Huffington Post founder Arianna Huffington says the site will no longer allow anonymous comments — but by taking such a step, the site will be giving up something of value, and may not even solve its troll problem.


According to statements made by Huffington Post founder Arianna Huffington in Boston on Wednesday, the site plans to end anonymous comments next month, as my Gigaom colleague Barb Darrow has reported. Huffington said that she has decided there are too many “trolls” using the site who hide behind anonymity in order to make violent or offensive comments, and that she believes people should “stand up for what they say.”

Is anonymity really the problem with online comments? I don’t think so, and there is at least some evidence that supports my argument — but more than that, anonymity has real value and giving it up has serious consequences.

The Huffington Post founder suggested that her feelings about anonymity have been colored by recent rape and death threats against women in Britain, even though most of these — including a sustained attack on freelance journalist Caroline Criado-Perez that took place over a number of days — actually occurred on Twitter (which has repeatedly defended its users’ rights to remain anonymous, or at least pseudonymous). According to Huffington:

“Trolls are just getting more and more aggressive and uglier and I just came from London where there are threats of rape and death threats. I feel that freedom of expression is given to people who stand up for what they say and not hiding behind anonymity. We need to evolve a platform to meet the needs of the grown-up Internet.”

Do we need a “grown-up internet?”

Arianna Huffington

Huffington’s views on comments — especially the part about using real identities being part of the “grown-up internet” — are very much in line with other arguments that have been made in the past about the dangers of anonymity, including former Facebook marketing director Randi Zuckerberg’s comments in 2011 that anonymity on the internet “has to go away” because of online bullying and other bad behavior. Google executive Vic Gundotra made a similar argument when the company launched its Google+ social network and required the use of a “verified identity.”

The logical extension of this argument is to make online anonymity impossible, or even illegal. At least one U.S. legislator has tried to float a bill that would do exactly that: Illinois senator Ira Silverstein proposed that web users only be allowed to comment if they verify their identity online. But would this actually reduce the amount of trolling or flaming or offensive behavior in comments — or anywhere else? Unlikely.

Not only is this fairly obvious from seeing all kinds of abusive behavior occur on Facebook,, but some evidence has emerged that shows requiring real identities from online users actually accomplishes surprisingly little when it comes to changing trollish behavior: in 2007, South Korea mandated verified identities for all users of sites with more than 100,000 visitors, but scrapped the program in 2011 — because requiring real identities only reduced the amount of nasty comments by .09 percent.

Requiring real identity loses something


The debate over anonymity and its value, both in comments and elsewhere on the internet, has been going on almost since the web was invented. I’ve tried to make the argument numerous times that there is real value in allowing anonymity, a case I also fought for when I was the social-media editor at a major Canadian newspaper, where I had to defend the use of anonymity almost daily.

For me, the question isn’t really whether requiring the use of verified identities or registration numbers or some other system might help reduce anonymous trolling — it’s what we would be giving up in return for that small improvement. As a number of other defenders of anonymity have pointed out, there are valuable things we can learn from commenters that they would never contribute if they had to attach their real identity to it. Comments about spousal abuse, sexual identity, religious persecution — the list goes on.

Is a modest decline in trolling worth giving all of that up? And it’s not just those kinds of comments — a recent survey by comment-software company Disqus found that pseudonymous comments provided some of the most value across a fairly huge number of sites using the company’s software.

Why not try to engage instead?

One of the last times this debate came up, Anil Dash made what I thought was a persuasive argument that if your site has too many trolls, that is your fault for not engaging more with your readers and setting the tone for your comments. Obviously engaging with millions of commenters every day — which is what the Huffington Post routinely gets — is virtually impossible, which is why the site has 40 moderators and a suite of algorithms to handle the flood. So why not try to look at comments differently?

The New York Times has made at least a small effort to improve its comments, by promoting some readers to ** status and allowing them to comment without moderation, and by highlighting some comments alongside its stories instead of leaving them all at the bottom. Both of these steps are a significant incentive for good behavior. Gawker Media gives its commenters their own blogs, and lets them contribute their own stories — something that should be appealing to a site like Huffington Post, which got its start by allowing almost anyone to blog for free. Why not create tiers of commenting to encourage better input?

Do we encourage trolls and offensive behavior when we allow people to contribute anonymously? Perhaps. But free speech comes with a price, and I think we lose something significant when we start requiring people to verify their identities before we listen to what they have to say. If that’s what is required for a “grown-up internet” then I would like to stick with the one we have.

In closing, here’s a TED talk from Christopher “Moot” Poole, the founder of 4chan, about the benefits of anonymity.

