Jessamyn on moderation

January 22, 2016 8:46 AM

Advice for the accidental community manager at The Coral Project ("a multi-year collaboration between Mozilla, The Washington Post and The New York Times designed to lead experiments in community growth and management, seeking ways to improve the state of user-generated contributions on the internet through open-source software.")
posted by beagle to MetaFilter-Related at 8:46 AM (18 comments total)

This is a nice piece, which is no surprise coming from Jess. And I think that core bit at the end—that having moderators with well-developed soft skills and actual community engagement trumps any tech toolset—is a key thing that doesn't always come through with community moderation that's built after the fact. You can hand anyone the keys to the car, but if they aren't a careful and attentive driver then giving it more horsepower won't help.

Which I say as someone who loves the toolset we've been able to build and loves pb for building it for us.
posted by cortex (staff) at 8:50 AM on January 22, 2016 [1 favorite]

(did anyone else subscribe to the coral email group when it was mentioned here a few weeks ago? i just realised i haven't seen a single email from them.)
posted by andrewcooke at 8:59 AM on January 22, 2016

To use the local vernacular here in Somerville, MA, Jessamyn's wicked smaht.

Tools are good, but it's hard to have purely technical solutions to non-technical social problems. Not that this hasn't kept a lot of people from trying.
posted by rmd1023 at 9:05 AM on January 22, 2016 [3 favorites]

I enjoyed writing this! I like watching what the Coral Project has been able to come up with. It's exciting because they have a combination of smart people working for them along with $$$ and buy in from big names you have heard of which helps.

That said, some of my UX with them has been a little rocky. Probably nothing to be unexpected but I wrote this article for a November 1 deadline (to be published around then) and it just bumped around as they got their blog going, went through some staffing changes, and a few other things. They have a newsletter which is great and really worthwhile (suggest subscribing) but they put their code of conduct up and asked for feedback on it via the newsletter (yay) but it was on Github and the only way to do it was via pull request (boo).

Which is not to poke at them too much only to say that even with the most dedicated and decently funded people, this stuff is still HARD and I'm looking forward to watching them assemble tips and tools and strategies. Sometimes I think one of the reasons MetaFilter is able to do so much is that there are some restraints on funding and time available so development is geared towards those restrictions which I think makes people look for creative ways to solve problems. Sorry for not calling everyone mod by name (especially cortex). I was working with fierce word limits (for a blog? why?) and had to pare down and pare down. Apparently I have a lot to say about this topic.
posted by jessamyn (retired) at 9:50 AM on January 22, 2016 [5 favorites]

Oh, god, the scale problem. That's the #1 problem in community management, imo. It's so far unsolved, I think - there's no easy way to prevent the workload from scaling geometrically with the active userbase.

And re: tools, yeah, ours *seem* minimal, but I have made strong women cry by showing them our admin panel. Try moderating a 100+ member Wordpress commentariat, or a large community on VBB. Our tools are so much better, it makes me want to burst into song.
posted by restless_nomad (staff) at 9:55 AM on January 22, 2016 [9 favorites]

I remember when I was a "guest editor" on BoingBoing for a week back when they had comments on-site and learned that, except for one disemvowelling plug-in they had nothing, nothing!, for dealing with any serious moderation issues. Just a Movable Type back end like any other normal blog. You could delete a comment. You could make a comment. You could, I think, hide a comment.

And scaling is only a problem if you need to make money. Like otherwise just hire more moderators when you get more users. But income (usually) doesn't grow proportionally with userbase and not needing more moderators is where a lot of sites want to make some of their money. And moderation stuff is generally not that public so you have to sort of read between the lines a lot to see it happening. I remember when Twitter was talking about making their abuse team larger and it was really clear that the abuse team got 30% bigger when the userbase got 100 or 150% bigger (don't remember numbers but it was something bullshitty like that).

