Fascinating article and one that echoes what I heard more than a decade ago from a woman who conducted public consultations about the dangers of large, public meetings: the strongly opinionated will take over, the moderates will get discouraged and go home, any hope of reasonable discussion will end.
Perhaps we can conduct a live test here: Thoughts on this article, everyone?
By Bora Zivkovic | January 28, 2013
Proposed alternative title: “This post is not about climate change”
Yes, we’ve all been there….
A couple of weeks ago, an article was published in Science about online science communication (nothing new there, really, that we have not known for a decade, but academia is slow to catch up). But what was interesting in it, and what everyone else jumped on, was a brief mention of a conference presentation that will be published soon in a journal. It is about the effect of the tone of comments on the response of other readers to the article on which the comments appear.
I have contacted the authors and have received and read a draft of that paper. Since it is not published yet, I will not break all sorts of embargoes by going into details, but can re-state what is already out there. An article about nanotechnology, a topic most people know very little about and usually have no a priori biases for or against, was presented to the test subjects. Half the people saw the article with (invented) polite, civil and constructive comments. The other half was given the same article but with uncivil comments – essentially a flame-war in the fake commenting thread. The result is that readers of the second version quickly developed affinity for one side of the argument and strongly took that side, which affected the way they understood and trusted the original article (text of which was unaltered). The nasty comment thread polarized the opinion of readers, leading them to misunderstand the original article.
The assumption is that on hot topics, like climate change, readers already come to the article with pre-concieved notions, and thus the civility of the comments would have no effect on them – they are already polarized. Choosing nanotechnology as a topic was a way to see how comments affect “virgin minds”, i.e., how the tone of comments starts the process of polarization in new readers.
They specifically chose a topic about which most people know very little and do not already have any opinion. Neither the article nor the comments contain sufficient information to turn the readers into experts on the subject. So they have to use mental heuristics – shortcuts – to decide what to think about this new subject. Uncivil, aggressive comments resulted in quick polarization. Readers, although still not well informed about the topic, quickly adopted strong opinions about it.
As many of you may already know, there is this thing called a 1-9-90 rule of online participation. In any given online community, about 1% of the participants produce most of the content, another 9% participate regularly by editing (e.g., on a wiki), commenting (on blogs and articles), occasionally producing new content (in forums, etc), and the remaining 90% are ‘lurkers’ who do not publicly participate but only read (though these days, many of them participate a little more publicly, if not creatively, by “Liking”, tweeting, and otherwise sharing the content in ways that are visible to others, but without adding any thoughts of their own). The exact proportions vary from site to site, but are usually close enough to 1-9-90 for the general rule to hold.
For sites like this one – a media organization and a blog network – the 1% are pre-ordained: our editors, staff, freelancers, network bloggers and guest bloggers. In other word, they are selected, not self-selected, and many of them can do it only once or very rarely. The 9% are active commenters, and the 90% read and perhaps share, but never say anything on the site itself.
Where are the comments?
Many people have noticed that the quantity of commenting, especially on blogs, has sharply decreased over the last couple of years. One reason is that discussion of the article or a post is now happening elsewhere – on social media (Twitter, Facebook, Google Plus…) or online communities (Reddit, Digg, Fark, Slashdot…), and are not physically attached to the original post. The trackback functionality is disabled on many sites due to enormous amount of spam.
Some new commenting systems are trying to re-attach such detached discussions back to the original post, but that is still not completely technically feasible – one can certainly not bring in a conversation happening on someone’s private Facebook wall. Some of those 9% of readers, instead of commenting on the post (at least a brief “Nice post, thank you”) are now sharing the link elsewhere and perhaps discussing it elsewhere, without the author of the original article ever being able to see that discussion.
Instead of “silent” participation leading gradually to more active participation as one becomes more comfortable with the site, it seems the opposite is happening: mildly active users are now becoming silent users as it is easier to click “Share on Facebook” than to post a brief comment.
But there is another problem here – most of the good, nice, constructive commenters may have gone silent and taken their discussions of your blog elsewhere, but the remaining few commenters are essentially trolls.
The question every blogger in this situation has to ask is – what to do next?
