Ban Group Member/Shadowbans?

groups

#1

We have an exile feature with groups (So we can boot out group members), but I think we should have the ability to ban people from joining/posting to a group if they were kicked out.

I ask for this because a group I help manage with a friend has encountered a lot of spam lately (Mostly hate speech/meme spam), and the users rejoin the group the moment they get exiled (Or create alt accounts if they're banned from Roblox).

I don't think that users take exiling seriously, and that the ability to prevent people from interacting/joining/requesting to join a group is the next step up.

For those who are repeat offenders (Who create multiple accounts to harass users on Roblox, even after getting banned numerous times), I think that a shadowban feature (Similar to what Reddit has) would be wonderful.

A shadowban is where someone can interact with a website (Posting on the forums, group walls, in-game chat etc.), but their interactions don't appear to other users (Example: User A posts to group wall, their message appears to them, but to everyone else, it's invisible).

The reason that I ask for a shadowban feature is because repeat offender users (Trolls, users with grudges, exploiters etc.), tend to create alternate accounts/harass other users more frequently when they realize that an administrator is intervening (They take pleasure in the factt that they're provoking a response from higher authorities).

Thoughts?


'Exile' in groups with option for permanent/temporal ban
Banning from groups
#2

This has been suggested many times before and was rejected by the "just set it to join requests" reason. In my opinion, this is not a good reason. We need a ban system.


#3

It wasn't necessarily rejected, we just never received a response (as is the case with most things web-related)


#4

This is actually the response I was given by a web developer in Maker Faire New York (2014 but it may be still relevant)


#5

I won't touch on regular bans, but you suggested something to take care of regular bans' weakness, so let's talk about that. As I'm sure you vividly remember, sailor hat dude kept spamming your PMs even though you had blocked a bunch of his accounts from PMing you (a form of shadowbanning). Yes, he could see that the account was unable to PM you, but that could have been because you set your PMs to friends-only. How did he verify that? He created new accounts to see if they could PM you, revealing that he was, in-fact, shadowbanned. People could do this on groups as well: the whole reason they're shadowbanned is because they create alternate accounts, so what's to stop them from creating another account to check if what they're posing is making it to the wall or not? Once they learn that shadowbanning exists (which won't take long once it's made public to every group's admins), they'll just change how they work and it won't do too much to stop them.

I won't spend too much time elaborating on this since it's not the topic of the thread, but if you want to stop the same people from repeatedly causing trouble in your group(s), the solution is to not give the first rank in the group posting permissions, and then hook up automated promotion (optimally supported natively by ROBLOX) to the second rank once they complete some sort of achievement. The achievement should take long enough to earn (maybe a full day) that it's simply not worth harassing your group. It's not an inconvenience to players who only have to earn the achievement once, but if spammers/etc get to make only one annoying outburst per time it takes to get that achievement, they'll soon lose interest. I'm more than happy of discussing this with you over PMs if you don't want to get into a discussion about it here btw.


#6

I asked for something like this a few months ago. As best I can put it, the problem is this:

User spams/trolls a group. You exile him, he just comes back for more. You set the group to request only, he possibly makes an alt. Even if he doesn't, you still have potentially hundreds of join requests to sort out per day just because of a few bad eggs.

I support this and I hope a Roblox employee does too.


#7

Yes, setting the group to join only is a very bad solution, especially so for large groups. Group banning is also a bad solution at achieving what it's supposed to achieve though. Banning has the same weakness exiling does, in the sense that they can just rejoin (albeit on a different account). The only way to stop these types of people from harassing you is to make effort expended > satisfaction gained by harassment -- that's not objective. One way to accomplish that is with what I suggested in my previous post, and there are probably some other ways to accomplish that as well, but group banning does not. Creating a new account only takes a couple of seconds, and I remember posting a gif in your earlier thread showing that. You need something that tremendously increases the amount of effort required to harass your group, and group banning does not accomplish that.


#8

That's why I said 'something like this'. I like this idea better.


#9

Shadow banning or preventing the first rank from posting?


#10

Shadow banning.


#11

In that case, my first response sums up why shadowbanning won't work.


#12

If their goal is to get a rise out of the officials in a group then how is shadow banning a bad solution? It takes them a while to realize that they've been forced into an echo chamber. When they do realize it, they'll make a new account, which will also be shadow banned. It'll take them a while to realize that they've been banned every time and saves the officials some suffering.

