Replace Alerts with a Report-Based Ban System

F4FNugget

Replace Alerts with a Report-Based Ban System

At the moment, Scratch uses the number of alerts a user receives to decide when to ban them. The problem is that alerts don’t always reflect actual harmful behavior — sometimes they come from false reports or misunderstandings. This can make the system feel inconsistent and less effective at protecting the community.

Instead, Scratch could use a report-based ban system where bans are tied directly to the number of valid reports a user receives. This would make consequences clearer, fairer, and harder to ignore.

Example System

15 reports → Temporary ban for 1 day

35 reports → Temporary ban for 1 week

70 reports → Temporary ban for 2 weeks

140 reports → Temporary ban for 1 month

500 reports → Ban for 1 year

1000 reports (or reports for terrible behavior) → Permanent ban

2000 reports total (if ban evading or using other accounts on the same IP) → IP ban

Why This Helps

More direct consequences than alerts, which can be ignored.

Fairer system: bans are tied to actual reports, not just moderator alerts.

Community protection: harmful behavior is removed faster.

Transparency: users know exactly how many reports lead to which ban.

This would make Scratch moderation more consistent and effective, while reducing confusion around alerts.

FeelingHammy417

F4FNugget wrote:
Replace Alerts with a Report-Based Ban System

At the moment, Scratch uses the number of alerts a user receives to decide when to ban them. The problem is that alerts don’t always reflect actual harmful behavior — sometimes they come from false reports or misunderstandings. This can make the system feel inconsistent and less effective at protecting the community.

Instead, Scratch could use a report-based ban system where bans are tied directly to the number of valid reports a user receives. This would make consequences clearer, fairer, and harder to ignore.

Example System
15 reports → Temporary ban for 1 day
35 reports → Temporary ban for 1 week
70 reports → Temporary ban for 2 weeks
140 reports → Temporary ban for 1 month
500 reports → Ban for 1 year
1000 reports (or reports for terrible behavior) → Permanent ban
2000 reports total (if ban evading or using other accounts on the same IP) → IP ban

Why This Helps
More direct consequences than alerts, which can be ignored.
Fairer system: bans are tied to actual reports, not just moderator alerts.
Community protection: harmful behavior is removed faster.
Transparency: users know exactly how many reports lead to which ban.
This would make Scratch moderation more consistent and effective, while reducing confusion around alerts.

I support because whenever I see projects get front-page for a Scratch Design Studio, multiple of them are removed because of Alerts. Then Alerts start adding up and then Scratch Team will decide based on the amount of alerts a user has whether to ban them (I think, my friend has gotten banned before), and some people can get banned for no reason. This would fix it! As long as the Scratch Team will actually review them, I support.

KittyLuca417

support cuz it sounds nice lol. it is quite a complicated thing but i read some of it and i like the idea

:p

hotcrystal

Semi-support. If this gets accepted, bad offenses like posting a very inappropriate project should probably count towards a high number of “reports” rather than just one. (also can someone post that essay by jvvg talking about how false reporting is a misconception here)

FollowMeASAP

Wow nice suggestion! I have got an alert before for a very dumb reason - “Spam”. That Spam was just literally me commenting F4F on some user's profiles who actually do F4F. WHYYYY!??!?!?!?!

kip23s

hotcrystal wrote:
Semi-support. If this gets accepted, bad offenses like posting a very inappropriate project should probably count towards a high number of “reports” rather than just one. (also can someone post that essay by jvvg talking about how false reporting is a misconception here)

Here

kip23s

Why do all these accounts support this and F4F? Conspiracy theory I think so.

Za-Chary

F4FNugget wrote:
At the moment, Scratch uses the number of alerts a user receives to decide when to ban them.

Not exactly. It is often the case that more alerts will increase the chances of a ban, but there's not a hard cutoff. Some people get banned without warning, while others keep getting warnings without ever getting banned. It all depends on not just how many alerts have been given, but also what those alerts were for, how often they've been getting alerts, and other things.

F4FNugget wrote:
The problem is that alerts don’t always reflect actual harmful behavior

And this is exactly why Scratch doesn't use the number of alerts a user receives to decide when to ban them.

F4FNugget wrote:
sometimes they come from false reports or misunderstandings.

Note that false reporting is harmful behavior and should not be tolerated. I see what you're saying about misunderstandings, though — they do happen sometimes — but is again a reason why Scratch doesn't use the number of alerts a user receives to decide when to ban them.

F4FNugget wrote:
15 reports → Temporary ban for 1 day

I don't think introducing automation to the ban system would be a good idea. That just means someone could report a project repeatedly in an attempt to ban someone.

F4FNugget wrote:
More direct consequences than alerts, which can be ignored.

But the difference is that alerts are sent to you, so you can see them. You can't see when someone reported your project, so you would have no sense of how close you would be to getting banned, and there would be no indication of how to prevent the report number from continuing to go up.

F4FNugget wrote:
Fairer system: bans are tied to actual reports, not just moderator alerts.

But again, false reports unfortunately exist. Bans should be tied to reports that are actually justified.

F4FNugget wrote:
Community protection: harmful behavior is removed faster.

Could you go into more detail about this one? I'm not sure if you've described how projects can be removed under this system.

F4FNugget wrote:
Transparency: users know exactly how many reports lead to which ban.

But again, users are never notified when they are reported, so this isn't a fully transparent system.

F4FNugget wrote:
This would make Scratch moderation more consistent and effective, while reducing confusion around alerts.

I'm not sure how this suggestion would change the content of alerts. Therefore it wouldn't reduce confusion.

Paddle2See

F4FNugget wrote:
Replace Alerts with a Report-Based Ban System

At the moment, Scratch uses the number of alerts a user receives to decide when to ban them. The problem is that alerts don’t always reflect actual harmful behavior — sometimes they come from false reports or misunderstandings. This can make the system feel inconsistent and less effective at protecting the community.
<snip>

I'm going to stop you right there - that's not how the system currently works. With the exception of very temporary blocks when the community reports something a lot, lasting blocks are placed by a human moderator. They evaluate how serious the offenses are, whether the reports are legitimate, how long the block should be - and so on.

Discuss Scratch