Post and thumbnail images courtesy of Shutterstock / Andrea Michele Pasquiado

  1. i just figured #nsa had put the screws to her/huffpo

  2. I agree with La Huffington … to a point. I agree that expensive human monitoring is wasted on people who are not prepared to be themselves. And I agree that it is hard enough to secure the rights of real people, without having to worry about the “free speech” issues of avatars. Nonetheless, I would allow anonymous comments, but would subject these users to the decision of unfeeling digital arbiters: fall foul of the algorythm and the account is closed, no questions asked. Call it a Platonic answer: allow two levels of digital citizenship, and let the robots deal with the unreal people.

    1. What if determining who’s “real” becomes too hard? Can we separate the ‘full-valued’ citizens from the ‘partial-value’ citizens by some easier to screen for characteristic? How about skin color? That’s pretty easy to determine and there’s a long history associated with discriminating on the basis of that characteristic.

      Whip up an algorithm and no individual would have to be responsible for any decisions reached by this method. Rather like Obama did with The Disposition Matrix for determining drone assassination targets.

  3. If the Huffington Post were to take the steps necessary to actually eliminate pseudonyms, they would simultaneously block and discourage the majority of their site visitors from ever coming back.

    If they go at it half-assed, they’ll just block the more obvious sorts of “unreal” names which will keep the ill will focused in that part of the Huffpo readership that most wants to engage with the site.

    In fifteen years, “What’s your ‘real’ name?” will replace the long standing “Keep off my lawn!” as the favored bon mot of the elderly.

    1. Good points, Snertly.

      1. I used to run a website that had a popular forum in which medical researchers and academics used to post pseudonymously. They did so because they feared posting using their real names would have resulted in both virtual and real world harassment from animal rights activists.

        Eliminating anonymity would reduce (maybe) some kinds of harassment, but also enable other kinds of harassment.

        1. Another great point — thanks, Brian.

        2. It’s more than that. People have varied opinions on lots of things. You might have some views on one subject, but quite different from your bosses. We don’t live in a fair world, and if those bosses see your point of view on the internet, odds are you’ll lose your job, or worse.
          I’m not talking about trolls, I’m talking about sensitive subjects, like abortion, gay marriage, political leaning and so on.
          Oh, and who the hell are they to decide who trolls are? I’ve seen some people on the internet who don’t understand the concept of sarcasm, or simply because they don’t agree with your views just mod down your comments.
          I don’t know what their game is, but it’s got nothing to do with trolling.

    2. Thank you, though in fairness, I must admit that I am somewhat biased in favor of pseudonyms and anonymous identities.

  4. Mathew, the forces for anonymous are running out of time. For almost two decades, the internet has been anonymous friendly, and what has come of it?

    What value has emerged from 4chan? Which anonymous reviews on Amazon actually helped people find better books or movies?

    Has there ever been a truly anonymous reporting of spousal abuse? How could that be?

    You have credibility, Mathew, precisely because we know who you are and what you’ve done in the past. The drive-by dispenser of snark, on the other hand, has no reputation and nothing much to lose.

    Last question: if you ran a bank, would you willingly do business with a customer wearing a mask?

    1. I don’t know about you, but I get value from anonymous — or pseudonymous — reviews on sites like Amazon and IMDB all the time. And while 4chan is an advertisement for debauchery much of the time, even it has produced value for society at times, as Chris Poole describes in the video I embedded.

      As for spousal abuse, I’m not talking about reporting of the actual act, but insight into the effects on a person’s life — the same with religious persecution and sexual identity. I have seen this happen first hand at the newspaper I worked for, and people told me they never would have made such comments without anonymity. That has real value.

      Your banking question is largely irrelevant — we are talking about discussion forums, not lending people money. The criteria are different, and with good reason.

    2. Inquiring minds might ask if the poster is indeed the real Seth Godin?

      I pose the question because for one the writing style seems different.

      Secondly Seth Godin doesn’t often post comments. I’d like to say never posts but I could be mistaken.

      Finally the use of 4Chan as a prime example? Hmmm……

      1. Well, he registered with Seth’s real email address, so I’m assuming it’s him :-)

        1. Ha, probably another hacked account, or am I just overly suspicious :-)

          I didn’t watch the video first time around, so I now understand the 4Chan angle. Sorry about that.

          Oh, scrolling down I see another skeptic. I can see why, in that even though I used my twitter account to authenticate (@highwick) there are no links for the avatars so we can’t see what you can from the backend.

        2. I think both arguments have valid points. Should people be allowed to make violent threats anonymously? No. Should people be able to express themselves and not feel vulnerable? Absolutely.