One option is to give up on comments entirely, and perhaps completely shut down the commenting functionality, trying, at the same time, to find and track discussions wherever they may be happening. A veteran blogger, Dan Conover just did that – go read his explanation why.
Commenting is not an essential element of blogging. Some of the most popular blogs never had comments, e.g., Andrew Sullivan’s The Dish, Josh Marshall’s Talking Points Memo (though his site has plenty of other spaces for the community to be active), or John Hawks’ blog. They always got plenty of reader feedback via email, and now also get it via social media where all of them are quite active.
Another option is to do some serious and time-consuming work on building the commenting community and motivating readers to post comments. It is much harder now than is used to be. And the presence of those few remaining comments – most of them vile and nasty – does not motivate serious people to add constructive comments into a cesspool of primitive discourse. Which brings us to the topic of…
What does it mean to moderate comments? Different people have different ideas about it, but many focus on technical fixes.
Spam filters – most spam filters already come pre-programmed to eliminate specific types of spam, e.g, those that contain words like Rollex, Vuiton, Viagra, Texas hold-em (unfortunately, sometimes just “Texas”) as well as various XXX words. Some spam filters allow the blogger to manually add or remove terms that trigger the spam filter. Other blogging software allows one to “teach” the spam filters over time, by sending comments to Spam, or by rescuing valid comments out of the spam folder.
Pre-comment moderation – in which the blogger sets up the software to send email notification each time there is an attempt at commenting. The blogger then goes to the dashboard, reads the comment and makes a decision: to set the comment live, to trash it, or to ‘teach’ the spam filter by sending the comment to Spam.
Post-comment moderation – in which all comments initially go live, but the blogger is generally online a lot, gets email notifications about each posted comment, and can quickly react and, if necessary, delete or spam the offensive comment. In some cases, the first time someone comments, their comment is held in moderation, but if the comment is approved, subsequent comments go through automatically.
Sophisticated graded moderation – some sites have been experimenting with more complex commenting systems which allow for, e.g., upgrading and downgrading comments (so upgraded comments get moved up to the top, while downgraded comments may become invisible); or letting valued commenters earn badges and special privileges over time, so their comments may show up on top, perhaps in different color, or with different types of signs or avatars; or allowing users to report inappropriate comments to the moderators. Other sites are starting to experiment with annotations in place of comments. Also see Dave Winer who separates main blog from comments – he requires an additional click or two to comment. I find this an interesting strategy, to make it harder to comment, thus filters in only who have something to say, and are motivated to do so, perhaps because they have been reading Dave for a decade and consider him a friend. Other bloggers may have the opposite problem – too few comments, so they want to make commenting easier in order to get comments, but then they are likely to get trolls first.
Modifying comments – leaving the inappropriate comments on site, but altering them in ways that makes them much harder to read, or making the commenter look silly, e.g, by inserting a picture of a bunny rabbit, or disemvoweling or using the Kitten Setting. The lightest ‘touch’ is to leave the comment as it is, but remove a link contained in the comment if it leads to a site you do not want to send traffic to. And yes, all of this is completely legal, and a very good strategy.
Engagement – the most important element of comment moderation is the presence of the author in the commenting thread. Responding to readers’ comments, thus showing that they are being read, observed and appreciated, is the most effective way to make sure that the discussions stay on topic and do not veer over the line of appropriateness. Sometimes a comment hurts, or makes you angry. Sleep over it. Then come up with a smart, witty, civil and firm response. Be in control of your own commenting threads:
So, why are so many comment threads so nasty?
Because they are not moderated! At all! In ALL of the senses listed just above. If commenters think your commenting thread is a public space where they can do whatever they want because nobody’s watching, they will do whatever they want. And that is not pretty. And then the potentially constructive comments never get posted, because normal people do not want to waste their time thinking and writing comments that will just get flamed.
If you don’t delete or disemvowel inappropriate comments, people will think you are not even reading the comment threads. If you don’t show up in person, nobody will know you are even interested in their thoughts. If you don’t delete the trolls, the trolls will take over and the nice people will go somewhere else.
Early online discussion spaces, e.g., newsgroups, were largely unmoderated affairs (with some exceptions). It was a Wild, Wild West. When first blogs appeared, the spirit of free speech permeated the early online discussions – bloggers felt they should let everyone have their say, because their blogs were rare spaces where people could do that.