One of the reasons people are paying The Coral Project is because they think they'll be able to find them some scalable moderation tools so they can get more (or lose less) money. I wish them a lot of luck.
posted by jessamyn (retired) at 10:11 AM on January 22, 2016 [2 favorites]

jessamyn - although it sounds complicated, i think that in practice that (github, pull, etc) means that you can go to the github page, click the edit button, change the text, and then submit the change. you don't actually need to use git (but you do need to be a member of gihub) - behind the scenes github will translate the web edit into a pull request.

but i do this for a living and find it opaque.... (and i just checked and indeed, this seems to work, but the UX is horrible because the "help" text is all written using terms like pull request, so you still need to understand that side of things just to be sure the "simple" approach is what you want).
posted by andrewcooke at 10:17 AM on January 22, 2016

behind the scenes github will translate the web edit into a pull request.

Yep, I get it. But if you don't want to edit the document or just make comments, you do what exactly? (not asking you or anyone, just posting a rhetorical question; I am actually good at computers) My point is there are multiple layers of messaging in a simple "We want your feedback!" request

A. "We really want input from all sectors and that is clear by our choice of tools and feedback mechanisms that we select"
B. "We want some narrowband feedback so we're going to use the tools that the community we're querying uses"
C. "We're all techies and so we're going to use the tools that seem normal to us" which can look a lot like
D. "We're actually looking for feedback from people who are more or less like us while using words that imply that we are looking for feedback from people not like us"

I think most people know that D is poor form, but they sometimes do C by accident when they're aiming for A even though no one would be mad if they just did B from the get-go.
posted by jessamyn (retired) at 10:32 AM on January 22, 2016 [11 favorites]

there's no easy way to prevent the workload from scaling geometrically with the active userbase.

I remember from a teacher talking about class sizes: the difference between say, 19 kids and 20 kids was not 1 kid, but 20 more relationships.
posted by the man of twists and turns at 11:06 AM on January 22, 2016 [30 favorites]

(did anyone else subscribe to the coral email group when it was mentioned here a few weeks ago? i just realised i haven't seen a single email from them.)

It looks like they sent out the first one in a month today.
posted by beagle at 11:49 AM on January 22, 2016 [1 favorite]

This was a good post, Jessamyn. Thanks for sharing it, beagle.
posted by Conrad Cornelius o'Donald o'Dell at 7:50 PM on January 22, 2016

This should be required reading for anyone getting into web-based community management. Great read for rank-and-file members, too.

We're really luck to have had Jessamyn here for so long (and still!).
posted by Joseph Gurl at 7:59 PM on January 22, 2016

"....it makes me want to burst into song."

And just why DON'T we have "MetaFilter- The Musical" yet? Is someone slacking???
posted by HuronBob at 3:24 AM on January 23, 2016 [1 favorite]

Like otherwise just hire more moderators when you get more users. But income (usually) doesn't grow proportionally with user base

Worse than that, you can't just hire more moderators proportionally to the user base past a certain point. If you want consistency in moderation, then you need some coordination and oversight. I mean, what if one mefi mod decided they didn't care so much about self links? I think mefi has stayed small enough to make it practical, but at some point you either go the subset route where the standards vary depending where you are and what mod you happen to get, or you have to have this big apparatus of oversight and appeals process.
posted by ctmf at 1:27 PM on January 23, 2016 [1 favorite]

A thoughtful, helpful post for those of us who are in community management (and I am definitely an "accidental" community manager, though now I guess I've been managing my company's social community long enough for it to be deliberate now), as usual. I will use this with my team.
posted by xingcat at 6:00 AM on January 24, 2016 [2 favorites]

And just why DON'T we have "Metafilter- The Musical" yet? Is someone slacking???

How does a static, old-school, page on the web with some
subsites, made at the tail end of the
nineties with some cat scans and bean plates
with membership for five bucks
end up as a community exemplar?

posted by rmd1023 at 11:39 AM on January 24, 2016 [5 favorites]

True fact: got my first job in community management in part by talking about, rather effusively, about metafilter in the interview.

Still keep coming back to a lot of the things I picked up as a member here, years later.

Thanks for highlighting the article, definitely an interesting read.
posted by slimepuppy at 1:06 PM on January 24, 2016 [1 favorite]

Advice for the accidental community manager: First, be smart from the very beginning.

I'll show myself out.
posted by duffell at 10:38 AM on January 27, 2016