Make moderator messages anonymous

Question

Currently, when a moderator messages a user, the user sees which moderator is messaging them. This includes when a moderator issues a timed suspension.

Generally, I have no problem with this. Sometimes a user needs to be gently reminded to be nicer in comments, or to disclose when they're advertising their own product, or to stop targeting their coworkers with upvotes. Usually, users are quite civil.

However, being a moderator occasionally involves dealing with some really unsavory individuals. Sometimes these people take suspensions very personally, and lash out in retaliation. On several occasions, this has resulted in some pretty intense social media stalking on my Facebook and Twitter, ranging from personal insults to attacks on my professional LinkedIn profile. Other moderators have had disgruntled users repeatedly phone their place of work trying to get them fired.

While I agree accountability is important, revealing who sends a moderator message in no way improves accountability. The entire moderator team can see every moderator message and all replies to it, and users are free to escalate moderation issues at any time to the community team.

I'd like to suggest that that we allow moderators to anonymously send moderator messages, via one of the following changes:

1. All moderator messages are anonymous
2. Moderator messages that involve a suspension are anonymous
3. Moderators have the ability make messages anonymous on a case-by-case basis

In any of these cases, mod messages could be signed "The Stack Overflow Moderation Team".

---

To head off a few counter suggestions and rebuttals:

• Changing my Stack Overflow username is a less appealing solution. My professional CV is tied to my Stack Overflow account, and I've invested a lot of time in Stack Overflow. I like having my Stack Overflow account tied to my real-world identity, as I imagine many users do. However, as soon as I was elected to moderate, I became somewhat uneasy about the imbalance of power when issuing suspensions to trolls who themselves are anonymous, and don't care about having their Stack Overflow accounts banned. These people can go from my Stack Overflow profile to my real world identity to my place of work with about 30 seconds of Googling, and generally that makes me pretty uncomfortable.

• I agree that moderators should be held to a high standard, but I'm not sure that exposing our account names (and indirectly real-world identities) to some of the more unsavory parts of the Internet helps that goal. We are still policed by each other and by the community team, and any user can come to meta to plead their case regardless of whether they know the identity of the moderator who messaged/suspended them.

• To be clear, I have no problem standing behind my actions as a moderator. This is about the very small number of users where my actions are appropriate, but their response is disproportionality abusive or outright crazy. It does happen.

• Won't this cause grief for all moderators instead of just one? Possibly; it's my hope that a disgruntled individual who doesn't have an immediate and obvious target won't go through the hassle of harassing all of the currently elected moderators. If they seek out and harass the wrong moderator, it's potentially easier to defuse the situation by being able to plausibility respond that they are attacking the wrong person.

• Regarding legal recourses for "cyber stalking" and harassment, this isn't really viable or desirable. The level of harassment I would have to endure before the police would care to get involved would be quite severe, and I'm just not interested in going through that to reach the point where I could pursue legal remedies, if any are even available.

Answer 1

Update August 7, 2017

Part 3 has also been implemented. A user is provided with an option to contact us (/contact), after they have already responded to a moderator message and did not receive anything back.

Update: June 2, 2017

Parts 1 & 2 of this have been released in the latest build. We're still working through the last suggestion, but messages will now appear to come from the moderation team instead of a single mod.

---

I agree with this but there are some concerns that need to be thought through. Moderators should always be accountable for their actions, including suspensions. We have processes in place when users believe a moderator abused their privileges, but if we make messages anonymous how will we keep moderators in check?

Any change made needs to keep that accountability in place, but should also provide some level of protection to the moderators from personal attacks. There is a balancing act between accountability and privacy, and we need to make sure we aren’t sacrificing one for the other when discussing making these messages anonymous.

While most moderator actions are public, we do have some anonymous moderator actions, including what moderator handled a flag, who imposed a review suspension, and others. When a user questions an action, they typically post on Meta allowing the moderator involved to respond publicly on why they handled a situation in a specific way. But mod messages and account suspensions have always been different. Good or bad, users have the ability to see which moderator sent a message or suspended their account. Most of the time, users take the messages in stride, but on occasion users respond poorly and/or escalate matters.

During my time as both a moderator and as a CM, I've seen far too many examples where a user freaks out about a moderator message, hunts down the moderator elsewhere (Twitter, Facebook, their place of work) and harasses them. This is very concerning to me because our moderators are volunteers who give their time to the community and shouldn’t be harassed for their actions. Lately, we’ve seen an increase in this type of behavior and want to get a change in place as soon as possible to offer some protection to our volunteers while we consider other changes to the messaging system.

With that in mind, I propose the following changes to the moderator message system:

1. The signature on the template moderator messages will be changed to say Site Moderation Team (meaning the elected/pro-tem diamond moderators). This makes the message appear to come from the entire moderator team instead of a single person. Moderators should not edit the signature and should avoid identifying themselves in the message to the user, unless they know they are already on good terms with the user being contacted.