          Sorry, it sounds unreasonable, but we need to monitor the Internet. It’s not realistic to ask people to use their real identities. The best alternative is to require people to act like human beings. The anonymous trolls who destroyed the young boy’s memorial on Facebook should never have gotten away with that. Is it too much to ask Facebook to monitor that type of behavior? They seem fully equipped to monitor women breast feeding babies.

          1. btw – I did not need to use a real email address to comment. :)

    3. The bank analogy is pretty weak…

      That aside. I would like to point out two things:

      First, real identities have not stopped bullying and harassment in schools or in the workplace. Why do people think it will stop it online?

      Second, the Internet is an amazing community for people who cannot find like-minded individuals within their physical community. These people can feel excluded by their community for political views, personal beliefs or just really being into video games. If these people already feel ostracized, why would outing them online — the only place where their anonymity affords them the luxury of “being themselves” — be a good thing?

      1. Thanks, Sarah — well said.

    4. Hmm. Well:

      -There has been truly excellent anonymous reporting from Syria, from Egypt, and from Tunisia. Also, from Bangladesh, from Russia, and from Cuba. But hey, if it doesn’t occur in America, it doesn’t matter, right?

      -Furthermore, as an outspoken woman on the Internet, let me tell you: The WORST harassment I have ever received came from people whose names and photographs accompanied their horrific comments. Hatred. Rape threats. All identifiable.

      -When I was a younger woman figuring out parts of my identity, the ability to discuss things anonymously on the Internet—on message boards, chat rooms, and yes, sometimes in comments—was incredibly helpful. It allowed me to form my identity without risking the permanence of some of the stupider things I’m sure I said. That’s been hugely valuable to me.

      But please, carry on with your white male privilege.

      1. Bravo, Jillian — thanks.

      2. Was that last line really necessary? If you think being white and male protects you from hate on the internet, that just means you haven’t seen it all. Privilege is relative and should never be used as tool to shut down discourse or to critique the commenter.

        1. Commenters should never be called out on their privilege? How do you assail the position “let them eat cake” if you’re never allowed to use the observation of privilege as a tool to critique the commenter? Calling it out is the only way to assail it because that’s the only rational response to an argument that completely ignores the circumstances of others.

    5. What has become of it? It has become the greatest thing human kind has ever created.
      4chan has done some pretty amusing things, many reviews on Amazon use usernames, and they’re just as good as those of people with ‘real names’. Whether a comment uses a real name or not it’s pretty meaningless to other users.

      Credibility has little to with wether that’s his real name though. For all I know that’s a fake name and a fake picture. You see him as credible because of what he has done in the past, if he had gone with “cheeseduck25″ the entire time nothing would be different.

      You go by “sgodin”, that tells me absolutely nothing. Your image looks like Seth Godin the author born July 10, 1960. But you have no credibility here, because anyone could post as Seth Godin. Mathew’s credibility here comes from being recognised by this site as being him.

      Banks do business with nameless and faceless organisations, shell companies and other legal constructions that make the person more anonymous than many internet users are all the time.

      Besides the philosophical standpoints it’s completely unenforceable.

    6. I’m biased because I’m with Disqus — (although we could easily go to real-name identifies only if we thought we should) — and I have a lot of respect for your thinking, but i don’t think your examples are making your case well at all. What does running a bank have to do with hosting a comment discussion on the web? You really think only real name reviews on Amazon are the ones that matter?

      Perhaps what’s happening here…as we often see happen…is that “anonymous” and “pseudonymous” are being treated the same when in fact they are very different. Reputation and accountability aren’t exclusive to real names.

      Or maybe this isn’t really Seth commenting? His own company, Squidoo, doesn’t seem to require real names: http://www.squidoo.com/lensmasters/d-artist.

    7. I’ve found enormous value in conversations with, and comments from, pseudonymous commenters, who would also be banned under this regime.

      Even if we granted Seth’s premise that somehow anonymity produces bad behavior, this doesn’t do anything to prevent bad behavior under real names.

      1. Out of curiosity, why then do you use Facebook Comments on your blog (if this is real Anil :)?

    8. I have no reason, on the basis of other stuff you’ve written, Seth, to doubt your good intentions, but let’s point out the obvious: A marketer making the case for the reduction of online anonymity, along with ad-driven companies like Facebook, Google, and HuffPo (don’t you think they’d love to have real identities to sell to advertisers?) simply cannot be taken without a healthy dose of skepticism.

      But first, the substance — the idea that real identities will actually be a cure for what ails the web … If we’re cherry-picking examples (like 4chan) to make our cases, let’s look at the comments pages on the Wall Street Journal Online, where real identities are widely used, and where, on any given day, you’re likely to encounter some of the most conspiracy-theorist/racist/nativist diatribes around. Tell me, exactly, how real identity helps there?

