Conversation

Notices

1.  (mint@ryona.agency)'s status on Thursday, 15-Aug-2024 02:49:16 JST 
   https://git.pleroma.social/pleroma/pleroma/-/issues/3306
   >yeah just take the pipeline made for static activity rejection/rewriting on the level barely above DB and make it apply dynamically upon API requests
   Deranged.
   • feld (feld@bikeshed.party)'s status on Thursday, 15-Aug-2024 02:51:54 JST feld

     in reply to
     @mint there may be a couple things we can move out of MRFs but I think this discussion has to happen to make everyone understand exactly what these do and that they're basically a firewall/ACL/content normalization policy set by the admin and really shouldn't be possible to bypass for security purposes
      likes this.
   • 0 (https://pl.absolutelyproprietary.org/users/0)'s status on Thursday, 15-Aug-2024 03:02:44 JST 0

     in reply to
     @mint
     
     > users should be able to override instance tier filtering.
     
     while i host the db, no.
      likes this.
   • Yukkuri (iamtakingiteasy@eientei.org)'s status on Thursday, 15-Aug-2024 03:37:34 JST Yukkuri

     in reply to
     @mint Local actor is cached anyway, so there is little overhead for MRF to check actor settings. That being said, it is difficult to think of any actual use-case, as normalizing MRFs are usually targeting posts of external actors/users of odd clients and there is no reason to permit some un-normalized posts. Maybe more gimick-y policies could actually make use of it, being opt-in rather than opt-out, such as group relays, forum-style signatures or some simple games.
      likes this.
   •  (mint@ryona.agency)'s status on Thursday, 15-Aug-2024 03:38:19 JST 

     in reply to

     • Yukkuri
     @iamtakingiteasy Well then it should be handled on the level of MRF itself, maybe with user tags similarly to TagPolicy.
   • pwm (pwm@pl.absolutelyproprietary.org)'s status on Thursday, 15-Aug-2024 03:38:28 JST pwm

     in reply to
     @mint The only proper response was, "Are you drunk?"
      likes this.