You can't expect a site to be self-moderating with click-buttons (Comment is helpful/not helpful). You can't expect machines, no matter how cunning then scripts to do it. You can't ask people who have other jobs to do it (e.g. the reporters and editors on a newspaper's site).

What you need is a dedicated crew. Yog and his Minions. With the power of high and low justice, and access to the back end. And you pay us.

Tell you what, Mr. Denton. Let me reassemble the crew I had on GEnie back in '92, pay me what GEnie was paying me then, and I'll whip any of those blogs you don't dare enter into shape.

I was just wishing for a handy index of these posts! Thanks.

This is where you guys and I get to be smug for a moment, right?

The problem is Denton and his chums, and anyone who listened to them, cheaping out on moderators.

Highly-skilled handwork by talented craftspersons costs money. Expecting it 24/7 costs more. You want your comment threads to be something other than an endless plain of toxic waste and gang violence? Step up and pay to hire the cops and garbage collectors and firefighters and EMTs.

And yeah, John, we get to be smug.

John Scalzi @ 3... Yes.

@3

There's a reason I hang out here and at Scalzi's place rather than in the comments section of my local newspaper.

The remainder of the problem is left as an exercise for the reader.

Newspaper comments tend to 'moderated' by robots that only know certain words must be banned - and don't always do that well: I've had a couple of comments that were perfectly printable in a family paper, and yet the robot wouldn't let them through.

On the other hand, the robots will allow comment through that would be disemvowelled here, and the commenter probably put on the don't-ever-come-back list.

All that moderation-by-machine gets you is a class of trolls who are adept at gaming the system.

Contrast strong human moderation at, say, Absolute Write.

obXKCD for what really great comment management robots might lead to.

Congratulations and a well-deserved moment of smugness to the lovely moderating staff, here and in other places that have one.

This site has a warped perspective. The commenters here are relatively few and very intelligent. A newspaper site might get a thousand times the comments, most of them from mouth-breathing morons.

If you tried to apply the Making Light model of omnipresent human moderation to a large site it would crumble. Your solution does not scale.

That's not to say that there isn't a solution. Denton might still be wrong. But don't dismiss him outright; he has a different perspective on the problem.

10
The newspaper site I was visiting claims to have human moderators. I haven't seen much evidence of that.

I spent two years moderating the SFRT at GEnie, with many times the traffic and many times the commenters we have here, using relatively crude tools.

I spent a year moderating the McClatchy comment threads. (That's a newspaper site, with the number and type of commenters you'd expect.) By myself, without staff, and without access to the back end. Been there, done that, proved by the experimental method that it does work.

This solution does scale. Just add more moderators.

I dismiss Denton, and mock him. He's trying to cheap out, and that doesn't work.

If this place is a warped perspective, let me stay warped.

Oh, and a plug for Ta-Nehisi Coates' blog. He moderates so well, his regulars moderate each other (and discourage commentary towards trolls) when things get heated

Remus @10: It's worth noting that several of the mods here have experience modding at places with much more traffic, e.g. boingboing.

And there are publications with large readerships who seem to manage this stuff reasonably well. Most of the NYTimes threads I've been in recently, for example, seem reasonably civil and intelligent. Sure, it's nothing like the carefully curated conversation at ML, but there's still lightyears of difference between that and the sort of fetid swamp you see in woefully inadequately moderated spaces.

My local paper decided that they had a problem with their comment section, the comments sometimes veered into flames. Not too bad, but theres no politics like local politics. So,the decided they wanted realnames, and they'd use facebook for that. There were a couple epic multi hundred comment threads with people taking that decision apart using many arguments that are old hat here. There were also a few "goodbye, thanks for the fish" flounces.

Now, they don't have a comment thread problem, but that's because they have maybe five comments a week. Down from hundreds.

Aaaand I just explained elseweb about the difference between anonymity and pseudonymity. Speaking of Thingness 101. It would be awfully nice if we could stop doing that someday, wouldn't it?

Remus:

I've been at Metafilter from when they had 2000 members. Now they have over 12,000 active posters, and more than 30 front page posts per day. It IS possible to have a site where large-scale intelligent, reasoned discussion and debate take place. It has its collective blind spots, especially in diversity issues, but it's also gotten a lot better about feminism over the last 10 years. And, hey, people get in bitter fights, but that will happen anywhere human beings coexist.

The fact that I still hang out there, 10 years later, is because the site owner addressed the growth of the community as it happened -- developing a culture of community self-policing, limiting the number of signups, and eventually adopting a model of omnipresent human moderation. It works.

I have to say that I largely agree with Remus @10, although I think that assuming that the world can be divided between "the few and very intelligent" and "mouth-breathing morons" [1] is entirely erroneous. Writing articulate blog comments is a skill that requires practice [2], and most of the people who read online newspaper articles and have opinions about them are intelligent people who do not have the benefit of that practice. Thus, you will get many comments that contribute a little to the conversation while re-hashing "101" sorts of things, while often failing to directly respond to other comments in a way that makes for coherent threads.

These are not trolls. They are not morons. The comments they write are not the sort of thing that one would filter out as harmful. But they do not make for things that are really worth reading, either.

What we have here on Making Light is, make no mistake, an elite community. (Elite in the better sense of the word, not elitist.) This is not a cross-section of the public; this is a self-selected subset of people on the far end of a bell curve who have put in a significant amount of practice at writing articulate comments, or at least are willing to. The site is clearly for that sort of conversation; people who do not value that sort of conversation will not seek it out, and will not stay here. We can see this in part in the small number of one-off commenters; most people who comment here do so regularly and frequently.

Whatever is (I think) less of a close-knit community than here, but it still has the feel of a community where a large number of the commenters are regulars.

I'm not convinced that moderation is enough to convert the sea of inoffensive-but-not-quality-conversation posts that I've seen in a lot of news comment threads into a community of people who value being articulate enough to work for it, unless you moderate people to high standards of quality well beyond the point where a person can reasonably assume that their comments will be posted. (That is, something in the direction of the usual "letters to the editor" page, though not that extreme; Shorpy is perhaps a reasonable example.) It seems to me that useful conversation requires a context where there is a strong sense of community among the commenters, and I simply don't see how you can get that when the bulk of your traffic comes from Google News searches and the like.

[1] As an aside, I'd note that "mouth-breathing" is the sort of ad-hominem that, like "lame" or "gay", sweeps up people it doesn't intend to sweep up in rather an offensive way.

[2] And what we have in the Making Light commentariat is, beyond intelligence -- and let us not delude ourselves into self-congratulatory elitism with that; the world is full of equally intelligent people -- astounding amounts of practice at writing comments.

Remus Shepherd @10 - Might I point out you're missing a part of the argument.

The part you're missing is that this isn't a new problem for the newspapers either. It's something which is soluble by exercising editorial control and choosing content carefully - something which used to happen as part of the "letters to the editor" section of every newspaper. Yes, it did mean that someone on staff had to go through the letters received and decide which ones were fit to print, which ones were relevant, and which ones would fit in the space available in today's paper. But again, it's about human intervention, human moderation and editorial control, and accepting provision of supervision as the cost your organisation pays for providing a playground for the public.

If you can't afford to employ someone (or several someones) to moderate your various comment sections, then you need to cut back the amount of comment sections you provide to the number you can afford to moderate, or send the comments to a single spot where you provide all the moderation in the one place. This is the model which is being used by the ABC here in Australia on their website (www.abc.net.au). News articles generally don't allow comments. Where they do allow comments, they're moderating the comment threads (not particularly strictly, I grant you) and generally shutting down the comments if things get too vociferous or obstreperous, or where the comments are repeating themselves. Otherwise, if you want to comment about something, you go to their "public playground" section - The Drum - which hosts their "write to us" section (otherwise known as "letters to the editor", I'd guess).

Public comments sections need to be supervised for the same reason public playgrounds need to be supervised: if you don't provide supervision, things get nasty.

Having read the posts between @10 when I started writing my reply, and when I posted it, I note the mention of some counterexamples (McClatchy comment threads, BoingBoing) that are more like newspaper sites and which have useful comment threads.

So, perhaps I am merely being myopic, rather than right, in my claims.

I am curious, though, if these sites do develop communities among the commentariat, or if their broadness (and ephemerality) of readership prevents that -- and, if they do develop communities, how that happens. And, more generally: Is it possible to have worthwhile conversations in an environment where a bulk of the comments are from drive-by readers, or do you have to structure things so that drive-by readers don't comment enough to be a large part of the conversation?