The main thing you're ignoring here is the long-term effects of group banning. In the short term, it'll get a rise out of the spammer and maybe you'll have to ban a few accounts (a much easier process since they'll never quite know for sure when they've been banned again). But in the long term, you won't have to deal with that person as much. Their main account will have been long-banned and then their alts will also have been banned. Maybe in the future they'll make another alt but then that one will also never be quite sure when it's shouting in an echo chamber.

Why deny group owners of a tool that has quite a bit of utility, especially against the less well-equipped spammers? As it stands, this solution has a bit more utility than the other two (constant exiling vs constant join approval), and combines the pros of both.

Remind me what your solution was in the previous post


#13


If ROBLOX implemented shadowbanning in groups, everybody involved in groups would know about it, including the would-be targets. If they're intentionally trying to get a rise out of group admins, they know exactly when to check on an alternate account to see if they're shadowbanned or not.

Because it doesn't work.

It doesn't include the pro of join approval, and has only the strengths and weaknesses of constant exiling -- it's as ineffective as repeatedly exiling.


#14

I'll admit your system is a decent solution but you also have to worry about group owners who can't script and how do you come up with a fair task to complete?

You also have a point with the alt checking for shadow bans.

Despite your insistence that it does nothing (a debatable point), I still think group banning is worth a try. Maybe not shadow banning in particular, but some sort of group banning. Because even though they can just create a new account, that's still an extra 30 to 60 seconds that will wear down on them much quicker than just 'Rejoin, post, refresh, rejoin, post, refresh', etc...

The thing you're forgetting is that most Robloxians are young; they are not smart like you. Having these tools to make it even slightly more of a hassle to spam is, in my opinion, worth it.


#15

<< I will address the need to resolve group banning immediately in the next post, since this one is kind of long. >>


You're right. Lots of group owners can't script. That hasn't stopped any of them from having bases, guns, fort capture systems, training facilities, etc though. If someone makes a free model for changing peoples' rank in-game, all group owners will be able to use it, just like they have guns/etc. If ROBLOX implements rank-changing in-game natively, you have my word that I will immediately publish a free model / plugin that hides config behind a GUI + a tutorial video that enables any non-scripter to use it in their game.

Your guidelines are:
- When done repeatedly, is extremely aggravating
- When done once, is not that big of a deal
- Takes little enough time to complete once that it's not that big of a deal to legitimate users
- Takes enough time to complete that when done repeatedly you have a significant break before you have to exile/demote someone again

It's going to take a little planning on your part to decide what works for your group, just as games need design planning and groups need to plan out guidelines/procedures/etc.

Recommendation:

If you're looking for a recommendation of how long the achievement should take to get, I'd personally start experimenting with total gameplay = 6 hours. I can't imagine anyone grinding 6 hours in-game to be able to spam/etc on the wall for a couple of minutes. If that doesn't stop them though, you can increase that. If it does stop them, you can try lowering it to make it easier for legitimate players to get through. Even at 6 hours though, that's not asking a lot of legitimate players since they'll only be making that wait once.

Behavior is contagious. Even if we assume generally all but a small few wouldn't think to create a new account on their own, all it takes is for someone to show them the way. If they see someone using alts to spam on a group wall, that will be in the back of their head if they ever want to do the same thing in the future. Implementing a "solution" and tackling the gaping hole in it by not tackling it, and hoping and praying that people don't find it isn't going to work. That's no more sound than assuming nobody is going to exploit the ROBLOX client's vulnerabilities because only "smart people" can find them. One YouTube video / tool designed to allow "not smart people" to use the exploit and it goes viral.


#16

I totally understand why you'd want these people to be stopped immediately, and that you're latching onto the only thing in sight because of it, but group banning is really not the way to go about doing that. While it may stop some of the less persistent troublemakers in their tracks, when it doesn't stop dedicated repeat offenders (arguably the most important, as anyone less than this will only cause issues for a short while), instead of "Group exiling doesn't work! To fight harassment we need group banning!", we'll be seeing "Group banning doesn't work! To fight harassment we need AnotherThing!" To stop people from continually returning to harass groups, we need a solid solution that reliably stops them.

