Tips on community behaviour issues for moderators
From Dar
Arising from a discussion on the E-Mint mailing list in June 2007
The original question asked for tips on dealing with with deliberate troublemakers.
Contents |
Well-being
Advice for moderators who wish to remain sane:
- "They'll get bored before you do."
- If you have no means to block at browser level (although even that doesn't keep out a savvy and determined pest) just keep deleting new names and don't stress about it.
- As long as the rest of your community can see that you are repeatedly taking action they'll know you're doing your best to protect them.
- Use your community to help you. Do you have any volunteers? If so, ask them to watch out for the troublemaker. If not, post to your members asking them to alert you as soon as they're suspicious.
- Smile from the wrists down. No matter how much someone gets to you, don't let it influence how you deal with them.
I think although we internally have the tools to locate people, your communities will often just "know" that the banned user has returned outright and so can alert you if you haven't noticed it yourself. It may also give you an indication of how close your community is and what they're willing to put up with.
However, if you just have an out-and-out troublemaker, you need infinite patience.
Feeling protective, territorial
Moderators can take great personal umbrage at any attempt to disrupt or devalue the Community.
Acknowledging and understanding your boundaries with regard to territorial feelings is part of the growth of a successful community professional
It's not only moderators who feel territorial. Users have the right to enjoy a community where people aren't being anti-social. It helps lurkers to move to commenting or becoming more engaged. Online communities are not the real world - we expect owners of them to "police" them.
Sometimes users have much higher or stricter standards than us as to what should or shouldn't be tolerated. They feel passionate about the communities they are in.
We instinctively feel that community is a fragile thing, but that can lead to overreaction. Often the burgeoning dispute which appears to be threatening to tear a community apart one day, looks like a storm in a teacup the next. The art is to hold back and let things run their natural course. It isn't easy.
Contexts
As with everything regarding interaction between humans, context is key.
Some people don't realise that the rules do actually apply to them as well.
I don't think there is going to be a simple recipe or toolkit which will work in all situations.
Ethics
- There is obviously a balance to be struck between management/moderating and exerting undue power from the centre.
- Abusing a site is unethical too.
- I find it unethical to administer sanctions against an online community participant without informing them of what the sanction is, how long it is to last and what they were supposed to have done wrong in the first place.
- trying to be reasonable can make things worse with attention seekers.
A first principle is to treat people in a way that you would hope to be treated yourself. Don't think it's unimaginable. One day you might join a new community, ask a perfectly straight question and be accused of trollish behaviour. How would you expect the moderator to treat you?
Be aware that established communities can be quite nasty to anybody who is a bit different, sometimes.
responsibility
Moderators have a responsibility to "keep the party clean"
Strategies
Redundancy
The end goal is having a community that runs by itself because the rules are implicit, the people apply them to themselves before submitting content and are also vocal in the community to any new members to ensure they are aware of the guidelines too. They are also not shy about alerting you to 'bad' content or people affecting the future of the community because they have ownership and know that something will be done if they report it.
These are some of the ways I think you can achieve this.
React quickly to reported content
One of the keys is how you react to reported content and also your day to day moderation. React quickly to reported content and be seen to take it seriously. That doesn't mean immediately removing it (let's face we know people abuse this feature), but if the guidelines are applied in a fair and consistent manner and actions are visible then it reinforces the guidelines and the knowledge that there is someone behind the scenes who will react when called on. Make your moderation actions visible if possible either on the site or if not possible due to the nature of the ugc - ensure they know why something's been removed e.g. email response, auto reponse, note n the submission itself. If content is removed without a user knowing why (or worse - banned) how would they be able to learn 'better behaviour' in order to be accepted into that community or take part in the advertising campaign etc etc.