Man, Gawker is so not the example to use - and I know that the interviewer was not, let us say, totally sympathetic to Nick Denton's perspective.

I was at a different panel at SxSW full of Gawker writers discussing, among other things, the difficulty of making a cogent point in your articles when the financial incentive is to write incendiary headlines and not much else because their bonuses are based entirely on the traffic they generate. Gawker as a business is not interested in intelligent debate. They're interested in ad impressions.

I've been reading blogs since at least 2001 and it became apparent, even way back then that human moderation was a must. Bit like all things that are worthwhile, there's an investment necessary to make them work. Especially something like human moderation, which is invisible when it works properly. But trying to explain this to bean counters never works. If they can't see the effects, that function is the first to go whenever there's scaling back to be had.

@Remus Shepherd #10
If you tried to apply the Making Light model of omnipresent human moderation to a large site it would crumble. Your solution does not scale.

I give you AbsoluteWrite.com/forums/

197,689 threads
6,776,400 posts
40,618 active members who have logged in within the last year
40+ active moderators. All human, even the battery, the dog, and the cat.

Human beings with good will, a modicum of wit, and the well-being of the conversation and the community at heart are amazing assets who make the world, and the community, better.

Having been a programmer in charge of comment engines at one time or another, I'd love to hear any opinions that Yog and other moderators might have on the design of moderation tools. Are there obvious improvements to be made?

I'd rather spend time making life easier for the moderators of the world than manufacture even more dubious auto-moderation AIs.

Of course, maybe the secret toolset is nothing more than "a readable font, an Edit button, a Delete button".

The first, simplest, cheapest, easiest thing that anyone who wants to have worthwhile comment threads can do is put new replies at the bottom of the thread.

There are only three blogs/forums that I will spend any time on. I don't have time or mental energy to waste on the garbage I have seen on some places.

All three places have good moderation and smart and friendly commenting. You must have the former to have the latter. It is no guarantee, but it is a necessary prerequisite. (did I just use that word? sheesh)

Oh, those three sites are:
ML
Whatever
Absolute Write

I, too, would like to give my heartfelt thanks to the mods here who have given me a nice place to hang out. I'll try not to track mud on the carpet.

The CNN headline is perhaps less negative than Jim interpreted it. Remember Betteridge's Law: "Any headline which ends in a question mark can be answered by the word 'no.'"

James Macdonald at #12: "The solution does scale. Just add more moderators."

I disagree. It doesn't grow at the same rate as the other aspects of maintaining a website. Broadcasting a message and taking messages in don't scale at the same rate.

If I gave a talk to three audience members I could have a conversation with all of them at the end of the talk. If I gave a talk to five hundred audience members, I couldn't have a conversation with all of them at the end of the talk. My listening won't scale as well as and at the same rate as my talking, so my communication will become more one-way.

Dealing with reader comments is, I suspect, analogous to that in terms of growth rates.

If it were less late at night I could probably say this more articulately.

Not to disagree with the general back-slapping at all, but I don't think this is really about how to make the internets a smarter place to live.

I don't think the point here is to seek out ways of controlling real commenters but to justify introducing created commentary. The pre-selected commenters (an idea not rejected out of hand) will make it look as though a discussion is taking place and persuade the non-commenting types into believing that what they're witnessing is a troll-free thread, while actually excluding those who might question the presented/accepted narrative too much.

Fox news seems (from a distance anyway) to prove that many people like their news and opinion without the messiness of there being alternative news and opinion, but shutting off coments altogether tends to make a site a less interesting place for people to gather (there's no way of knowing that other people feel the same as you do and so no illusion of community). Moderation (as a way of controlling the opinions presented) risks having people spot that certain comments get removed and so can become a source of conflict (which again unsettles those who like to pretend they want the news but only want the olds). But if you have pre-chosen commenters... that's simply ensuring there are no annoying trolls and loonies, and if certain questions never get asked that's because the selected commenters didn't happen to ask them, not some kind of censorship. (Plus certain questions can be asked with the plausible deniability of a commenter being at fault not the outlet itself).

Yup, my paranoia is showing -- but I suspect advertisers would be very much happier to put their money into sites that run this way.

Despite the infrequency of its use, my moderation policy has always followed what I call the Marion Ravenwood Model: I would rather watch it burn than let the Nazis have it.

Haters? Get the Katt Williams treatment. If they're persistent, they get another dose. Plus banning.

Of the Gawker network sites, io9 is the best. Not just because its content is great, but because Annalee, Charlie, and the others are smart, sensitive humans beings. Moderation is one big Turing test, and they ace it every time.

#28 @Erik Nelson

Broadcasting a message and taking messages in don't scale at the same rate.

If I gave a talk to three audience members I could have a conversation with all of them at the end of the talk. If I gave a talk to five hundred audience members, I couldn't have a conversation with all of them at the end of the talk. My listening won't scale as well as and at the same rate as my talking, so my communication will become more one-way.

You pick moderators from the active members of the community. You look for the people who are already modding without having access to software to make their actions more effective. You look for mods who mod by force of personality, and communication skills.

You don't have all of them responsible for the entire universe; you assign them niches. You pair them up in teams of two or three. You let them support each other and figure things out. You don't second guess them.

You give them tools, and a place to communicate and support each other. You are honest about what you expect. You back them up when they have to make hard decisions.

And you have someone who's willing to say "The buck stops here."

You encourage a culture based on writing truth, and being true to your words.

This isn't new. It's essentially what's built into the back-end of all the LMSs ever, with the ethos of scholars everywhere as a basic truth. But the thing that's hard is finding the mods who are right for each other, and the community. That's black magic, that is.

#25 - not only new comments at the bottom of the thread, but disable threaded comments. Force everyone to post in a single column like y'all do here. For some reason it seems sites that support threaded comments are more prone to trolls.

Denton at least grasps that he needs to get the commenters to share a common culture, though he clearly has no idea how to imbue this.
Emily #17 is right- MeFi is another good example. We talked to Jessamyn West about this at TummelVision, also how Heather Champ and George Oates did this for Flickr too.

Okay, some of the personal anecdotes have swayed me, but I'm not completely convinced yet. Jim's experience that he listed in #12 is the most convincing.

But anecdotes about Metafilter and AbWrite don't mean much -- moderators there are volunteers. If you have enough volunteers you can do anything. But for-profit corporate sites can't use that solution.

Or can they? I wonder if it's been tried. BoingBoing is a for-profit (if low-comment) site, I don't know if they pay their moderators.

In any event, the Denton perspective is that the solution doesn't scale for large corporate sites. He stated that Gawker gets 2 million unique visitors a month. That not something your staff can moderate in their spare time, and hiring an army of moderators hurts your bottom line.

I believe there still may be a solution, but I don't think strict all-human moderation is the answer for large sites. Hopefully someone will figure it out soon.

But anecdotes about Metafilter and AbWrite don't mean much -- moderators there are volunteers.

Moderators at Metafilter are not volunteers - I'm one of them. We're paid very, very well, actually. (Until last year we had one volunteer who occasionally checked in during US nighttime to make sure the site hadn't burned down, but that was it. We now have a full-time Europe-based mod for those hours.)

I dropped a comment on the CNN thread.

One thing I pointed out was that operations such as Denton's depend on minimising costs. They get what they pay for. And the advertisers will eventually notice that they're not getting as good value from confusing the subway station with the sewers.

Are these things sustainable, or is Denton's business part of Bubble (n+1)?

Think of ML as an instance of emergent behaviour. Some of the specifics of this place might not scale: we have a community and the moderators are part of it.

What we do have works.

What Denton, and many newspaper sites, has isn't working. They don't need to have an emergent community, but if you genuinely want debate you have to allow different opinions. You have to distinguish between the bad idea and the bad person. And, however you control the debate, you have to be trusted.

ML is a special case: we have seen the moderators and they are us. That isn't likely to scale. But, because we know Jim, we are able to believe him about the possibility of doing effective moderation on these large sites. But, when Denton talks about 2 million unique visitors a month, I have questions to ask.

1: How many of those visitors are advertising spammers?

2: A visitor is not necessarily somebody who makes a comment, spammer, troll, or "real". If you're talking about comment threads, shouldn't it be comment numbers that matter?

3: What's the long-term uniqueness? Do people come back? Is it the one-offs or the repeats who give you problems?

