additional vulnerabilities in the protocol that create a larger attack surface.
 
Merging them might be technically possible, but ultimately pointless.
Imageboards and big ap servers such as poast, nicecrew tend to use threads as chatrooms so other than cultural conflicts, which I agree, poast is unlikely to handle gracefully, you could probably merge, or at least facilitate interop between, the two softwares sanely, but it might be more like, 4chan becomes a sort of imageboard style frontend for data spread via activitypub, or 4chan is extended to export and federate its data to the activitypub network.

A development effort like that would take a serious team of real developers some time to fully realize into a commercial site (and they would not be working for free, either). Wouldn't happen overnight, and wouldn't ultimately come from the existing codebase. Other than buying the domain name and brand, there would be little purpose in actually buying the thing, since so much would have to be made from scratch.

My understanding of ActivityPub is that it slaughters your database
Not really, especially not if you are not storing posts/users/federation data forever. For an imageboard, the database requirements would not be much bigger than they already are, as once threads are pruned, you just drop the data once it falls off the archive. Fediverse admins already do a similar process to manage their db size, though on a much longer/wider scope.

In the meantime, if graf wants to buy 4chan, he should run it as it as an imageboard separate from poast, then pay a team of developers to rewrite the thing to support fediverse interop. I would even say the wise approach is to restrict it by board. E.g., once feature complete (or for beta testing) add a board that does federate, so that users can acclimate,

Just my $0.02. I support it inasmuch as the fediverse needs stuff going on to not die out, even if it is ultimately a failure.
 
Posted this to the 4chan thread as well but Graf has considered buying 4chan from hiroshimoot. Going as far as to email him a 1 million dollar offer. (Which is surprising considering the occasional funding issues he has)
Besides this he seems to intend to federate it via ActivityPub. I'm not entirely certain about how scalable the protocol is, but for the volume it seems insane. Much like attempting to buy it in the first place.
Graf is always on some megalomaniac manic episode where he's gonna partner up with X and do the Y.
 
Why doesn't this retard manlet try making a small imageboard first, and then seeing if he likes it and wants to graduate to literally the biggest imageboard in the world?
That would involve putting actual work in over a period of time with little reward at first and no guarantee of success. graf loves to be that guy who makes bold announcements but never actually delivers on anything as it's just an easy way to feel good. He never reflects on the broken promises and just moves onto the next harebrained idea.

Announcing something when you have nothing but ideas guarantees a project will never get anywhere.

I'm sure his ISP contact (probably just the owner of Incognet) is used to this kind of shit now and just says yes to everything knowing that nothing will come of it other than a few short term hardware leases that'll be left idling.
 
Imageboards and big ap servers such as poast, nicecrew tend to use threads as chatrooms so other than cultural conflicts, which I agree, poast is unlikely to handle gracefully, you could probably merge, or at least facilitate interop between, the two softwares sanely, but it might be more like, 4chan becomes a sort of imageboard style frontend for data spread via activitypub, or 4chan is extended to export and federate its data to the activitypub network.

A development effort like that would take a serious team of real developers some time to fully realize into a commercial site (and they would not be working for free, either). Wouldn't happen overnight, and wouldn't ultimately come from the existing codebase. Other than buying the domain name and brand, there would be little purpose in actually buying the thing, since so much would have to be made from scratch.

Well, NiceCrew really isn't that big of an instance, we're just pretty active and terminally online. I am not entirely sure how 4chan would become federated, unless they used something like the Nostr protocol to arbitrarily append IDs to names. However, with that being said, I feel that it would have massive implications considering the difference between the federated community and 4chan - the latter userbase probably not understanding the cultural differences. This is of course not withstanding the issue of moderation for an image board. Due to the issues with moderating an imageboard the size of 4chan, I would likely block all activities from it because I don't want to deal with it. Nostr is already sketchy enough to federate with. I wish Graf the best of luck, but that is one hell of a responsibility.
 
  • Informative
Reactions: byuuWasTaken
There is an existing federated AP imageboard software I think, FChan.
And all its nodes (mostly hosted by FCE/sleepy.cafe people, I recall) closed down because of CP spam, with people deeming the whole suite radioactive since. Also there were no interoperability with anything other then FChan itself.
 
1745216919218.webp
He didn't even get a response and he upped his offer by 25%. That shows desperation and is not a good look in negotiation.

This deal isn't going to happen. Even if it does, I don't think graf appreciates what he would be getting into. Personally the idea of rewriting 4chan in Go and federating it sounds fun as hell as a project but impractical and not likely to ever work as a product. I don't even know what the tie into the fediverse would be since image board and micro blogging experiences are so fundamentally different. It's just a square peg/round hole situation. Not to mention how totally unprepared he is for moderating a big pond with so many malicious actors as 4chan attracts.

All that said I fully support the continuation of this effort as the worst case scenario is something funny to watch unravel and the absolute best case is a functional image board with poast culture which would be an improvement on modern 4chan.
 
And all its nodes (mostly hosted by FCE/sleepy.cafe people, I recall) closed down because of CP spam, with people deeming the whole suite radioactive since. Also there were no interoperability with anything other then FChan itself.
They weren't closed because of a spam problem per se. People stopped posting cause the joke got old, and no one was posting or looking at the boards anymore. It would normally be fine to keep up a dead website, except for that spam, which was rare but forced the admins to check on an empty board every day just in case. So that's why they took them down. If you're actually paying attention to your fchan instance, it's not different to any other imageboard in that respect. The fchan dev was really nice and I wouldn't want people thinking his software is "radioactive", people just didn't want to moderate dead empty boards after the bit got old so they closed them.
 
You could not pay me to buy and operate poast. Maybe you could pay me to ooperate it but no one is going to touch that shit. 4chan is at least a famous site. Poast is... in a couple news articles no one ever read.
 
Back