Right to reply
This in my mind is crucial to the longevity of a community/consumer driven campaign/social network etc. Moderation is the application of guidelines and relies on interpretation of guidelines based on experience and instincts. If a user has the right to challenge a decision or a the very least ask why it is an excellent opportunity to engage with the user to explain the reason. Sometimes at this point a decision may be reversed and that can actually be a very positive outcome. It can also help refine guidelines for the future. Two way dialogue is a wondeful thing ;)
further reading
An excellent ugc whitepaper that eModeration co-published with Jake McKee (www.communityguy.com) which covers a lot of moderation issues but aimed squarely at the ugc campaigns and branded social media aimed at consumers.
http://www.emoderation.com/news/press-release-ugc-whitepaper-released
disruption
One analogy for trolls/idiots/disruptors is that if they are like a fire, then starve them of the oxygen of attention and they will die away. Educate your community not to take the bait, quietly remove and thwart their attempts to disrupt, and do not show signs of perturbation at their actions. The more fuss they have made of them, the more they will play to the attention.
Useful article by Cory Doctorow from Boing Boing " How To Keep Hostile Jerks From Taking Over Your Online Community" "If you want to fight trolling, don't make up a bunch of a priori assumptions about what will or won't discourage trolls. Instead, seek out the troll whisperer and study their techniques."
Summary of useful points / tips from this article
Trolls can infect a small group, but they really shine in big forums. Discussion groups are like uranium: a little pile gives off a nice, warm glow, but if the pile gets bigger, it hits critical mass and starts a deadly meltdown. There are only three ways to prevent this: Make the pile smaller again, spread the rods apart, or twiddle them to keep the heat convecting through them.
Making the group smaller is easy in theory, hard in practice: just choose a bunch of people who aren't allowed in the discussion anymore and section them off from the group. Split. Or just don't let the groups get too big in the first place by limiting who can talk to whom. This was Friendster's strategy, where your ability to chat with anyone else was limited by whether that person was your friend or your friend's friend. Users revolted, creating "fakesters" like "New York City," whom they could befriend, forming ad-hoc affinity groups. Friendster retaliated by killing the fakesters, and a full scale revolt ensued.
Spreading the group apart is a little easier, with the right technology. Joshua Schachter, founder of del.icio.us, tells me that he once cured a mailing-list of its flame-wars by inserting a ten-minute delay between messages being sent to the list and their delivery. The delay was enough to allow tempers to cool between messages. A similar strategy is to require you to preview your post before publishing it. Digg allows you to retract your messages for a minute or two after you post them.
But neither of these strategies solves the underlying problem: getting big groups of people to converse civilly and productively among themselves. Spreading out the pile reduces the heat -- but it also reduces the light. Splitting the groups up requires the consent of the users, a willingness to be segregated from their peers.
Troll accusers
Once the pattern of accusing outsiders of being trolls is established, it is hard to break, even if it's only a minority who are doing it. The 'troll accusers' can be just as damaging as the destructive trolls.
Sometimes the moderator needs to take a stand agaisnt the lynch mob too.
Sanctions
Very occasionally intervention is necessary. In one opinion this should be done in full view of the community, not through any back channels and with the unacceptable behaviour and sanctions clearly explained.
Sanctions should be clear and part of the joining instructions or "terms and conditions" a statement which establishes the norms etc.
TOS
"Terms of Service"
From previous discussions about how many people read terms of service, there is no way that you can force someone to read TOS, and act on them, you can only force them to click a box to say that they have.
Reams and reams of TOS are also far less likely to be read and are always subject to many interpretations by users who claim their behaviour was not breaking the rules.
Quarantine
Technology
How do you prevent someone malicious from just rejoining when you ban them?
- Persistant trolls who seem to be happy spending their entire lives setting up fake ID's in order to be trollish.
multiple account cookies
"Tachy goes to Coventry list"
If you add a user to this, their posts will not show up to anybody but moderators+ but the user posting will still see them and think that they are simply being ignored.
This was considered "The best disciplinary feature [of a certain software]"
and also "highly unethical"
- Some like the idea of making the trolls posts invisible as nothing annoys trolls more than being ignored.
Is it the moderators job to annoy people?
Slow server
- For an identified offender, the system can be set up to serve the Web site to that person only verrrrrrry slowwwwwwwly (like dial-up on a bad day). Eventually they get bored with the endless waiting.
Some people considered this unethical too.
Disemvowelling
There is a technique called "disemvowelling" -- removing the vowels from some or all of a fiery message-board post. The advantage of this is that it leaves the words intact, but requires that you read them very slowly -- so slowly that it takes the sting out of them. Disemvowelling part of a post lets the rest of the community know what kind of sentiment is and is not socially acceptable.