It's arguable that Denton can't talk about some of this without revealing "commercially confidential" information, but the figure he gives is actually pretty useless, whether you're judging the value of advertising on the site, or the costs of moderation. And it is rather hard to hide the number of comments made. It's not hard to discover that there have been at least 780 comments published here this month. I could pretty easily find more by checking all the active threads at the end of February.

So rather than use an auditable figure for comments made, which is the amount of work to be handled for a moderation system, he's used something almost irrelevant.

I don't think it's paranoia to wonder what he's really trying to do.

Human moderation seems to scale quite well, but as far as I can tell it is always in the context of a community. You need a scaffold of people willing to work within the boundaries of propriety, and willing to work with the moderation team.

I have no idea whether that means that news sites, which probably don't have a community in the same sense, are impossible to moderate using the same system or that a relatively much larger moderation staff is needed.

Regarding commercial sites, the basic rule has been repeated often enough: If you're not willing to invest in a community, you don't get a community. At least not your community; if you totally blow off your responsibility, you may well get an emergent community of trolls and other abusers.

Remus, you shouldn't be too impressed by numbers: Yes, modern forums can have a lot of comments, but the flip side of that is that your computers can help out a lot, -- not doing the job for the moderators, but providing tools that multiply the moderators' effort -- keyword, regex, and Baysian searches, sensible interfaces that allow mods to handle batches of messages at once, log filtering, and so on. (On the flip side, If banning each individual troll requires 24 clicks and getting a response from your supervisor, you might as well go home.) Yes, that means your mods need to be technically literate.

Crowdsourcing likewise acts as a multiplier -- you still need sentient supervision, but again, simple tools (like a "report" button) can make the mod's job much easier.

Tools still don't mean that one moderator can handle an unlimited load -- that goes back to "cheaping out on moderation". Also, the mods need to have backing from the owners -- if SOP is to fire a moderator anytime somebody complains, then your site will be SOL.

more later, maybe -- I'm getting called out by my dog.

no links, even

[Lack of an apostrophe in a common contraction got you. Many of the Madlibs-style comment spams lack quote marks and apostrophes where you would expect them. I believe it's because the word-substitution lists the spammers use are delimited with single or double quotes. -- JDM]

Shorter Denton: “Your people, sir, is nothing but a great beast.”* Obviously the solution to the problems of freedom is authoritarianism. (Original source: Alexander Hamilton.)

Here's hoping Teresa gets that book out soon.

As a corvid, I think I'm more of a small beast.

Both McClatchy and GEnie paid me. My staff at GEnie was all paid (even down to the lowliest minion who got a free account out of it).

As to tools:

A big red button to delete any post. Ability to see IP numbers, user's email address, and dated/time/ipn when they joined. Ability to edit any post. Ability to see all new posts. Ability to build my own filters.

eric @ 15: Your local newspaper sounds like mine. Of course, since I never read the comments there before or after, I can't tell what the switch to FB did to the comment threads.

In general:

I like the concept of crowdsourced moderation, but I suspect it only works when you've got an actual community, as opposed to a group of commenters on the same site. A site owner can't just say, "Oh, we'll crowdsource the moderation, and then we won't have to pay a moderator/can cut the number of paid mods in half/can do the whole thing with volunteers."

But once that community is in place, I think it gives the community a sense of ownership. "This is my part of keeping this place nice." Here, we have the spam-flagging that anyone can do. On a forum my husband visits, regulars will tell obvious fraudsters that they are "selling kayaks"*, to flag the thread for the moderators. The mods can then run a saved search for "kayak" at their leisure, and find all the flagged threads.

*It's an automotive forum. Their issue is people who sign up pretending to have cars/parts for sale, at prices obviously too good to be true. A regular adds a comment to the thread that the original poster appears to be "selling kayaks"**.

**Word may not be "kayak", but it's something that's not otherwise discussed on this forum.

I think Denton is acting like even more of a mean-spirited, arrogant jerkwad than people here are giving him credit for. His sites live and die by page views and clickthrough. Well-considered, reasoned, polite debate does not necessarily make for huge page-view numbers, and the people who engage in it are typically not the best of impulsive targets for advertisers. So he profits from articles and moderation (ahem) policies that generate free-for-alls or one-liner snark. And then he has the nerve to act as if it's a bug rather than a feature.

When my spouse managed moderation teams for some tv-network sites (years ago now) the situation was similar: management wanted to maximize pageviews consistent with avoiding legal liability. And to foster loyalty to the site and the network's shows. The cheap, low-road method of doing that was much easier.

If your goal is to have sensible, respectful discussion you can do it, but few and far between are the publications and sites that really do.

View All By is a wonderful tool for supporting the formation of a community of people who take some pride in their words.

One problem for commercial sites: the owner of the site does not own the community, and trying to pretend he does will end badly.

Community is clearly part of the answer but I don't think it's the full answer. Not all online discussion should require a long-term investment in community for the participants.

I have online places where I consider myself a regular and tend to follow the discussions wherever they go (ML and a handful of author LJs, mainly), and I have places like news sites or local community forums where I might be interested in occasional discussion of a limited topic if a reasonable one could be had.

The first are true communities; I am there often enough to know the other regulars and the in-jokes, know local custom, know what topics have recently been beaten to death - and participate in maintaining the desired tone. I value them highly, but they are time-consuming, and I don't want that level of depth in everything I do online. On the other hand, I think I'd become too insular if I never ventured beyond my home turf.

The media sites are, for the most part, unbearable at this point. Even setting aside the trolls, bullies, etc., the lack of commonly understood ground rules causes a painful amount of thrashing in the discussion even among well-intentioned people and leads to discussions being sidetracked onto Thingism 101.

In the middle ground, there are online places that are clearly communities, but where I am only an occasional visitor. There, a clear-cut commenting policy and responsive but not overzealous moderation by the regulars can help my visit be pleasant for all involved.

I am envisioning the online equivalent of a scene where an elegantly-dressed visitor from the East Coast walks into a Wild West bar and orders a glass of wine.

Also, when mods are visible parts of the community, the behavior they model seems to strongly influence the behavior of regulars.

That's been a failure mode on a few otherwise cool sites I've seen -- mods employed a lot of mean sarcasm in response to "problem" comments or commenters (not only trolls, but newbies who weren't aware of community norms, or regulars who screwed up). So mean sarcasm became the accepted mode of discourse, and things spiraled down. Commenters started trying to win social status by tearing each other down. A lot of people who did not enjoy that style of interaction left.

Compare Scalzi or Ta-Nehisi Coates or Our Esteemed Hosts. The usual response to problem comments (if they're not gross enough to delete, disemvowel, or ban the commenter outright) is "That is not okay. Here is why that's not okay. Do not do it again. If you do, it will be deleted and/or you will be banned." In response to a regular who's temporarily going off the rails, it's "You are doing X bad thing. Take a break from this conversation now." Sometimes that needs enforcing with temporary suspension; here, at least, the commenters told to step away usually seem to do it themselves.

Sarcasm is certainly not out of the question, but it's unlikely to be toxic.

Moderation is all very well, but what if the commenters, despite being reasonably civil, are idiots?

Here's my latest blog entry. Read the comments and weep.

(TL;DR: Charlie decides to buy a new car, specifically a 2005-06 Volvo V70 D5 Estate (station wagon). Charlie then discusses how buying a car has changed over the past two decades. Peanut gallery respond: "why don't you use Zipcar/use Hertz/buy a (Ford Flex|Lambourghini|Fiat|bicycle|bizjet|zeppelin) instead?" No, numbwits, I was not asking for advice on which car to buy: I already know what I'm looking for. If you'd read the OP with brain in gear you would have seen the reasons why options A-F above are inappropriate. Sigh ...)

They do this with the best will in the world: sort of a benevolent dogpile of spurious and unwanted lifestyle advice. I think I am going to crawl under the bed and hide now.

Charlie Stross @ 48... Besides, it's really difficult to find parking spots for zeppelins. :-)

I'll note one difference, in my mind, between a newspaper comments section and a Making Light comment section; the point of an ML comment section is discussion, but a large portion of what I want from a newspaper comment section is fact-checking and additional facts. Twenty-five comments pointing out that the contractor invloved in some scam is the mayor's brother-in-law may be boring, but it is useful information.