Then traditional media got into the game and started allowing comments on their articles. And that is where everything broke down. Due to a grave misunderstanding of an old court case by the legal departments (I really, really hate to use nasty words like “idiotic”, but my Thesaurus has come up empty with suggestions for apt synonyms) at some media organizations – which then spread to all of them – newspapers decided not to moderate their comments at all! Not using any of the methods listed above. Actively preventing authors and editors from deleting, editing or responding to comments! Really! What did they expect to happen – intellectual treatises occurring on their own in each comment? Perhaps they thought the comments would be just like Letters to the Editor, but already edited, chosen and filtered automatically before they even arrive?
Yes, all methods of comment moderation are perfectly legal and don’t let any media lawyer tell you otherwise – they keep getting it wrong.
Free Speech is a very American concept. Most of the other 200 nations on the planet do not provide constitutional protection of free speech. And Internet is global.
And even within the USA, the concept of free speech does not mean everyone has the right to say everything everywhere. It does not mean you have the right to say your stuff on my blog. It means you have the right to start your own blog. Just because I have commenting functionality on my site, does not mean you have the right to post whatever you want on it. Every host of every site has the right to delete, edit, or modify any comment in any way, to ban users, and to implement whatever moderation norms and techniques one wants.
Commenting is a privilege, not a right. You have to earn it.
While early bloggers were generous, giving their rare online spaces up to public discussion, there is no need to feel so generous any more. Starting one’s own blog is easy these days, and ranting on social media is even easier. There is plenty of space for people to discuss stuff, and that does not have to happen on your site – the era of such generosity is mostly over, and most veteran bloggers have severely tightened their commenting rules over the years.
And yes, some blogs are still rich with vibrant commenting communities – e.g., Atrios, Pharyngula, etc. They have people there who talk to each other every day, often ignoring the topic of the actual post. And then mega-blogs, like DailyKos, are a completely different animal, where community provides most of the content, in form of diaries (as well as lots of comments).
My blog is my living room in my home. I set the rules. I determine the tone. I determine the topic of conversation. When you post a comment on my site, you agree to abide by my rules, you stick to the topics I determined, and you keep the tone I deem OK to be used in my home (imagine reading out loud your comment in front of my wife, mother and kids). I have the right to warn you and to kick you out of my home – it’s my party, after all. You have no right to be here, no right to say anything – it is up to me to welcome you here, and up to you to ensure you are welcomed.
Another way to think of your blog (which I heard from Anton Zuiker) might be as a classroom where you are the teacher. It is important to keep control if you want to facilitate a constructive discussion. Unless you are willing to use your time and energy to keep control and engage in the discussion yourself, you may be better off not having comments at all.
If I delete your comment, it is not censorship (and if you cry “censorship” I will laugh out loud). You are free to start your own blog and start working on increasing its Google Rank so it becomes visible in searches. It takes time and effort, but I will not lend my Google Rank to you to use for your blatherings. Do the work yourself.
In this comment on a recent post of mine, I was commended for having a good, vigorous, constructive discussion. How did I manage to do it? By swiftly deleting about a dozen trolling comments as soon as they were posted. If I did not do that, half of the good comments would not have been posted as their authors would not have bothered. The discussion would have veered off-topic onto some silly tangent, and trolls would have taken over.
And it is not about blocking every opinion that is different from mine. Obviously, some of the commenters disagreed with me on the content and conclusions of my post. And there are a couple of comments there by obvious climate dieniers that I left standing because they were on-topic and civil in tone, despite me not agreeing with them one bit. It is not about censorship, it is about tending the garden of one’s commenting threads, by nurturing the good flowers and removing the weeds.
“You are entitled to your own opinion, but you are not entitled to your own facts.” – Daniel Patrick Moynihan
Different websites and blogs have different goals. This one, see the banner above, has the word “Scientific” in it. This means something. This means that the discussions are about science and about the way science and society interact. This also means that the content of our site tries to present factual information about the world, as best discerned from scientific data.
This magazine is 167 years old. The magazine and its website and blogs often cover the latest studies that add just a new twist to an old, established body of work. Some articles summarize larger, established bodies of work.
