A neat finding about pseudonymous commenters—and why you should question it

Here's some interesting data that I would like to believe is true—mainly because it matches up with what I've experienced here at BoingBoing. Many of you use some kind of pseudonym in the comments, whether it's first-name-only, an Internet handle, or a completely fake name. My experience here has taught me that, despite this, you all are perfectly capable of writing fascinating, informative, worthwhile comments and having good discussions that add to the usefulness of the original post. (That doesn't always happen, as I'm sure Antinous will attest. But it happens often enough that I talk y'all up to other journalists and bloggers who are nervous about having a comments section on their site.)

After an analysis of 500,000 comments, Disqus now says that pseudonymous commenters are the most prolific commenters—and that the quality of their comments are actually a little better than the quality of comments from people who logged in through Facebook, using their real names.

If this is correct, it's pretty cool. It might not be correct, though. So do think about that before you start touting this as absolute fact in the #nymwars. For instance, the key measure of quality here is whether or not a post generates "likes" and replies, and, if so, how many. Another thing I've learned from watching the comments on BoingBoing: Likes and replies are not necessarily indicative of actual quality. Likewise, the measures that branded a post as "low quality" seem designed to really only address the worst-of-the-worst: Comments that get flagged, deleted, or marked as spam. There's a lot of room left over for comments that are low quality, but not outright trolling/spam.

Another issue: "Real identity," in this case, means "logged in through Facebook. I can think of several of you, off the top of my head, who I know use real names in the comments, but don't log in through a social media site.

Finally, I can't find anything about where the 500,000 comments were pulled from. Depending on the site(s), this may or may not be a representative sample. After all, the site you're posting on—what the content is, what the community is like, how well moderated it is—probably does a lot to influence how you behave there.

So, basically, what I'm saying is this: Disqus has published an infographic confirming my personal beliefs. Hooray! The problem is, I don't really feel like I can trust it.

Image: jack masque, a Creative Commons Attribution (2.0) image from speculummundi's photostream


  1. It would also be more interesting, to me, if there were a way to compare long-term pseudonyms with short-term ones; anyone can make up a name and use it once to comment if a blog has disabled anonymous commenting. Or register multiple accounts. But there’s a distinction there between a random pseudonym and an established online identity.

    1.  Or somebody who’s had previous accounts banned can pretty quickly create a new account and continue to troll the comments section.

    2. Just occurred to me that I use a pseudonym yet it is tied to an actual picture of me.  Does that make me anonymous or not?

      I guess I have this vague hope that someday I’ll bump into someone on the street and they’ll say “OMG you’re dagfooyo!”  Well a guy can dream.

  2. I tried to think of something insightful and interesting to post on this article but just couldn’t do it.  – MB44

  3. One difference that stands out is that FB logins may not be as devoted to the site as site-specific logins. I’ve had a BB login since way before BB went Disqus, but many other sites that didn’t used to accept FB logins, I just didn’t bother commenting. Now that FB / Google / Twitter logins are more common, this distinction may be going away.

  4. Ah yes, the science of likes. The best ways to get them are to repeat conventional wisdom in one sentence, or give a quote from media that’s popular among the audience for that site. Also, try to get in at the top of the comments. The best way to get replies is to troll.

    I think sites that force a trackable identity tend to be more mild and polite, with the exception being Tea Party types. I think this is because they get so much validation in their local real-life communities for being rabid that they see no need to self-filter.

    PS: Overall on Disqus right now, I have 790 comments and 1984 (!) likes. That is not too bad a ratio I think, though it’s been higher in the past.

    1. When comments are threaded, the first post will get more replies.  I really don’t care much what Mujokan has to say here, but it’s the first post, so I’m responding to it, and my comment will be read before those who posted sooner than me but below.

  5. Maggie, that last paragraph of yours is a perfect example of why I look forward to your posts most of all on this site.  Generally, you at least attempt to provide a balanced perspective and failing that, you’ll own up to your own personal bias.

        1. Seriously, you’re one of my favorites here. That ability to glance around the corner of your own ideology and formulate a thought like “I WANT to believe this, but it’s a bit too easy” is pure gold. I wish it were contagious.

      1. Oh, great.  This must be one of those “one guard always lies, one always tells the truth” riddles.

        1. Yes, but the word “untrustworthy” at least provides an out whereas “always lies” wouldn’t. A random bit string is, in some sense, more untrustworthy than the negation of every bit of a true string.

  6. ‘For instance, the key measure of quality here is whether or not a post generates “likes” and replies, and, if so, how many. Another thing I’ve learned from watching the comments on BoingBoing: Likes and replies are not necessarily indicative of actual quality.’

    I can just imagine applying that standard to Youtube. Any given argument between people with enough followers, or pre-polarized topics like whether gods exist, could result in comments getting lots of likes and dislikes even if the comments are along the lines of “Yes you did,” “No I didn’t.”

    1. Although I generally looked on Likes as a metric with a jaundiced eye, the YouTube comments that get Liked to the top do tend to be pretty hilarious.