*patpats poor Charlie*

That's ok, my neighbor asked me for a jump yesterday. I had to apologize because we don't own a car. We use bikes for everything instead. I think a lot of folks just assume their favorite choice is the best one for everyone, and they get a bit flummoxed when other people pick different ones.

A Volvo wagon sounds like a great idea to me. Those suckers are really bulletproof. (well, ok apart from the whole driving thing...)

I read a blog, which has helpful interesting comment threads, and I feel burned by the fact that when I tried to contribute, I got an immediate "cannot accept this content" to a lot of the things I wrote. I think they must be using a fairly aggressive Bayesian filter - I did better if I kept comments very short, but there didn't seem to be a particular word that was getting me banned.

Perhaps it would benefit me as a reader to more tightly formalize which sites are "read-only to me" and which I may comment on - there's some level of "ow, it hurts when I do this" that I'm not learning well on my own.

When I worked at Yahoo, I read ValleyWag fairly often (corporate memos appeared there shortly before they showed up in my company inbox, and their rumors about layoffs were a good supplement to the grapevine). My impression was that the comments weren't that much worse than the articles, and when the articles' authors engaged in the comment threads, their tone fit right in with the other commenters. My impression was that ValleyWag was far from the worst Gawker site in that regard. Given all that, I find it amusing that Denton's so shocked by the low intelligence and lack of gentility of his commenters.

I'm pretty sure it's not original, but Ravelry's "report this post" function automatically opens up a private mods only discussion area for that report. This is really handy.

So for the group I mod, a lot of the flags end up being due to the moderation team. (for a volunteer team, we have really excellent time zone coverage) If a mod needs to tag themselves as off duty, we report one of our own posts. See something fishy that doesn't necessarily take action? Flag it. Participate in a hot discussion and need to step AFK? Flag it.

It doesn't take much time. There's 6 of us for a very newbie heavy group, and a half hour of reading per day is usually enough to keep up. And due to our comparatively heavy mod coverage, we tend to post more as regular posters... which heads off a lot of trouble right away. The new users figure out very quickly that the mod team is there and active.

There are other tools, and we do use them. But 95% of the time we depend on talking with each other and occasionally saying publicly that personal attacks are not ok, or that it's really not ok to conduct scientific arguments by fiat.

Charlie @48:

You have been hleped. They are hlepy. That was hlepiness.

That kind of behavior always makes the skin on my arms itch. But, circling back to moderation, it can be very difficult to curtail it. The commenters are trying so hard to be nice that telling them to STOP IT AAK! feels like kicking a nest of puppies.

One of the toughest areas of moderation is not dealing with the trolls, who mean ill, but with those who mean well but are still impossibly disruptive.

But you know that. This incident notwithstanding, you've got a good community over there.

Abi @55, thanks: it's useful to have a term of art for this phenomenon. "Hlepfulness." Yes, that cap fits!

Charlie @56:

Actually, "hlepiness", on the model of "truthiness".

It's a term of art in the Dysfunctional Families threads, where many people have extensive experience of being on the receiving end of it.

I know of one forum (http://www.rmweb.co.uk) where the moderation team are all anonymous. They post as regular posters using one account, but use separate moderator accounts when posting officially while wearing their moderator hats. The site owner goes to some length to protect those anonymous identities.

What's the Fluorosphere's take on this as a moderation technique? I'm not sure if it's a good idea or not.

Charlie, the hlepiness comes in avalanches around medical issues, too.

When I was using the walker and therefore couldn't outrun them, lots of people were hlepy in person too, but most of them were online. It's easier to deal with in person ones since I took a suggestion: have a notebook and pen, and ask the hleper to please write down their suggestion "because I'll never remember it otherwise." This at least saved me from having to lip-read endless in person infomercials for how nutritional supplements, hydroponically grown herbs, and wearing a scarlet fez while standing on my head would cure this, that, and the other thing.

Nowadays I don't post much about my health stuff in the clear at all, and hlepiness is a major reason why.

I'm sorry people were annoying and tedious about the car-buying post.

I've been a part of volunteer-moderated communities (skewing younger -- teens to twenties) where moderation became extremely cliquey; there was a lot of resentment of the Cool Kid moderators and their popularity, and a lot of "why do THEY get to be a moderator and not me?" and a lot of feeling (some of it fair) that decisions about moderating posts were driven more by the urge to take your friend's side in a fight than anything fair.

Indeed, one of the things I appreciate about Making Light is that moderators will tell even their friends to take a break from a conversation that's going badly.

In that kind of situation, then I can see where anonymous moderation would be a good choice, but I prefer the relatively transparent and open moderation here and at MeFi.

Tim @58:

My usual reaction to hearing about a given moderation scheme is, "an amazing variety of moderation schemes work for their own particular communities."

Having said that, I am leery of any scheme that relies on secrets staying secret. The damage that outing someone as a moderator could do constitutes a risk to the community, and the pursuit of the information could have a distorting effect on the conversation. People get obsessed enough even without secrets to act as foci of their attention.

Really, though, it depends enormously on the community.

On cheaping out: Why should it be surprising that proprietors who have gotten used to filling their sites with free copy should resist paying moderators? As the controlled-circulation tech-biz magazines that I used to write for were bought up by big media and converted into websites, I saw the kind of features I did for the print versions replaced by material supplied by PR flacks, "consultants," and other corporate-copy-mill employees. (And, of course, all the editors but one were canned.) "Free" content means reducing the number of people paid to generate, vet, and edit it to the absolute minimum. Between the crude economics of pageviews=revenues and the urge to maximize profit by minimizing costs (which mostly means labor), the scrawls-on-a-wall situation is almost inevitable.

There. I feel much better now.

As a volunteer mod elsewhere, I find that moderation in the clear can have an unfortunate effect on the moderators' reception by the community.

We usually moderate in the clear when acting directly with individuals. But we have a jointly-used moderator account that we use when moderating entire topics. That way, no individual can be "blamed" for shutting down or removing a discussion--we've been there and individual mods have been the recipients of a lot of flame as a result. We also use the joint account when communicating with individuals who are known to be prickly and short-tempered, for similar reasons.

When we were on yahoo, all moderation was in the clear, and many of us were burned as a result. Now we're on a platform that allows us to have this moderator account, and it's come in quite handy.

We were also often accused of using our power to steer a discussion or put forth a point of view that "represents" the "official" word on something. That was a matter of perception on the part of the members; the mods did and do not make official statements outside of statements relating to moderation/netiquette. Official statements of the organization's positions are made by the organization's founder; the mods often disagree on individual issues--and disagree publicly, on the Forums. When we post publicly, we post as ourselves, offering personal opinions and experiences. (It's easier to be clear about this in the new format where we have an official font color for official pronouncements.)

Several mods withdrew from personal participation almost entirely as a result of these accusations, which deprived them of the benefits of the community they were volunteering to moderate (we all get free membership). One mod quit.

Interestingly, once we moved to our new platform, most of this went away. Some of this is due to the fact that some of our most volatile members did not move with us, for a variety of reasons. Some of it is due to the fact that moderation is much easier/quicker, so that flashpoint posts can be removed from the forums before they grow into flamewars. I think the difference in format, which allows individual topics to be read individually, instead of all jumbled together, lets people approach subjects with more focus and attention.

The second interesting byproduct is that membership has grown dramatically. When we made the move, we had, as a guesstimate, about 900-1,000 members (not all of whom moved with us). 6 months later, we're approaching 1,400. While only 600 or so have posted, more than 300 have posted between 2 and 50 times, and overall we have far fewer lurkers than we used to.

The third interesting thing is that because there's been such a huge influx of new members, the prejudice against the mods seems to have disappeared, so we are all posting a little more than we were, and we feel more integrated into our own community again.

albatross @ 45: View All By is (at least in my experience) really useful for people who are new to a community. But I don't think I've come across anything similar elseweb, which is a pity: it would definitely improve my experience at some of the other sites I comment at.

When I started commenting here, I found VAB really useful for getting a sense of who I was talking to at any given time (and for me that was really essential for getting the sense that I was talking to real people, and not just - in TNH's words - a bunch of electrons.) From that point of view, it's really much better than real names.

I suspect that if more sites had something similar it would make a lot of people more appropriately skeptical saw the 'real names (or real identities) are required for civil discussion' fallacy. Is there any chance it could be hired out or sold to other blogs?

We have the equivalent of "View All By" over at Absolute Write.

You can click on a poster's name to get a drop-down menu that will take you to the poster's profile, all previous posts by that person (in reverse-chronological order) and a couple of other options.

