There is also another complexity involved in the authority chain of federated messages. For example, you created a post to /c/science@mander.xyz, and you are from lemmy.nz. The spammer is from lemmy.blahaj.zone.
As the admin of mander.xyz, I am able to purge the user and all of its posts. However, even though the community is on my instance, I do not have the authority to rule over this content with respect to the other instances… So, even if I take an admin action it is not enough to stop the problem. The structure of authority is useful because it helps prevent abuse from admins, but it also means that we often need multiple different people to deal with the same issue.
Also, since the community is in my instance, I still get all of the reports from the other instances that can still see the spam comments. I can’t take any action, as the messages show up as already removed, but I still need to manually click on them one by one to consider them resolved.
I am just trying to explain to you what it looks like from my perspective. It’s not that I am complaining about how things work now. I understand that lemmy is still being developed and it needs to be polished, and as we experience these systems in practice we can come up with ways to improve the ways things work. I just want to share some of my perspective so that you can get an idea of why situations like this one might evolve the way they do. You really do need a combination of volunteers spread across different actions to take some coordinated action on a Sunday 😅 It is not optimal, but no one knows for sure what “optimal” looks like and how to get there.
Where you acting as a mod or an admin? Perhaps mod actions do get federated. I am the mod in that community, but I took an admin action to fully ban the account and remove the content. Perhaps if I would have taken mod actions instead by deleting every comment manually it would have federated… I will test next time!
Because for admins, while your reasoning makes sense, the main issue is probably trust. Adding an admin to your instance is basically giving them the keys on everything, and allowing them to destroy the whole thing.
I know instance admins tend to be really reliable, but I also understand why some might be reluctant to add 4 different admins (let’s say 2 more per time regions, and assuming you need 2 additional time regions to cover 24/24) to their instances.
Well, since this spammer has gone crazy on several instances
I was thinking we need mods from different time zones.
There is also another complexity involved in the authority chain of federated messages. For example, you created a post to /c/science@mander.xyz, and you are from lemmy.nz. The spammer is from lemmy.blahaj.zone.
As the admin of mander.xyz, I am able to purge the user and all of its posts. However, even though the community is on my instance, I do not have the authority to rule over this content with respect to the other instances… So, even if I take an admin action it is not enough to stop the problem. The structure of authority is useful because it helps prevent abuse from admins, but it also means that we often need multiple different people to deal with the same issue.
Also, since the community is in my instance, I still get all of the reports from the other instances that can still see the spam comments. I can’t take any action, as the messages show up as already removed, but I still need to manually click on them one by one to consider them resolved.
I am just trying to explain to you what it looks like from my perspective. It’s not that I am complaining about how things work now. I understand that lemmy is still being developed and it needs to be polished, and as we experience these systems in practice we can come up with ways to improve the ways things work. I just want to share some of my perspective so that you can get an idea of why situations like this one might evolve the way they do. You really do need a combination of volunteers spread across different actions to take some coordinated action on a Sunday 😅 It is not optimal, but no one knows for sure what “optimal” looks like and how to get there.
Interesting, I didn’t think about it that way.
But wait, I had a similar issue:
Is this normal? It seems to contradict your experience
Also, if that may help, I don’t see the spammer messages on the post in !science@mander.xyz
I can still see the spammer messages, look: https://lemmy.ca/post/10554277
Where you acting as a mod or an admin? Perhaps mod actions do get federated. I am the mod in that community, but I took an admin action to fully ban the account and remove the content. Perhaps if I would have taken mod actions instead by deleting every comment manually it would have federated… I will test next time!
Ah, that’s interesting! I was indeed a mod, not an admin, so probably that’s why the action got federated.
Example: https://jlai.lu/post/2612405 https://lemmy.ca/post/10554287
Well, yeah I understand we can’t manage across other instances 😅
I mean at least we can stop them fron spamming our instance quickly. Even on bigger instance (lemmy.world) they spammed for over an hour.
I think lemmy need something like user made bot just like reddit to stop spammers instantly.
In addition: as far as I know, only Lemmyworld has a timezone distributed admin team, but they are the biggest instance, so it makes sense for them.
That’s difficult for smaller and localised instances. I’m more thinking plain simple spam detection.
Yeah but it takes time to create bot to detect spam. And lemmy need user made anti spam bot just like reddit.
As for now adding mods is much easier
Only mods can ban users at the instance level based on my experience (I’m a mod on a few communities, and can only ban users on the community level)
TIL lemmy doesn’t have Supermod
Months ago, I was banned from world news on lemmy.ml, and I can’t post in any community on lemmy.ml
I thought it was a new mod on world news, now I know it wasn’t them 😅
Do you mean mods, or admins?
Because for admins, while your reasoning makes sense, the main issue is probably trust. Adding an admin to your instance is basically giving them the keys on everything, and allowing them to destroy the whole thing.
I know instance admins tend to be really reliable, but I also understand why some might be reluctant to add 4 different admins (let’s say 2 more per time regions, and assuming you need 2 additional time regions to cover 24/24) to their instances.
Mods not admins, at least we can stop spammers quickly.
Only mods can ban users at the instance level based on my experience (I’m a mod on a few communities, and can only ban users on the community level)
I see, I thought lemmy has something like Supermod. They can ban users on instance level, but they don’t have all admin’s privileges.
Unfortunately, not at this moment. Hopefully something we’ll see in the future. Would help with this kind of issues.