2. While the name of the moderator will be hidden from the user, it will remain visible to all the diamond moderators on a site (elected/pro-tem moderators and SE staff). If another moderator views a message, they will see, as they do today, who sent the message. In the event, there are questions about a message or suspension, this allows all diamond moderators to see who was involved in the incident.

   User View of Message:

   

   Moderator/Employee View of Message:

   

3. In addition to the existing reply to message option, we will add a very obvious and difficult to miss means to contact us directly in reference to the message. Users will be advised to respond, calmly, to the message explaining why they believe the situation was misconstrued by the moderator, and reminded that they probably want a positive outcome. By adding this link to contact us it will give the user an easy way to contact staff, if they believe a moderator crossed the line with the message and/or suspension. Our hope is that by providing users a way to contact us immediately, it will be far easier than going to another platform to rage on.

Keep in mind that the community team already receives copies of moderator messages, which provides an additional level of oversight on moderator actions. This proposed change doesn't make the process completely anonymous. Each site always has more than one moderator which means all actions will be visible to the remaining moderators. Every moderator will still be accountable for their actions, but this provides an additional layer of protection from users who might go off the rails in response to a message. We already do this with support tickets to the team and it seems reasonable that we'd extend this to the moderators.

We are still discussing all of the details of this proposal, and there might be some changes in the final implementation, but the end goal will be to make sure that the majority of the rotten, nastiness gets directed back to us and no longer at the moderators. We'll be working on getting some of these measures in place immediately as a stop-gap while we decide how to overhaul the moderator messaging system entirely.

Answer 2

There is another reason to change this: the single name below the message sometimes gives the impression that the suspension was a lone decision by that moderator. While many suspensions are decided by a single moderator, those are usually the completely obvious cases. In any of the more complicated cases multiple moderators are often involved, and sometimes a few SE employees as well.

In a few moderator messages I've explicitly added all the other moderators to the signature as well, because we all decided it together and we knew that the users would complain about the suspension.

Signing the messages with "moderation team" could help to avoid the impression some suspended users have that a single moderator is "harassing" them.

Answer 3

I think there's something pretty important to say about the opinions that folks have expressed here, none of them are wrong.

I also want to let everyone know that we're giving this very careful consideration, and should have a decision by the end of this week. The way that our sites are moderated by the communities that they serve has always been an amazing and rather rare balance; we're determined to keep that balance while adapting to the way that the world, particularly people that use the Internet, behave.

I want to take a moment and distill out what I see to be the roots of the concerns here on both sides. When you boil it down, essentially, well, people want to feel safe, and they should feel safe. That amazing balance of moderation that I spoke of earlier is precisely what makes our sites some of the safest places for civilized discourse on the Internet.

Moderators shouldn't be abused for doing the strictly volunteer work that they do, and the incidents of this happening is on an alarming and sharp incline. It's going to get worse, and we need to do something about that. We can't fix the fact that people behave like jerks online, but for the people that give us so much of their time and take so much flack for it, we will not leave any stone unturned in order to support them.

On the other hand, as I mentioned, everyone needs to feel safe. Knowing that should you have an unpleasant or unprofessional encounter with a moderator that you're able to take the situation into your own hands makes you feel safe. It gives you some degree of control over how much authority you'll take. Just like you could really let those bozos at Buy Lots know you mean business with 20,000 shares on Facebook after they bait-and-switched you.

But that's the thing. The people that have caused this to come to a head never attacked us as a company, they attacked the people that volunteer their time to make sure our sites stay great. So there is quite a bit of merit into seeing what we can do in order to steer abusive outbursts back on us by default - and this might be the best way to do that.

We don't know yet, but the minute we do, we'll let everyone know. This is a tough spot for all of us to be in, none of us asked for it and it's super damn unfortunate that this is now a 'thing'. But we're going to take everyone's interests to heart, and quite seriously, and see what we can do.

I'm unlocking this now, please be good to one another, especially right now.

Answer 4

I agree with this and I specifically disagree with Telkitty's view that this removes "taking responsibility with your actions".

Our community moderators don't run for re-election. They're volunteers and we have placed trust in them. In cases of genuine mistakes, they generally come forward on meta. In cases where it really is necessary, the SE staff can see what they're doing. We're not talking about people who on any regular or consistent basis would maliciously abuse power.

By any reasonable metric, the kind of person who becomes a moderator is someone who is very responsible and has no issue with accepting mistakes if they make them.

There is no information that is lost by removing the name. For the vast, vast majority of people, the moderators are seen as a whole anyways. To have them face harassment over other channels or force them into hiding as much information of themselfes as possible is in my opinion highly unnecessary.

This isn't about accountability, every moderator decision is subject to public scrutiny at all times, and the team always knows which moderators did it.

Answer 5

I agree with this notion; you're doing your duty and you shouldn't be subject to harassment of any kind in the process of that.

However, I wonder if it's fair to take it a few steps further.

I'm no lawyer, but I would imagine if someone were to go to the level of harassing one's social media account(s) and ultimately be beholden to the level of cyberstalking, that should come with a bit more legal bite to it as well. (Note: this is already an option.) I would say that the user doing that sort of thing isn't the kind of person that we want in this community at all, so I would hope for their permanent removal as well.