I used to be a regular on a knitting mailing list, but I stopped reading it when it got into a drawn-out and bloody fight over whether it was appropriate to put "Merry Christmas" at the end of a message. These were nice, well-meaning people but that hit hot buttons all around, and then the moderators weighed in instead of using their Hammers of Correction. When I tried to get involved again, they were in the midst of a bloodfest over the ban on taking knitting needles onto planes, and again, the moderators were in the scrum.

I think that the success of successful moderation efforts I've been involved in is based on a having a team of moderators who communicate a lot with each other and who can keep themselves meta on the fights, around the clock mod coverage, and a strict ban on discussion of moderation within the forum aside from announcements from the mods themselves. The moderators also have to accept that a chunk of people will be annoyed with them pretty much all the time.

This is probably a lot like a successful towing company.

praisegod barebones @64

Ravelry also has an equivalent to view all by, with some variations: you can see all of a person's posts in a given thread, on a given board, or everywhere on the site. (This is handy for moderators who want to know if someone is breaking Ravelry rules about crossposting.) You can also search a given person's posts for particular terms.

I'm sure there are other sites elseweb that have similar features, though Ravelry's forum software is homegrown.

We have a "view all by" equivalent at our place as well.

One important point that I haven't seen addressed here yet is that the moderators sort of need to be relatively easygoing and thick-skinned at the same time, which isn't easy! For example, in the unlikely event that I were to be offered a mod slot here, I would not accept the invitation -- I know perfectly well that I'm too hot-tempered. (And I think the other mods know it too, which is why the possibility is unlikely.)

Yes, there are times when the Hammer Of Loving Correction needs to be applied -- but the ability to determine when it's appropriate requires an even temperament. Or at least that's how it looks to me.

Lee #70

Exactly - I've turned down an invitation for precisely that reason. I know I don't really have the temperament for it either.

Ravelry also has a board dedicated to pointing-and-laughing at flamewars and hissyfits (Ravelry Rubberneckers). It is against the board's rules to comment on both a flamey thread, and the Rubberneckers' thread on it.

I've seen people on other boards comment about the Rubberneckers coming over to look; I wonder if fearing being Rubbernecked keeps some people from becoming involved/ starting flamewars?

This site has a warped perspective. The commenters here are relatively few and very intelligent.

What, you think that's an *accident*?

BTW: for quite some time I couldn't figure out why someone's nostril width and/or lack of sinus congestion was supposed to have anything to do with their intelligence, but if you figure that the person breathing through their mouth is doing so *because they've been physically exerting themselves* and therefore need more oxygen, then it reduces to a straight-up class slur. People with nostrils so narrow they have to breathe through their mouths to get adequate oxygen even while sitting still are just the collateral damage.

And then there's faux civility, which is based on the idea that one can be as hostile as one wants, as long as one clings to a few pretenses---no 'swears', as some people call it, which seems like what a grade school kid would call certain expletives. That kind of thing----the 'bless-your heart-Minnesota-Nice-passive-aggressive seething but sweet hostility'----can't be caught by a bot, but pointing out such a person's striking resemblance to various unwashed orifices and ending the pretense is a violation of all of Emily Post's rules.

And the 'like/adds to the discussion' buttons are useless without degrees for the other ends of the scale: "hate/despise/smite'.

58/61
Over at Slacktiverse, the board administrative team (TBAT) will post using that name (and avatar) as mods, but if they're posting under their own names, they're not being mods. (It's not a secret who they are.)

Remus 34: Or can they? I wonder if it's been tried. BoingBoing is a for-profit (if low-comment) site

BoingBoing made a choice to be a low-comment site. They got rid of the Recent Comments feed on the front page, which makes it EXTREMELY difficult to find a conversation you'd been taking part in and rejoin it. They knew exactly what would happen if they did, and they made that choice.

I don't even read BB anymore. It's too frustrating. I read things and want to comment, but I know I'll never find the conversation again to see if anyone responded, asked me a question, told me I'm wrong, whatever. What's the point of shouting down a well?

I occasionally hang out at Tech Support Comedy, and I used to wonder how it stayed such a civil place with almost nonexistent moderation. But this discussion has helped me to realize what it does have:

• Login required to comment (although pseudonyms are encouraged over real names).


• New comments display at the bottom.


• Comments to a given post are not threaded.


• There is a view all by.


• Experienced members explain the customs to newbies.


• There is a sense of community. Members organize get-togethers from time to time. Sometimes we pass the hat for a fellow member in need.

I came within a hairsbreadth of flaming out as a moderator, once. I posted a very testy message, and then backed away, had another admin remove my mod powers, scrambled my password, and never came back to the forum again. Because I could feel myself about to go into a public frothing snit, and had just enough self-awareness to not do so. That time. Thank god.

But in that case, it was a matter of moderation rules that...well, didn't suit my idea of what a good forum should be. Making sideways jabs at large groups was always fine, calling someone on doing so was always Rude and Improper, because, see, that was personal. I'm glad to be well away from the place.

Emily @ 60

It can go the other way too. I modded a pretty dysfunctional group for several years (with plenty of in-person crossover and hostility, for added fun!), and quit when I noticed I'd slipped down into the muck with everyone else.

Chris @ 72

Nothing so one-sided. People with merely decorative noses also get nailed, but for being "overbred" elites.

Lee @ 69: I sort of talked towards that in my comment at 47 -- hot-tempered mods being the ones who modeled mean sarcasm as a high-status behavior.

Damping the shouty feedback loop by being highly non-shouty -- extremely reasonable and/or matter-of-fact -- helps.

This ties into DFD in a big way. For me, anyway. Successful moderating seems to have a lot in common with successful parenting -- at least from the perspective of someone who's only been a child and a commenter, and not (yet?) a parent or a mod. I don't mean that moderator/commenter is exactly the same relationship as parent/child (obviously it isn't); just that some of the same skills seem to apply. Guidance, understanding, de-escalation, redirection, consistency, firmness, compassion, educating in the moment.

I'm interested in whether people who have been both mods and parents think this simile makes sense, or whether it's different from that perspective.

Caroline @79:
Successful moderating seems to have a lot in common with successful parenting -- at least from the perspective of someone who's only been a child and a commenter, and not (yet?) a parent or a mod.

I find the parent/mod overlap is most noticeable in two contexts:

1. The use of the Mom Voice: the ability to stop an argument dead in its tracks with very few words. Mind you, plenty of non-parents have the same power (Teresa, I'm lookin' at you). But one does get the chance to hone it as a parent.

2. A certain way of reading quarrels. Kids, particularly young ones, have their motivations very much on the surface. When a two year old pokes her brother on the arm over and over again, you can see in her face what she's up to. Adults are prey to the same impulses, but they hide them better. Having seen them in the clear makes it easier to pattern-match.

Where parenting and moderation part ways, however, is that I am responsible for the long-term emotional development of the children in my care. I make decisions that are intended to help them grow up to be good, thoughtful, responsible and courageous people. I have no such long-term duty of care to any commentariat; my strategies, even those for avoiding emotional damage to people in the area of my authority, are all short-term.

It would be insanely intrusive and entirely inappropriate otherwise.

Caroline #79: Adding to Abi: Also, you're trying to shape the behavior of folks who at any given moment may not be "of sound mind", while also minimizing use of outright force (to which you have only limited access anyway).

abi @80 I have no such long-term duty of care to any commentariat; my strategies, even those for avoiding emotional damage to people in the area of my authority, are all short-term.

It appears to me that implicit in the way you approach the task is the view that your long-term duty of care is to the nurturing of the site rather than the individuals. Not that you are indifferent to the individuals, just that, as you say, the long-term focus is not on them in the same way it is with children. Mod as gardener, perhaps.

Fade Manley @77:

But in that case, it was a matter of moderation rules that...well, didn't suit my idea of what a good forum should be. Making sideways jabs at large groups was always fine, calling someone on doing so was always Rude and Improper, because, see, that was personal.

Oh, that drives me absolutely bonkers as well. I'm most familiar with this stance from LibraryThing, where it's perhaps the thing I like least about the site. At any point, anywhere on the site, someone can (for instance) say that all LGBT people should have their children taken away, because we're all child molesters; that, from the viewpoint of the site owner, is just a political stance. Respond directly by calling the poster a bigot, though, and you're liable to run afoul of the "no personal insults" rule. There is no moderation per se on the site -- members can flag messages as violating the no-insults rule (or counterflag them) and after too many flags a message will require a click to be seen. The climate is still very unwelcoming to people who don't have the luxury of viewing those attacks in the abstract. (I got flagged for personal insults once for pointing this out. I said that someone had straight white male privilege that allowed him to see these attacks as abstract political issues, which was viewed as an insult.)

