Regarding moderation transparency #113

oatbiscuits commented 2024年02月05日 10:00:23 +01:00

I have read the moderation process document.

Given that a decent chunk of content has been recently removed, can there be a public moderation log (it could be delayed to avoid knee-jerk reactions) to document the content that was removed?

I have read the [moderation process document](https://codeberg.org/forgejo/governance/src/branch/main/MODERATION-PROCESS.md). Given that a decent chunk of content has been recently removed, can there be a public moderation log (it could be delayed to avoid knee-jerk reactions) to document the content that was removed?

algernon commented 2024年02月05日 10:07:42 +01:00

I do not think this is a good idea. The goal of moderation is to remove the noise, the content that goes against the code of conduct, or is otherwise problematic. Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too.

I do not think this is a good idea. The goal of moderation is to remove the noise, the content that goes against the code of conduct, or is otherwise problematic. Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too.

algernon commented 2024年02月05日 10:18:47 +01:00

As a prime example, ban evading and derailing topics is not something that should be given a platform in a public log.

As a prime example, ban evading and derailing topics is not something that should be given a platform in a public log.

oatbiscuits commented 2024年02月05日 10:20:04 +01:00

Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too.

Putting a delay (30 days, for example) on it, possibly removing the identifying bits, and posting it as a markdown or log file (no comment section) might address these concerns.

I understand the trade-off and the effort involved, but I believe some form of a compromise to bring a level of transparency into moderation actions is worth it.

I also do understand that ultimately Forgejo members decide on such matters, but I thought it is important to bring up nonetheless.

> Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too. Putting a delay (30 days, for example) on it, possibly removing the identifying bits, and posting it as a markdown or log file (no comment section) might address these concerns. I understand the trade-off and the effort involved, but I believe some form of a compromise to bring a level of transparency into moderation actions is worth it. I also do understand that ultimately Forgejo members decide on such matters, but I thought it is important to bring up nonetheless.

oliverpool commented 2024年02月05日 10:28:11 +01:00

@oatbiscuits you can write moderation@forgejo.org to get the moderation report.

In this case it is not the content of the messages that is problematic, but the fact that the poster has been explicitly banned (so any message the poster may write will be moderated during the ban period - independent of the content).

If you want a log of the messages, you should activate email notifications (and add a sorting rule in your inbox to prevent becoming overwhelmed :).

@oatbiscuits you can write moderation@forgejo.org to get the moderation report. In this case it is not the content of the messages that is problematic, but the fact that the poster has been explicitly banned (so any message the poster may write will be moderated during the ban period - independent of the content). If you want a log of the messages, you should activate email notifications (and add a sorting rule in your inbox to prevent becoming overwhelmed :).

algernon commented 2024年02月05日 10:32:24 +01:00

Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too.

Putting a delay (30 days, for example) on it, possibly removing the identifying bits, and posting it as a markdown or log file (no comment section) might address these concerns.

Removing the identifying bits is hard, and error prone. When it comes to moderation, I believe the straightforward approach is to err on the side of safety, and remove everything. That's a trade off, yes. Some useful bits may be lost, indeed. But the cost of failing to remove identifying bits properly is, I think, greater than any value we may possibly gain from a public log.

There's an internal, non-public moderation log. Not being a moderator, I have no clue what's in there, but based on the moderation process, it should contain all the information necessary for a review, should the action be appealed. I suspect it might contain enough information to restore a removed message too, would the appeal be granted. I think that's fine, and this information does not need to be public.

I understand the trade-off and the effort involved, but I believe some form of a compromise to bring a level of transparency into moderation actions is worth it.

I think the current practice of forgejo-moderation posting a message at the place of removal is adequate and sufficiently transparent, and protects all parties: it does not reveal who did the moderation, and it does not reveal the target either. Once the message is removed, identifying the target without seeing the original first is going to be difficult, as it should be.

> > Reintroducing those, even partially, in a public place defeats the purpose of moderation, does it not? It gives a platform for spammers, too. > > Putting a delay (30 days, for example) on it, possibly removing the identifying bits, and posting it as a markdown or log file (no comment section) might address these concerns. Removing the identifying bits is hard, and error prone. When it comes to moderation, I believe the straightforward approach is to err on the side of safety, and remove everything. That's a trade off, yes. Some useful bits may be lost, indeed. But the cost of failing to remove identifying bits properly is, I think, greater than any value we may possibly gain from a public log. There's an internal, non-public moderation log. Not being a moderator, I have no clue what's in there, but based on the moderation process, it should contain all the information necessary for a review, should the action be appealed. I suspect it might contain enough information to restore a removed message too, would the appeal be granted. I think that's fine, and this information does not need to be public. > I understand the trade-off and the effort involved, but I believe some form of a compromise to bring a level of transparency into moderation actions is worth it. I think the current practice of `forgejo-moderation` posting a message at the place of removal is adequate and sufficiently transparent, and protects all parties: it does not reveal who did the moderation, and it does not reveal the target either. Once the message is removed, identifying the target without seeing the original first is going to be difficult, as it should be.

forgejo-moderation commented 2024年02月05日 12:54:20 +01:00

A message from an account created to circumvent a ban from Forgejo spaces was deleted, send a mail to moderation@forgejo.org for more information.

A message from an account created to circumvent a ban from Forgejo spaces was deleted, send a mail to moderation@forgejo.org for more information.

earl-warren commented 2024年02月05日 12:59:48 +01:00

@oatbiscuits all moderated content is archived and available on request to moderation@forgejo.org. It will just not be made public.

@oatbiscuits all moderated content is archived and available on request to moderation@forgejo.org. It will just not be made public.