      The Web is a messy place. Always has been. When you ask “What has come of [anonymity]?” you seem to be suggesting that, somehow, the Web has become a cesspool, and that anonymity is its root cause. That if we only adhere to real identity, somehow the 4chans will go away, everyone will be well behaved, and the Web, as a whole, will become some more productive, more morally defensible place. What a fanciful idea. You know, we have an analog. It’s called the physical world, and last I checked, people have been doing extraordinarily stupid, unkind, violent things for millennia. Why should we view the Web any differently?

      Now, real identity is great for marketers. This whole multi-screen, on-demand, fragmented attention world is really difficult for you guys to master. But you know what? It’s not our responsibility to make your jobs easier. It’s your job to make us pay attention. “Real identities” don’t serve a people problem. They serve a marketing problem and, while I have no disdain at all for markers, you’re not bound to find a whole lot of sympathy on that one.

  5. I always troll with my real name. I have nothing to hide from the internet, and I figure any illusion of anonymity is quickly shattered when push comes to shove anyway.

  6. James Harradence Wednesday, August 21, 2013

    My wife runs a large business unit (multi $B revs).

    She does not & should not (IMO) post her actual name on say, Trip Advisor, for lots of reasons.

  7. I am shocked each time I hear smart people make this argument, in the teeth of the evidence. It swiftly devolves into ‘Identity Theatre’ where what is attacked is not abuse but things people think look like they are associated with abuse. I’ve written about this at length before: http://epeus.blogspot.com/2011/08/google-plus-must-stop-this-identity.html

    1. Thanks, Kevin — and thanks for the link to your post as well.

  8. I’ve been arguing for years (but nobody listens, sigh) that comments to a media property are, in fact, a franchise for that company, and an important one at that. I have zero problem with curating comments by removing those that offend, and the reason people don’t is simply liability. The lawyers have told us that editing comments in any way makes them a part of our product, and therefore, we’re liable. So what? Listening is the new skill that media companies need. Every newsroom monitors police scanners, because they bring “news” into the shop. Why not apply the same logic to something so obvious as feedback from viewers, readers, followers and fans? Call me a nut, but I believe that 21st Century news organizations owe it to themselves and their “audiences” to monitor, curate and report about the feedback they get, whether it’s via their website, Twitter, Facebook or any other form of “media.” The best way to deal with trolls is to ignore them by removing them. To simply ban all anonymous comments, however, is lazy, and that’s being kind.

  9. Dennis D. McDonald Wednesday, August 21, 2013

    I say, try both approaches to see which one is better. Anonynous comments are the practice now, why not try something else for a while. There are costs and benefits associated with each approach. I, for example, have difficulty debating with someone who assiduously refuses to reveal his or her identity. Maybe that’s old fashioned, but I engage online to engage with people, not avatars. But I admit there are arguments for anonymity but there are also arguments for openness. So run a test and openly evaluate the results.

  10. People need to comment like they are being watched and that their name will be associated with the comments. Having trolls spout non-sense, only degrades the discussions and the value of the conversation. If you wouldn’t say it to someone’s face, then you shouldn’t be allowed to spout it off on the internet, because nobody knows who you are.

    People need to be responsible for their comments and having anonymity is not helpful in the end. It only provokes people to say things that are more ignorant than anything, but who cares if no one knows them…

    1. Where’s your data to support those contentions? Oh, right–there are no data suggesting that requiring ‘real ID’ in comments forums enhances the quality of the debate or limits nonsense.

      Some of the most vile, bigoted and nasty comments you’ll find anywhere are the Facebook comments under any news story. These are proud trolls, man! They think they’re clever, and correct.

    2. What about things you would say to someone’s face, but not to EVERYONE IN THE WORLD’s face? For example, I would probably happily debate many things with a stranger, whether in person or online, that I would not debate with my parents, my boss or my ex-boyfriend, all of whom are easily able to monitor that conversation online if your real name is used. Imagine you’re having that conversation with the people on the comments board in person – but you’re not alone. Every single person you’ve ever met or ever will meet is standing there watching and listening to every word, and recording the whole thing so they can go back and access it again at their leisure. Suddenly you’re having to watch what you say in even the most benign of conversations, nevermind conversations about private, personal things or controversial topics. For example, if you want to discuss problems with your sex life with others who are having the same problems on a medical forum, you had better also be OK discussing the same things with your boss, your parents, your grandma, your children still to be born (because nothing ever really goes away on the internet), your next partner you haven’t met yet, that guy down the hall at work who hates you, and your neighbour across the street.

Comments have been disabled for this post