Ooh, nice, Abi. The big difference, then, is that for some moderation at least, you can choose whether you're in the iterated or single-shot Prisoner's Dilemma scenario. If the longterm health of a community involves some commenters not being part of it, so be it, whereas you mostly can't say the same thing about the longterm health of an immediate family.

Of course, some commenters are more or less willing to take hints -- when I was moderating a usenet group, there were some people who seemed absolutely dedicated to the group in question, just with a seriously different view of what it should look like than the moderators and the charter.

abi, thank you so much. That makes tons of sense. The Mom Voice and reading quarrels seem to cover most of what I was thinking about there -- the enforcement and conflict-handling skills.

The point about where your long-term responsibility lies is a crucial difference -- one I obviously hadn't thought of from my perspective.

Clearly, most of my experience with conflict-handling has been in a family setting, so I reached for that analogy. But it's not the right analogy, because you pointed out that conflict-handling and enforcement in a parenting situation are just part of a much bigger long-term project of guiding the emotional development of your children (the people whose conflicts you're handling).

In a moderation situation, conflict-handling and enforcement are part of a different project. Guiding the development of the culture of the board/list/blog comments section, maybe?

Caroline @ #85 said:

In a moderation situation, conflict-handling and enforcement are part of a different project. Guiding the development of the culture of the board/list/blog comments section, maybe?

One of the enormous challenges we faced at AW in creating a politics room was just this.

We decided, behind-the-scenes, that we wanted a room culture that rewarded posters who supported their positions with careful citations, logical coherence, and rhetorical grace; likewise, posters would be discouraged from flame-baiting, posting drive-by one-liners, and merely regurgitating talking-points.

While it's certainly not a room that's everyone's cup of tea, and it's been a long slog to get the regulars to the point where they're helping self-police, propagate, and nurture those values with each other and with newcomers, and while admittedly imperfect -- it's mostly worked.

The room skews a great deal more politically moderate than I personally tend to be, and I get the occasional note from a newbie accusing me variously of either right-wing or left-wing favoritism. But factions from all over the spectrum are represented in the room, and generally things stay pretty civil.

We generally manage multi-page conversations around some very hot topics that are neither full of echo-chamber ditto-heads, nor partisan-political conversational grenade-lobbing.

One of the unforeseen benefits was when that culture of "cite your sources for that assertion" started making its way into other rooms, as well.

At any point, anywhere on the site, someone can (for instance) say that all LGBT people should have their children taken away, because we're all child molesters; that, from the viewpoint of the site owner, is just a political stance. Respond directly by calling the poster a bigot, though, and you're liable to run afoul of the "no personal insults" rule.

Oh, that crap. I'm not on LibraryThing, but I recognize this from a board I thought about joining but then did not because I could see that going on. That is: One poster had been put on suspension -- she couldn't post, and others had to click to read her posts -- because she had said to another poster, "You sound like an abusive boyfriend." Personal attack, see? Whereas that post of his he responded to, in which he gave other guys on the forum the advice "Don't ever let your girlfriend have male friends"? That was A-OK.

*shudder*

Let me just say "fuck LibraryThing, the horse they rode in on, and any other site that behaves that way."

Oh, and anyone who supports that behavior? Fuck them too.

Xopher @ 88:

Frankly, I think the world would be better off if everyone refused to have sex with people who say things like that. Even aside from the improvement in the gene pool, I mean.

A fair point. Spay and Neuter them, then.

I would be surprised if anyone associated with Gawker believed moderation could work - look at thier site. The news they print is gossip, contentious and incendiary at the outset. Intelligent comments are not as valuable to them as sheer click volume, I suspect. Nothing wrong with that, but asking a Gawker principal about moderation is like asking Forbes about financial regulation - they have strong incentives to believe it both unworkable and immoral.

Scott:

For good and ill, some topics of discussion, and some types of commenters, will just behave better than others. Being able to include and exclude topics and people makes for a better environment for everyone else.

This maps back to realspace communities, as well. Some communities that get together tend toward more bad behavior and annoying of the neighbors than others--the obvious example being any activity that's mostly done by young men and involves alcohol.

Often, the practical answer to this is to exclude the kind of people who cause problems, usually in some backhanded way like refusing liquor licenses to new establishments in some neighborhood, or by having a store pipe in classical music to irritate the after-school teenaged crowd who are mostly perfectly nice kids, but who in a large group will cause more trouble than most other large groups. Or even the high-pitched irritating noises that mostly aren't audible by older people, but which make younger people mostly decide to leave.

To have a community is to be able to exclude some people who aren't part of the community. In one sense, this really sucks, since it sucks to be excluded and many of the people excluded might have been able to eventually become part of the community and add a lot. And yet, I don't think it's possible to have a community without some level of ability to exclude people. Deciding when it's time to move people along, as abi commented before, is very important for being able to keep a community going.

albatross @92:
For good and ill, some topics of discussion, and some types of commenters, will just behave better than others.

I'm increasingly aware that there are certain tones in which one can write posts that drags the subsequent commentary down. I've not really discussed this with anyone, but I'm beginning to be able to tell on reading the OP whether the thread that follows it is more likely to go south.

When I have this properly quantified and understood, I may write about it more. But it's still mostly instinct right now.

To have a community is to be able to exclude some people who aren't part of the community.

If it comes down to a choice between a newcomer, however high-potential, and the existing community, I'll choose the existing community. (Though "choosing the existing community" also includes "making the existing community a welcoming place for newcomers"; it's good for us.)

abi:

Do you think the effect of tone is separate from the effect of the topic and the information content[1]? My intuition now that you've mentioned it is that tone probably drives a lot, both the tone of the original post and the tone of many of the comments in the thread[2].

I wonder how this interacts with the makeup of the community? My sense is that an OP that deals with some split in the community can have a big impact on how the comment thread discussion goes, but I haven't really thought this through much.

[1] Like, if you wrote two different original posts, on the same topic and with the same information, but with a different tone, and then flipped a coin to decide which one to post, the thread would somewhat predictably go in one direction or another based on the outcome of the coin flip? I understand the IRB (aka, the mods) would object to this experiment applied to the ML environment, but it's a useful thought experiment[3].

[2] I also have the sense that comment threads and other discussions have local gravity wells of badness, into which they can fall, and from which they can't really escape. The tendency for threads to die a natural death seems like a natural defense against this, analogous to the way limited length of telomeres is a natural defense against cancer.

[3] The IRB for thought experiments is considerably more permissive.

albatross @94:
Do you think the effect of tone is separate from the effect of the topic and the information content...?

Yes, I do. A mocking tone, for instance, is more likely to elicit a mocking discussion thread. Which is not to say that mocking, or angry, discussion threads always go wrong. But there are certain subsets of mocking and angry that are more likely to slide into the dark side.

Like I said, I haven't analyzed this well enough to present a clear explanation (yet). At the moment, it's a pattern I can recognize but not verbalize.

Much of moderation (as I practice it, anyway) is instinctive. Teresa talks about "natural moderators", and points out that most moderation information comes out in anecdotal contexts. This is because many of the people who do it don't understand the how's and why's until they've hit the real-world examples. The rules and guidelines we talk about are the result of pattern-matching and analysis after the fact.

And I'm not quite there yet with this one.

abi @95 At the moment, it's a pattern I can recognize but not verbalize.

To me, one of the most interesting things that's come out of psychology research in the past 25+ years is an increasing recognition of and respect for the extent to which the human mind does this. Huge amounts of information processing go on at the nonconscious level with only an occasional ripple on the surface. I love to see someone who's tuned into something - the weather along a stretch of coast, the sound of a particular engine, the tone of a discussion - use that expertise. Rules and teachers and explanations can shorten the time for acquiring that level of knowledge (or sometimes confuse things completely) but there really is no substitute for experience.

OtterB @96: but there really is no substitute for experience.

But for some few of us, exposure to that "tuning" in action can result in acquisition of the "tuning". (Think Sylar and superpowers, only benign and without the head-razoring.)