Early in the magazine’s history, during its first couple of decades of existence, it was established that biological evolution is a fact. Since then, we cover, daily, new twists and turns in details of exact mechanisms of how evolution works, or about its results, e.g., discoveries of new fossils or new living species. But the fact of evolution is long established. Thus, almost none of our articles discuss this fact. Thus, a constructive comment thread on any of those news articles would involve discussions of those fine points of mechanism, or details of the new finding. A comment thread that debates the fact of evolution is off-topic. It is not “vigorous debate”, it is a comment thread hijacked by creationist trolls. They are entitled to their beliefs and opinions, but they are not entitled to their facts. Evolution is a fact. Questioning the fact of evolution is not a part of discussion of a particular new finding or mechanism. There are other online forums that discuss that – this is not one of them. Thus, such comments are trolling, and need to be deleted so people who do have a scientific mindset have a free and welcoming forum to discuss the details of the study described in the article.
The same goes for climate change. It is a fact that global warming is true. And it is also a well established fact that humans played a big role in it. And the notion that if we broke it we should fix it is what responsible humans do. Thus, an article about a new study about climate or weather or energy or infrastructure is not a proper forum for debating the well-established facts. There is no debate there. Thus, such comments need to be deleted.
Now let’s go back to the very beginning of this post and the forthcoming article about the effect the tone of comments affects readers. If we leave the creationist or denialist troll comments up, what does it do to the rest of the readers? It polarizes them, it makes them more certain about things than what their actual knowledge warrants, while at the same time repelling experts from wading into the mud-pool to correct, over and over again, the untrue statements and anti-facts posted by denialist trolls.
How do you decide what is a trolling comment?
The first definition of trolling is ‘posting comments in order to derail the discussion’, to take it away from the topic of the original article and onto a topic the commenter wants to discuss – his/her own pet peeve.
If you want your comment threads to remain clean and civil, and to stick to the topic in the article, you HAVE to delete off-topic comments.
So, if I write about a wonderful dinner I had last night, and somewhere in there mention that one of the ingredients was a GMO product, but hey, it was tasty, then a comment blasting GMOs is trolling. Any comment that contains the word “Monsanto” instantly flies into the spam folder.
If I write about a wonderful weekend mountain trek, and note I saw some flowers blooming earlier than they used to bloom years ago, then a comment denying climate change is trolling. I am a biologist, so I don’t write specifically about climate science as I do not feel I am expert enough for that. So, I am gradually teaching my spam filter to automatically send to spam any and every comment that contains the words “warmist”, “alarmist”, “Al Gore” or a link to Watts. A comment that contains any of those is, by definition, not posted in good faith. By definition, it does not provide additional information relevant to the post. By definition, it is off-topic. By definition, it contains erroneous information. By definition, it is ideologically motivated, thus not scientific. By definition, it is polarizing to the silent audience. It will go to spam as fast I can make it happen.
For a science site, every comment that insert non-scientific, anti-scientific, nonsensical, ideologically or religiously motivated anti-facts, is by definition not just trolling, but spam. Like online Viagra sales. Literally! There is more and more evidence that a small subset of trolling posts (e.g., those aggressively promoting climate denialism) may be paid for by astroturf organizations funded by some vested interest groups. By peppering every article and post that can remotely have anything to do with their topic of choice, they provide an illusion that their pet movement is bigger than it really is, or that support for their position is more widespread than it really is (which then, if it works, results in the actual rise in the support for their anti-science positions). This then encourages the others (after they got persuaded quickly, without having their own sufficient knowledge, as the nanotechnology paper showed) to keep posting additional comments for free. The first troll comments are supposed to be seeds for more trolling. Which is why it is essential to cut them at the root. You do not want to provide a free platform for a paid political operation.
I am certainly not using cowardly, mealy-mouthed He-Said-She-Said mode of writing my own posts, so I will also not allow for a He-Said-She-Said pseudo-debate to develop in my comment threads. You don’t like it? Deal with it. Go and complain in the comments on Watts’ posts, or on your Facebook wall.
And the idea that deleting inappropriate comments reinforces the formation of “echo-chambers” is a complete myth. Plenty of different opinions are out there, many more of them much more easily available than before the Web. The commenting threads are not a place to showcase the whole spectrum of opinions, no matter how outrageous some of them are, but to educate your readers, and to, in turn, get educated by your readers who always know something you don’t.
