| 22 May 2025 |
 uep | event storage, yeah. | 07:04:38 |
| uep | image cache storage is of course an entirely separate problem and there's no garbage collection there | 07:04:58 |
 Zhaofeng Li | that sucks, so they can still DoS, just more slowly | 07:05:11 |
 Cat | Thats where ACL kicks in. | 07:05:26 |
| Cat | ACL can help prevent that problem as ACLs dont give a shit about DAG level matters they act as network level blocks now ofc the ACLs have to be enforced by everyone or else they can leak and well current tooling for spotting leaks sucks as it is close to non existent. | 07:06:31 |
| Zhaofeng Li |
the ACLs have to be enforced by everyone
but now there's no built-in "block together"-like mechanism, right?
| 07:08:19 |
|  ayyz12 joined the room. | 07:08:31 |
| Zhaofeng Li | (I imagine the personal homeservers being the most difficult to get on board with such things) | 07:09:38 |
| Zhaofeng Li | *
the ACLs have to be enforced by everyone
but now there's no built-in "block together"-like mechanism, right? welp, I actually thought it was something like homeserver-level ACLs that block federation entirely
| 07:13:51 |
 f0x | In reply to @uep:matrix.org
image cache storage is of course an entirely separate problem and there's no garbage collection there there is? https://element-hq.github.io/synapse/latest/admin_api/media_admin_api.html#purge-remote-media-api | 07:14:22 |
| Cat | Turning off federation wholesale at room creation is also a thing. | 07:14:55 |
| Cat | But yes you can also enable allow list federation on a room level via ACL. | 07:15:09 |
| Zhaofeng Li | yeah, but it's not really practical | 07:15:42 |
| Zhaofeng Li | I know about the room-level homeserver ACLs, but they seem to me to be less effective than real homeserver bans | 07:16:41 |
| f0x | In reply to @zhaofeng:zhaofeng.li
I know about the room-level homeserver ACLs, but they seem to me to be less effective than real homeserver bans what would be the difference? | 07:18:26 |
| Cat | Matrix doesnt have de federation of that nature and it wont work due to how the protocol works. | 07:19:10 |
 Gnuxie 💜🐝 | In reply to @f0x:pixie.town
what would be the difference? the difference would be huge | 07:19:23 |
| Gnuxie 💜🐝 | lemme get you a link | 07:19:32 |
| Gnuxie 💜🐝 | @f0x:pixie.town https://github.com/matrix-org/matrix-spec-proposals/pull/4124 | 07:20:36 |
| Zhaofeng Li | I was thinking of activitypub/mastodon-style bans - you don't need to add the blocklist to each room and just ignore anything with a sender from the blocked homeservers | 07:21:07 |
| Zhaofeng Li | (i.e., blocked applied by homeserver admins, not room admins) | 07:21:27 |
| Zhaofeng Li | * (i.e., blocks applied by homeserver admins, not room admins) | 07:22:57 |
| f0x | In reply to @zhaofeng:zhaofeng.li
I was thinking of activitypub/mastodon-style bans - you don't need to add the blocklist to each room and just ignore anything with a sender from the blocked homeservers I think the best you could do is automatically setting server-acl (ish) things on all rooms your users have permission to. There's a fundamental difference with ActivityPub in how a block needs to affect other users/servers: AP servers are free to not send events to arbitrary other servers, and hide/drop events from blocked servers, but this approach doesn't work well in a shared chatroom, especially when it needs server cooperation to arrive at a shared room state | 07:28:59 |
| f0x | In reply to @gnu_ponut:matrix.org
@f0x:pixie.town https://github.com/matrix-org/matrix-spec-proposals/pull/4124 right, that's less about homeserver bans though, and rather a mechanism to allow for pre-screening? | 07:31:14 |
| Gnuxie 💜🐝 | In reply to @f0x:pixie.town
right, that's less about homeserver bans though, and rather a mechanism to allow for pre-screening? it's both | 07:31:36 |
| f0x |
We will embed m.server.knock_rule in m.room.create if it someone raises
small error 'it someone'
| 07:31:37 |
| Gnuxie 💜🐝 | oki | 07:32:19 |
| f0x | In reply to @gnu_ponut:matrix.org
it's both ah yeah, by making it explicit which servers are allowed to interact with the room | 07:34:24 |
| Zhaofeng Li |
this approach doesn't work well in a shared chatroom, especially when it needs server cooperation to arrive at a shared room state
It's ugly theoretically (homeservers may never converge to the same state), but in practice it may not be that bad?
| 07:39:02 |
| Zhaofeng Li | right now servers are already missing events naturally and having different views of the room state | 07:39:25 |