This is a skill I seem to have. The "tune" manifests in the whole person. I can sometimes deduce pieces of it from verbal cues and pattern-deduction. But in order for it to work well and efficiently, I almost have to be in the physical presence of the person exercising the target skill. Ironically, my meta-skill is entirely non-conscious as well. I can turn it loose, and sense it after it's happened, but I have no conscious or deliberate involvement in the process.

This is one of the things that tickled me about NLP. Their "submodalities" model makes for a good framework for studying these kinds of skills. It's still spooky magic, though.

albatross @94, Abi #95: If memory serves, we recently had a "natural experiment" along those lines, with two "politics" threads in December and January. There were, as always, complicating factors, but there was certainly a major difference in the tones of the original posts -- and it was the one started with an angry sarcastic tone that flamed out.

Fade Manley @77: I think that's an easy trap to fall into. The two essential ingredients that I can recognize are these: (1) A general and naive pro-freedom-of-speech slant that the moderators buy into, and (2) "Personal attacks are a bad thing".

Now, both of these are understandable impulses. There's appeal in the thought of being an open forum for ideas and debate, and the desire to stop personal attacks is a good one.

What it primarily lacks is the realization that personal attacks are only part of the problem; the real problem is more to do with the lack of treating fellow forum members with reasonable respect and manners. There are plenty of ways to disrespect others without being explicitly personal, and if you ban direct personal attacks but not disrespect that doesn't name names, then all you're rewarding is passive-aggression and baiting.

In addition, if you make it so that nobody can criticize anyone else's behavior, then you'll end up with situations where tempers and hatreds build up in secret until things explode and you have to ban people you'd rather not have had to ban.

I'm one of the moderation team at TV Tropes, and we've been slowly dealing with getting past these moderation mistakes and instituting better policies. The typical growing pains of a place that's outgrown "everyone knows each other and everything that's going on".

Matthew:

It seems to me that another aspect of this is that in each community, there are some targets that are seen as fair game for at least rude and dismissive comments in passing. What gives offense in that context is a disagreement about whether your group is fair game. Go to Respectful Insolence and you will see practitioners of homeopathy and integrative medicine and creationists being slammed both in detailed fact-oriented ways, and also in snide remarks made in passing. Or look here w.r.t. "dominionists." Or go wander to the right end of the blogosphere, and you'll see the same thing w.r.t. liberals or gays or Muslims.

A lot of speaking up when some minority is being bashed is about establishing some kind of bound, saying "I don't agree that this community should be a place where it's open season on gays/Christians/SF fans/etc.

Matthew Brown @99: What it primarily lacks is the realization that personal attacks are only part of the problem; the real problem is more to do with the lack of treating fellow forum members with reasonable respect and manners.

Also, a confusion between "personal attack" and "challenging/questioning/disagreeing with people's beliefs/behaviors."

Jacque:

I think it is honestly hard to draw a line between those, at least some of the time. Some real topics of discussion will inevitably hurt people who have as much right as anyone else not to get their toes stepped on, and it's a matter of community standards (sometimes shading into laws, risk of lawsuit, or risk of some kind of nasty personal retaliation like getting beaten up or fired or added to a watchlist somewhere) which of those discussions may or may not happen in a given place and time.

For example, it's a perfectly legitimate question to ask, in some communities, "What is wrong with all these people who believe in some magic invisible friend in the sky that they pray to for aid in their daily lives?" In others, it's perfectly legitimate to ask "What is wrong with all these liberals that has made them hate Christianity and America so much?"

Indeed, in some communities capable of having interesting and valuable discussions, that kind of question has become a part of the shared community. In a community where bashing on X is part of the community standards, someone complaining about all the bashing on X is being oversensitive, not having a sense of humor, demanding censorship, whining, etc.

Now, the specific assumptions underlying those questions may or may not be right. I'd hate to live in a world where either one couldn't be asked or discussed anywhere. And yet, I also don't really feel like spending a lot of time in communities where my basic beliefs and who I am are attacked constantly--that unwelcome feeling I'm getting there stems from the fact that I am, in fact, unwelcome.

I think this is a near-universal property of communities. There will always be more-or-less widely accepted premises in a community--stuff you don't feel like you have to defend or justify, stuff you think you can dismiss out of hand without argument. Some of that stuff will be the subject of perfectly legitimate disagreement, but not in that community.

albatross @102: Expanding on your thought: communities exist as much because they exclude people as because they include them. The problem is similar to the problem with 'pataphysics, which requires the treating of every event as if it were unique. No community can include everyone in the world comfortably: the world is too big for almost anyone to feel a kinship with everyone in it.

To quote the first verse of Sussex by Kipling:

God gave all men all earth to love,
But since our hearts are small,
Ordained for each one spot should prove
Belovèd over all;
That, as He watched Creation’s birth,
So we, in godlike mood,
May of our love create our earth
And see that it is good.

It's as true of groups of people as it is of places. That is, pretty much, and there are a few exceptions.

Matthew, #99: I'm glad you showed up, because I have a question for you. Is there any relatively straightforward way to look up "that trope where..." if you don't already know what it's called? I love the creative names for the tropes, but the last 3 times I've been over there trying to figure out what a specific trope is called so that I could link to it, I've given up in frustration. Generally after an hour or two of wikisafari that didn't get me any closer to what I was looking for. :-)

Lee #104: I'm not him, but IIRC there's a forum actually named "That Trope Where...". That is, indexing by the kindness of strangers....

We know it's a problem. It's kind of an inevitable one, given that no matter how obviously we title something, not everybody will think of it using the same words.

One approach is trying to get rid of the worst offenders in the "unobvious title" category: things named after characters in works, for instance. Every fan thinks everyone knows their show or book or comic, but for those who don't, "The Wesley" (for example) is a completely impervious lump; they'd never even think to look there.

Another is to give things aliases, to try and cover various ways to put it. Sometimes a funny trope name is too good to lose (humor helps you remember it after you know it, for instance) but needs help; other things, there's just so many ways to say it.

Also, all tropes should be indexed. Starting points for exploring the indexes are on the left sidebar.

There's also a title search and google site search available.

If you're really stuck, there's "Lost and Found" (http://tvtropes.org/pmwiki/query.php?type=lnf) where you can ask and people will try and get you an answer.

albatross@92

I've wondered about those high-pitched noises to drive away teenagers-- might the noises also torment small children and babies?

albatross @ 100
there are some targets that are seen as fair game for at least rude and dismissive comments in passing.

This.

It's a big part of why I'm spending less and less time on a lot of blogs I used to read, including ML.

i'm biased, of course, but i'm a big fan of giving readers the tools to permanently hide trolls.

it's quite a nice feeling to be able to say "yes. that person is a troll" or "i don't want to read anything else by that abusive ass", click a couple buttons and... voila - hidden for good.

it doesn't stop other people from reading or responding to the troll, but it keeps me from ending up on the aggravating side of "someone's wrong on the internet!"

@109 Those are great unless the troll is getting a lot of action, in which case (for me) it's equivalent to being on the bus with somebody having an argument with their spousal equivalent over a cell phone.

I really like the disemvoweling approach - if I'm really determined I can make out what the troll is getting on about, but visually it's like listening to an argument in a language I don't understand: just clutter that I can skim over.

Albatross @ 100:
there are some targets that are seen as fair game for at least rude and dismissive comments in passing

SamChevre@ 108:
It's a big part of why I'm spending less and less time on a lot of blogs I used to read, including ML.

You mean like Xopher @ 88 and 90, and Bruce Cohen@ 89?

I agree with the sentiment expressed - the LibraryThing behaviour totally failed to recognize some genuinely sickening and damaging remarks whilst punishing a far lesser transgression in a way that IMHO calls for at least that much snark, and that's before we get to the actual misogynist - but still that particular run of comments popped into my mind instantly on reading Albatross's comment, and not just for proximity. And more so for SamChevre's.

Thing is, there have certainly been productive discussion with similar sentiments in the OP, or even the OP *title* much less the comment section ("Fscking Ralph Nader" jumps to mind; ditto something at least as rude re: Barack Obama from before he was even a presidential nominee.) So clearly these kinds of passing attacks are not in themselves cause of detriment to open discussion.

Can someone who has a little more brain than me explain why? Is it because, as albatross says, these targets are accepted by the full community as fair game? Because there were facts and opinions and direct personal damage to back up the rude opinions? Or, as I most sincerely hope, because we recognize that anger is sometimes a perfectly appropriate emotion, even when we disagree as to its target or its expression, and thus we let our friends say angry things when they need? A bit of all the above?

