!meowlnir-v11:maunium.net

Meowlnir

77 Members
Discussion about Meowlnir | Latest release: v0.3.0 | https://github.com/maunium/meowlnir53 Servers

Load older messages


SenderMessageTime
17 Mar 2025
@tulir:maunium.nettulirno18:21:25
@nex:nexy7574.co.uknex (she/it)🦈🐈️Is there an obvious reason why that shouldn't be done that I'm not thinking of? I'm conscious of how limited ACLs are in terms of size constraints18:22:25
@nex:nexy7574.co.uknex (she/it)🦈🐈️Redacted or Malformed Event18:22:35
@tulir:maunium.nettulirthere shouldn't be that many duplicates 🤔18:22:59
@nex:nexy7574.co.uknex (she/it)🦈🐈️you clearly do not watch 10 policy lists 🚎18:23:16
@nex:nexy7574.co.uknex (she/it)🦈🐈️no but fr, one deduplicated pattern means an extra server can be banned18:23:34
@nex:nexy7574.co.uknex (she/it)🦈🐈️it'll definitely mean something to someone somewhere at some point18:23:41
@tulir:maunium.nettulirdeduplication like that sounds like a lot of effort18:24:20
@nex:nexy7574.co.uknex (she/it)🦈🐈️it does?18:24:27
@nex:nexy7574.co.uknex (she/it)🦈🐈️A lot of CPU cycles sure, but effort?18:24:33
@gnu_ponut:matrix.orgGnuxie 💜🐝it sounds like it, and it'd suck to get it wrong x318:24:49
@nex:nexy7574.co.uknex (she/it)🦈🐈️true18:25:01
@gnu_ponut:matrix.orgGnuxie 💜🐝 * it sounds like it, and it'd suck to get it wrong x3 18:25:03
@tulir:maunium.nettulirACL size probably needs to be fixed on the spec level sooner or later anyway, saving a couple hundred bytes won't make much difference when you get close to filling it up18:26:41
@nex:nexy7574.co.uknex (she/it)🦈🐈️I mean wouldn't changing the ACL event to accommodate larger ones require a new room version?18:27:21
@gnu_ponut:matrix.orgGnuxie 💜🐝please try get SCT to add dag server bans rather than just patching acl and leaving it a job well done meow18:27:46
@nex:nexy7574.co.uknex (she/it)🦈🐈️im not sure how you could reasonably implement an increased ACL size in a backwards compatible manner since older servers won't accept them18:28:02
@tulir:maunium.nettulirno, you'd just need to ban outdated servers18:28:13
@nex:nexy7574.co.uknex (she/it)🦈🐈️and I doubt a new room version would be on the table until this is already a problem18:28:13
@gnu_ponut:matrix.orgGnuxie 💜🐝 e.g. https://github.com/matrix-org/matrix-spec-proposals/pull/4124 18:28:16
@nex:nexy7574.co.uknex (she/it)🦈🐈️right, which is kinda what a new room version would do18:28:32
@nex:nexy7574.co.uknex (she/it)🦈🐈️unless you expect everybody to start running my implementation banner 🧌18:28:40
@tulir:maunium.nettuliracls already expect you to ban everyone who doesn't respect them18:28:56
@nex:nexy7574.co.uknex (she/it)🦈🐈️finding out who doesn't respect them in a PITA though and it's rarely done proactively18:29:22
@nex:nexy7574.co.uknex (she/it)🦈🐈️Redacted or Malformed Event18:30:53
@nex:nexy7574.co.uknex (she/it)🦈🐈️ I was also going to ask about implementing bans for invite/knock/join memberships that match a server ACL as a kind of backup "this user leaked in" but I'm not too sure how effective that'd even bet 18:31:28
@nex:nexy7574.co.uknex (she/it)🦈🐈️Redacted or Malformed Event18:31:35
@gnu_ponut:matrix.orgGnuxie 💜🐝the advantage of this MSC is that it is proactive and newly discovered servers can't interact with the room automatically because there is no ambient permission that lets them do that like there is currently.18:35:45
@nex:nexy7574.co.uknex (she/it)🦈🐈️I'm still giving it a read, it's a lot to digest lol18:36:10
@gnu_ponut:matrix.orgGnuxie 💜🐝yeah i don't believe it's possible18:36:32

Show newer messages


Back to Room ListRoom Version: 11