Again, I entirely understand why you'd want the issue to be solved right now, and that's a completely respectable wish, but that's not how software development works. You take careful time to plan something effective, and then implement it. It takes a lot less time to plan things out the first time around and then implement it than it does to rush in with hastily devised "solutions" that need to be duct-taped with more "solutions" to get something that's remotely usable. It will take a lot less time to totally quash group harassers by planning a total solution the first time around than to keep tacking gimmicks on like group banning.

If you think planning a total solution to group harassers is difficult, I imagine a good bit of that is due to the case where you're 5 minutes into thinking and "Wait, why am I trying to find a solution again? We can just use group banning!" If you took off your rose-tinted glasses, and realized the weaknesses of group banning (i.e. not just that it won't stop some people, but that it won't stop the most severe harassers), I don't imagine you'd have as hard of a time coming up with a solution. I hastily devised the idea of a buffer rank with automatic promotion in just the time it took me to respond to another thread, so I'd imagine with some planning over the course of a week you'd be able to find a solution, and one that is probably better than a buffer rank at that.


#17

A viable solution would be something that makes it a little more cumbersome for spammers to do their job. Even if everyone knew about the fact that you could just make an alt, it would still be an extra half a minute to a minute before they can get back at it, as opposed to just clicking 'rejoin', like I said earlier. As things currently stand, we are ill-equipped for combating this. Every potential solution currently just makes it harder on us as well as them. This solution, while not much, as you've said, does help a little, and hurts the spammer more than us, especially in the long-run.

The user is discouraged from going on a rampage again because it wasn't quite as fun last time and now all their mains and alts are banned. For example, I know a user who hates me. Why? Who knows. He visited my games all the time and ruined the gameplay. So I banned him from the game. Not a solution, right? He can just make an alt, right? Well, yes. He did make alts, but it wasn't as hard to deal with him when he couldn't just click 'Play' again. And to top it off, he didn't come around for a long time because his main was banned and we had power to stop him if he tried again on an alt. So I talked to him and unbanned him. He didn't spam or rage again after that. Take from that what you want.

I know this is your ENTP debate side speaking. I know mine bursts out onto the scene as devils-advocate quite often too. But you have to admit that the short term hindrance to their process and the long term discouragement from repeating it is at least handy.

Nothing currently available to us is an antidote to this problem. The only thing that would ever 100% solve this issue is the use of AI, which we won't have for another 5 to 10 years. Until then, band-aid solutions is what we get.

In my opinion, group banning is a viable solution, even if it's not perfect. That being said, I also think your hypothetical ideas are solutions as well. I rejected the idea of a buffer-rank because that just shifts the pain of manual approval into an even more painful process, unless I install some sort of stupid non-vanilla script or hire people. It's a good idea, and Roblox should support at least one of these.


#18

ITT: Let's not fight a war on drugs because you cannot stop it 100%

I don't see a bad side to this. I mean people join groups specifically with names meant to insult others, so a ban would stop that kind of behaviour at the very least.
EG https://www.roblox.com/users/119965535/profile joined a group that I am a HR in because the original owner disagreed with some group policy.


#19

Okay, just so I'm understanding this, your entire premise for wanting the implementation of group banning is based on helping ease the pain of the problem quickly, right now, to hold us down instead of leaving us ill-equipped until we find a more absolute solution, right? Even though it won't stop the most severe harassers, at least it will stop some of them until we find a way to stop all/near-all of them?

Actually I'm INTJ surprisingly.

Big difference. If we're comparing this to the war on drugs, I'd be voting against a hastily-drafted policy that would stop 50% of drug use in favor of planning a better solution that stopped close to 100%. Naturally, if we can't come up with anything better, I'd be okay with the 50%, but there's no way I'm going to rush to do something when it can be done better with careful planning. Better to spend extra time planning the best solution the first time around and being done with it then and there than to spend even more time tearing your hair out trying to duct tape a lesser solution up because it isn't working.


#20

Yes.

Swagbomb. The INTJ strengths as well as weaknesses are showing.

There is the problem. We don't want a total solution. We just want relief. Implementing something that cuts drug use by 50% would be a godsend if the goal is relief from the effects of drug use, whatever they may be. We can implement the band-aid solution until sometime, 50 years in the future, when we have a total solution. You seem like such a logical person. But you're treating humans like they're computers.

Like I said, the only total solution is AI. Until then, I think you refusing us a '50% solution' just because it's not the '100% solution' isn't cool.