Genuine questioning, genuine confusion where and what the line is.

Lenora Rose @111:

The role of anger is one of the less-settled matters in this community culture. That's mostly because different people here (both in the threads and on the front page) have different relationships with the emotion, and different uses for it in their lives.

As one of the more anger-averse people here, I try to divert and deflect those expressions of it that I think will damage the conversation and the people in it. But cutting off others' ability to express their anger is also damaging, so I can't (and wouldn't) just cut it all off.

First and foremost, I am concerned about expressions of anger directed at people actually in the conversation. My second concern is for anger inappropriately* expressed against people or things dear to members of the conversation.

It's difficult when people are attached to institutions or establishments that have directly harmed others. It's also difficult if I don't know about the affiliations of community members. And it's not very easy, either, if people defend what is dear to them in inappropriate ways. That last one is difficult, of course, because there's nothing like feeling alone in a hostile crowd to make a person argue badly.

It's possible that some combinations of sensitivity and anger are irreconcilable. It's possible that some affections and affiliations cannot be accommodated in the community culture, and that some forms of anger can't fit in here either. Where that's the case, I try to move people on as undamagingly as possible.

I wish it were otherwise. I wish it were possible to have everyone who wants to be here in the same conversation. But I know better than that.

(SamChevre, albatross, do we need to talk?)

-----
* This is the keyword. For instance, four of the front-page posters are members of the Roman Catholic Church. There's plenty to be angry about with regard to that; I can't say as how I'm delighted with things are they are myself. But expressing that anger inappropriately—for instance, by making derogatory comments about ordinary members rather than the actions of the hierarchy—is not acceptable.

Nancy@107 - those high-pitched noises to drive away teenagers
Yes, they also bother small children, babies, and potentially dogs*, unlike playing Mozart or Sinatra which doesn't bother the young kids but gets teenagers really annoyed at the noise their parents call music these days. The real question is whether it's at a volume that's actively uncomfortable or just nagging and annoying, but you'll have to ask a kid about that.

A while back I was working on some electronics, and a twenty-something person came by and told me it was making loud high-pitched squeaks. I was surprised, because while I have trouble picking up speech clearly when there's background noise, I can usually hear high-pitched noises that my wife can't, but I guess it was out of my range.

(*"dog-whistling" isn't just for right-wing politicians and their comment-trolling minions, after all...)

Bill Stewart @113: It all depends on the teenager, of course, but in my daughter's social group, Frank Sinatra and Tony Bennett are considered musicians, not noisemakers. They like some classical, though most of them haven't been exposed to much of it and generally they prefer jazz and swing and things from what is widely considered "The American Songbook."

That's not to say that they don't listen to Lady Gaga, Amy Winehouse, Rhianna, Ne-Yo, Usher, and other more contemporary artists, but playing Great Singers of the Past is not going to drive these kids out of anyone's store. They're more likely to sing along . . . .

:-)

Lee @104:

Apologies if you're already using this technique which is why you ask for something straightforward, but can you come at it sideways? Think of a classic and obvious example of the trope, and look at the page for that source? (Look up James Bond if you want to find out the name for the one where the villain monologues at the captive hero to explain his fiendish plan, for instance.)

abi @ 112:

If that sort of comment really does bother people, I'm quite willing to stop. My intention in posting that sort of remark is more than just to express anger. In this particular example, I wanted to make it clear to Xopher that I agreed with his disgust at the behavior of the mods at LibraryThing, and to add some humor to the discussion, so as to vent a little for both of us. My exchanges with Xopher in the past have shown that we tend to get angry about the way some people treat others, and that we'll cool down after awhile, but it helps for us both to vent a little steam in the process. I understand that this might not be clear or acceptable to other people.

Bruce @116:

I thought that exchange went fine, including your contribution to it. It's within the parameters of acceptable in this community, unless someone wants to clearly and directly (ie, not just using it as a recent example in a wider discussion) argue otherwise.

An argument otherwise should, I hope, differentiate between the expression of anger (which is a necessary part of the community life, as discussed) and the terms in which it is expressed (for instance, my request for people not to use "die in a fire", which seems not to have damaged the conversation in the long run).

It's a fine balance, always. And part of my role is to help make the community culture so coherent and clear that most of the norms are normal, unthinking behavior.

Melissa Singer @ 114... The San Francisco Chronicle's Phil Frank had an episode of his "Farley" comic-strip where a gas station used classical music to chase away the darn kids who kept hanging around, only to then be infested by old foggies.

Serge @118: It being San Francisco, old foggies should have been expected . . . .

Melissa Singer @ 119... And old geezers in Yellowstone Park? :-)

Too little time and too much stress today to answer in depth, but I just want to say, for clarity, that comments #88-90 are not what bother me.

abi 117: I thought that exchange went fine, including your contribution to it. It's within the parameters of acceptable in this community

I'm really glad to hear you say this. I was pretty sure you would have said something at the time if you felt otherwise, but having it be explicitly mentioned is reassuring.

OTOH I'm now aware that such things make some people uncomfortable. That may temper my venting in the future. In fact it has in the past...I rarely vent HERE now. My Facebook friends see more of it.

114 Melissa Singer: it is to be said that most people that are playing Mozart, Sinatra, et al as a teenager irritant are not going to have issues with most teenager(s/ groups) who are antigen-resistant. At least not when I was said teenager nor in the situations I've been involved in as a no longer-teenager.

Mycroft W: I see what you're driving at, but in other respects these are "typical" teens, fully capable of breaking stuff and shoplifting and being loud.

They just prefer to have dance battles instead of fist fights.

Xopher, Bruce Cohen: FWIW, I wasn't bothered by the exchange, and actually hesitated to use it as an example because I expected it would make you self-conscious of your actions in the wrong way.

Abi @ 112, your list of priorities, even as vague as it was, is exactly the sort of thinking I was curious about. It makes perfect sense, and not just from the moderator view.

I'm not very anger averse. Still, I chose at one point to excise expressions of anger in the form of threats of violence from my vocabulary, for a number of personal reasons, and since then, I seem to have become more alert to them in others' rhetoric. They only occasionally bother me, varying usually on how much I think the person would willingly commit the violence and sometimes on misogynistic triggers. To beat a dead example, Xopher's words didn't bother me because he's been over the "rhetorical flourish" vs. "want to do" vs. "will ensure he is never in a position to do" territory many times.

Firstly, I do not think that scaling is as straightforward as you seem to think. The dictatorial system requires a person with sound judgement, and mod teams that try to establish a collective voice spend a lot of energy in the process.

But some sites do do it well. I'm fascinated by Money Saving Expert, which has (and requires) a huge forum. Human moderated; most of the mods are volunteers who are active in a particular sub forum. Rules that are clear, often quite onerous (to prevent legal challenge), and applied strictly. Many, many active members, from a wide variety of backgrounds. And different texture in the different sub forums; some are quite aggressive but most are strongly supportive.

Secondly, Ravelry is full of forum innovations. It's partly a good site because it filled a need; but it's underpinned by a genius coder.

Forum moderation, like education, appears to be a realm in which it NEVER OCCURS TO ANYONE to say, "let's see if we can find a place where this is being done well, and find out how they're doing it."

In education, at least, the approach appears to be either (a) "Let's try something random for no particular reason" (supported by research that shows that pretty much ANY change in educational routine will cause a temporary uptick in student performance) or (b) "Let's find somewhere this is being done well and then explain why it won't work here so no need to bother trying" (Iceland).

Ida @128 links to a pretty cool site, but has questionable google-fu, and no history here. This gets close to the XKCD model of comment bots, if it is one. If not, please say more Ida?

in my previous comment....

"Great read! Thanks!" is a totally-typical comment-spam comment.

Lila @ 127:

'Research that shows that pretty much ANY change in educational routine will cause a temporary uptick in student performance.'

I find it very plausible that this is true, and that there's research to support it, but would appreciate being pointed in the direction of some.

praisegod barebones@131: I don't know about education specifically, but every management study from Taylor on down has confirmed that changing routines at work causes a brief upturn in productivity. Since learning is a kind of work, I'm not surprised it works in education too.

praisegod barebones @131, to follow on what mjfgates said, the term in the industrial psychology literature is "Hawthorne effect." Although it appears possible that the effect wasn't actually there in the original study if the data are reanalyzed.