I've never had to receive a moderator message (that I can recall anyway), and I personally would not wish to lash out against a moderator individually anyhow, but it makes sense to anonymize these messages. No sense in introducing the risk of abuse.

Answer 6

One thing to keep in mind is that sometimes it is useful to have the moderator message come from an SE employee. This has come in handy for us on Workplace several times.

I agree with the sentiment here, however, so my recommendation is to make the message come from "The moderation team" when sent by a non-employee site moderator but allow SE employees/CMs to send this message as themselves should they choose.

Also, it would be useful for other moderators to be able to see who sent messages.

Answer 7

Yes, agree with your points. Moderators becomes open and soft target for such users and specifically suspended users. Moderation is not an easy task, they have to take strict decision against some wrong actions. Their words may sound strict at first but what they are conveying is not pointless at all. So, this is what brings all moderators in a risk where they can be directly or indirectly abused in one or more ways,

• From social media profiles
• Their website or blogs
• By editing any stackoverflow post
• By downvoting post (specifically questions)
• ...

In moderator private message it should be common name like Stack Overflow Moderator and even more not keeping it only up to name we can have common moderator profile like Community where all the stuff can be described.

Answer 8

I think this only makes sense as an opt-in from the moderator that wishes to write the message. The thing is, this veil of anonymity is probably not going to work very well.

I've seen this in action with Impraise, where all of the feedback you receive is free-text, but you do not "know" who wrote what. In reality, it's usually pretty trivial to tell. There is science behind that: this talk claims that even with something as dry as source code, even when you ignore the variable names and only look at the structure of the code of the program you write, even when you make it obfuscated, a computer is able to attribute who wrote what between a pool of 20 authors or so with 80% accuracy in the worst case scenario.

In other words, unless you pick a template and don't touch it, you're not going to be "anonymous."

I would suggest that if you feel uncomfortable at all contacting a user — that's what other moderators and community managers are off. A veil of anonymity is probably only going to give you a sense of false security and the receiving user a burning urge to discover who wrote what rather than focus on the message that has been received.

Answer 9

Well, firstly I think this is a huge issue that points to a glaring problem: Stack Exchange isn't protecting its moderators.

I think that the best solution is for Stack Exchange to employ their customer service, and in some cases their legal, teams to stop abuse and harassment. Once these bad users find out that Stack Exchange protects its moderators the abuse should be cut way back. Perhaps the police wouldn't become involved due to the low level nature, but this doesn't mean that Stack Exchange can't get involved.

The stories told in the OP sound huge warning bells - does Stack Exchange really do nothing when its moderators are being this harrassed to this extent? It boggles my mind that the only possible solution is to hide them behind a curtain like the Wizard of Oz.

That's the best solution? It really does sound like a huge failure of Stack Exchange.

But if Stack Exchange can't be reliably expected to respond to moderator harassment, then I disagree that reducing transparency to this degree is a good choice. Too much of Stack Exchange is now handled behind closed doors/chatrooms as it is, and the community is often only involved once decisions have already been made.

I can see how this is a significant problem, though, and given that our moderators are volunteer positions we don't want to leave them open to such terrible abuse.

So I'd say "no" to a blanket obscuration of the sole moderator responsible.

I'd suggest instead that when you want a moderation message, ban, or other action to come from the "team" with no single moderator being shown, at least two or more moderators should perform that action. If it is a single moderator performing an action, then that moderator should allow their name to be used. If they are dealing with a user who poses a threat then they should work in pairs or greater to "vote for" or sign off on the ban. Only at that point would it be labelled and treated as coming from "the moderation team".

So this would essentially modify option 3:

4. Moderators have the ability make messages labelled as coming from the team on a case-by-case basis when approved by a second moderator team member.

I would then further suggest that all significant moderation events be performed by pairs. I don't know what the system looks like now, but having a moderation queue where significant events that moderators believe should happen anonymously or have additional oversight would go into, and another moderator would then be able to approve it, whereupon the action would take place would make the flow go more smoothly.

I'd prefer the messages to be signed by the multiple moderators rather than a blanket "team" signature, but I suppose at this point it really doesn't matter as long as two or more sign off on the action.

Honestly, though, if a moderator experiences even mild online harassment, Stack Exchange should be johnny-on-the-spot redirecting the user to customer support and away from the moderator, even to the degree of legal threats if the abuse continues. We cannot allow our moderators to be treated like this, but we also cannot reduce transparency.

Answer 10

I am, for one, against this proposal. If you have used your power, at least take the responsibility.

If a member of the police force guns down someone who might or might not be a suspect, (s)he shouldn't ask the whole police force to take responsibility for it. Any misjudgement was on that person's behalf, not a reflection of the whole police force.

I have witnessed once that a CM was hiding behind a moderator. It took me a while to figure out what was going on. Surely you don't want to be a scapegoat for someone else in your team?

Personally I agree on the message being anonymous if two or more moderators reach the same decision.

After all, the community is not a single person.