My own moderation rules
You are reading this on my own, personal blog. I know, the distinction is fuzzy. The blog is hosted on Scientific American, and I am an editor at Scientific American, thus this blog is in some way a public face of the organization. But writing this blog (or even hosting it on this site) is specifically not included in my contract and in my job description. This remains my own, personal blog. I host it here because it makes sense to me – it is easy (I am here all the time anyway), it feels natural, and it provides me a greater visibility than if I self-hosted it elsewhere.
Now, I am aware I represent the organization in public. Thus, I am very careful that everything that goes up on this blog is following the range of topics and the discourse standards of SciAm. If I think something I have to say does not really fit here, I post it on my Twitter or Facebook or Google Plus or Tumblr. And even there I am aware that I am still seen as a public face of SciAm so I am careful what kind of language I use, how I behave, etc. Deleting trolls, and not providing a platform for anti-science ideas, is good behavior for a scientist, a science writer, and an editor at Scientific American. It demonstrates I care for the truth.
If I want to say something that does not fit my public role as an editor, and if we cannot meet in person or chat on the phone, I will send you an email. I treat email as the last remaining channel for private, confidential communication, so if I tell you something in an email, it is between you and me, and not to be shared with others.
So here, on my personal blog, I like NOT to have pre-written Moderation Rules. Sure, I can be capricious, which keeps commenters on edge. But, really, I am putting a lot of thought into comment moderation, and I think carefully about each comment. I make decisions on a case-by-case basis. What is my post about? The same comment may be appropriate on one post and not on another.
Blasting in here with “Al Gore!” makes it easy for me to decide – you are not coming here to do anything but start a flame war and perhaps get a rise out of me. Perhaps you are paid-per-comment (or you were duped into doing it for free). Your comment is going straight to spam. A few more times, and I’ll ban you (and as there is no way to ban a commenter from just one blog, I’ll ban you site-wide, after consulting with colleagues, something I did only a couple of times in the last two years – I am careful, and do not ban easily, though there are dozens of other regulars who are under my careful watch and coming dangerously close to getting banned, which is fine, as that opens the field for better commenters to take their place).
But if you come in with strong language, or ideas that are not constructive, but I detect that you are just uninformed about my commenting standards (perhaps you just arrived from YouTube or DailyMail or 4chan), I will welcome you in a comment, explain that my kids are watching (so mind your language), and explain my “three strikes and you’re out” rule I sometimes employ on commenters like that. The commenters who get greeted that way either disappear forever, or they make the three strikes so get banned, or they tone it down and become regular, productive members of my commenting community (even though some of the same people may have been banned from many other science blogs).
I also have moderation powers on several other spots on the site, e.g., on Guest Blog, Expeditions, The SA Incubator, The Network Central and if I post myself on some of the other spots (Observations, @Scientific American, or on the main site). I employ pretty much the same rules there, though I am even more careful, and put even more thought into each comment. I especially want to protect our guest bloggers – some of whom are academics and not used to rowdy online behavior – from at least the worst core of our regular trolls.
The rest of the bloggers on the network employ their own rules, use their own judgement. Some pre-moderate, some post-moderate, some are very strict, some are very lenient. And we do not (yet) have more sophisticated commenting tools here. Especially female bloggers often have little choice but to be strict in their moderation, as the Web seems to bring out the nastiest mysogynists in droves.
I do not moderate the rest of the site – articles or posts written or edited by other SciAm staff are moderated by them.
Now, I know that I used the example of Global Warming Denialism here the most – mainly because it is currently the most acute problem on our site – but the same goes for people harboring other anti-scientific ideas: creationists, anti-vaxxers, knee-jerk anti-GMO activists, and others.
This post is not about climate denial, it is about commenting and comment moderation. It is about the fact that eliminating trolls opens the commenting threads to more reasonable people who can actually provide constructive comments, thus starting the build-up of your own vigorous commenting community.
There are seven billion people on the planet, many of them potentially useful commenters on your site. Don’t scare them away by keeping a dozen trolls around – you can live without those, they are replaceable.
Thus, on this post, comments about climate will be deleted.