Okay. It’s still unclear to me why piracy was not picked up by default. It’s a very big comm.
Anyhow, good luck on the merge request! Would be great to let the admin decide what to block instead of the weirdly random selection of comms and words on the list now.
Not block. Just karma-devalue. I’m gonna rename the variable to something like “unwanted reputation sources”. If an admin doesn’t want people farming rep by posting porn, they can put porn on the list. That’s not blocking, it’s just a data filter for admin eyes only.
Yeah, I was editing the code yesterday and realised we were both wrong. What it actually does is prevent the automatic bulk federator, which is a tool only admins have, from fetching those communities. It doesn’t affect the manual community fetcher, which is what users use. So it doesn’t limit user capabilities, only admin capabilities for automation. Also, the commit you linked is super out of date, the code is very different now.
Okay. It’s still unclear to me why piracy was not picked up by default. It’s a very big comm.
Anyhow, good luck on the merge request! Would be great to let the admin decide what to block instead of the weirdly random selection of comms and words on the list now.
https://codeberg.org/rimu/pyfedi/issues/1618
Already passed it on, and it’s in the pipeline now.
Not block. Just karma-devalue. I’m gonna rename the variable to something like “unwanted reputation sources”. If an admin doesn’t want people farming rep by posting porn, they can put porn on the list. That’s not blocking, it’s just a data filter for admin eyes only.
So there are two purposes for this “bad list”
correct?
The federation filtering is blocking IMO. It affects what comms users of the instance can find (via search).
Anyway maybe in Your PR You should make a separately configurable list for each purpose. I don’t see a reason why it should be the same list.
No, just one purpose, the second one.
That contradicts what @[email protected] told me.
Yeah, I was editing the code yesterday and realised we were both wrong. What it actually does is prevent the automatic bulk federator, which is a tool only admins have, from fetching those communities. It doesn’t affect the manual community fetcher, which is what users use. So it doesn’t limit user capabilities, only admin capabilities for automation. Also, the commit you linked is super out of date, the code is very different now.