  7. I can confirm this not a new phenomenon. I’ve been researching academic debates from about a century ago that were held in various Scottish newspapers. The same thing occurred in the majority of cases.  Many scholars of the time wrote excellent insightful articles under pseudonyms. Of course there were exceptions: people writing rubbish under pseudonyms and people writing excellent things using their own name, but as a rule it seems to have been the same.  Of course it could take weeks or months to generate the same amount of comments that I see above in this thread.

  8. I know people on Facebook using pseudonyms…  I guess that doesn’t help.

    Hey, if you can’t trust a 6 foot plus white guy who signs himself “Itō Ittōsai” who can you trust?

  9. Judging quality by “likes” is absurd, though I’m not sure I can think of a better way. Here on BB you can generally garner a raft of likes just by commenting “Christ, what an asshole” on any subject whatsoever.

    1. Christ, what an… 

      …insightful reality check on the positive reinforcement mechanism that incompletely reflects actual emotive ebb and flow in a discussion.

      In the long run, I think of the “like” as well as retweets, etc, as something like synaptic weighting in a neural-net… in the vast churning ocean of communication, it’s a tiny bit of energy pushing what’s important toward the surface, into the consciousness of the hive mind, the working memory of the web.

      1. That is the most brilliant thing I’ve read all day.  Wish I could like it more than once to push this concept closer to the surface of the neural net.

  10. We talk about pseudonyms as if they were merely false name tags, but in fact many people have full fledged personas to go with their pseudonyms.  When they use a particular pseudonym, they also assume that identity.  It helps focus one’s attention.

    There is a long history of public writing under an assumed identity.  I’m sure Hamilton, Madison and Jay had many other areas of interest in which they could wax eloquent for hundreds of pages, but when they wrote as Publius they maintained a core identity that stuck to the subject at hand.  There are so many academics who have switched to a different moniker when they write fiction that anyone can think of at least a few off the top of their heads.  Closer to the here and now (online blogs), I think of someone like Cuttlefish on ScienceBlogs.

    Or even myself.  I have several personas, with separate names and even writing styles for each.  I only use one persona per forum or group, so I don’t believe I’m doing anything “wrong” (no sock puppetry, for example).  There is literally NO overlap between my different personas….and yet, they are all legitimate representations of me.  And I’m no one….no blog, no public following, etc.

    We contain multitudes.  (apologies to Walt Whitman)

    1. I have used this tag since I think 2006, with this icon since 2008, and I’m sure I have tens of thousands of comments under it. It is just me except that I’ve been drunk or stoned for a lot of posts. If the man ever enforced real-name commenting though, I would quit immediately without a backward glance, as appropriate for the destruction of Sodom and Gomorrah.

    2. Could not agree more. I’m a writer with a nom de plume, I have three distinct, long term pseudonyms on BB, Reddit, etc and a fake FB name. I consider each of those names to represent me just as much as my real name.

      In fact, my real name isn’t even my own, because I tacked my husband’s name on to the end of it.

    1. Sure he would.  Mark Twain is a brand.  Facebook loves marketers (aka brands).  Whether he could have a regular Facebook account in the name Mark Twain is a different issue altogether.

  11. my real name is the most common name in the US when i was born (i’m a brit so damn you you name thieving rebel scum). everyone knows me as bob brown. thats a hell of a lot more anonymouses than dreddpiratebob.

    its wierd that i have to use a made up name to be MORE traceable.

  12. I don’t sign in through Facebook and I use my real name. I wonder if I’d be counted as pseudonymous via these metrics. Like others have said, you can game “likes” and replies by getting near the top, and strongly taking either side of an issue. If you get in early on, say, an abortion post, you’ll get a lot of likes with either “abortion is murder” or “women have the right to control their own bodies”. As a bonus, you don’t even need to read the original article to come up with these, which helps you get that “first” response. Meanwhile, if you take ten minutes to read the article and half an hour to compose a thoughtful, multi-paragraph response, you’re on page two of the comments and many people will skip over your reply for its length.

    1. This is true of the internet in general, and particularly sites like Slashdot and Reddit which attract much more comments than BoingBoing does on a regular basis.

      I think the dynamic here in the comments is different (and, as we know, is carefully cleaned up by Antinous and the moderators in a way that other sites don’t do) and I regularly see posts further down the page that have as many or more likes as the ones near the top.

      By the way, I find it funny that you mention not reading the article in order to write a comment quickly, because if you had read the entire article you would know that Maggie addressed the issue of people who use their real name but who don’t sign in through Facebook :)

  13. Boing Boing wouldn’t provide a very represenitive example either. The only people commenting on Boing Boing are those that Antonius has not gotten around to banning yet. The survivors can best be described as the ultra liberal counterpart of Rush Limbaugh’s “Ditto Heads”.

    1. Hey, if people can’t abide the moderation policy, they can take their eyeballs elsewhere.  It has nothing to do with liberal vs. conservative.  If you can’t be civil, you will get banned. 

      1. Yeah, but he’s just kinda hanging out in the lobby with his hands stuck in his coat pockets, making a pointed huff and scowl occasionally. :)

        Oh, wait, you mean he hasn’t been banned yet. Yeah, funny how he’s still here to complain about how y’all banned everyone who complains. :)

    2. You know, I’ve been pretty harsh on Antinous myself — I have generally not enjoyed the way he’s addressed people who have disagreed with him here — but that “only people” you’re referring to… seems to be a hell of a lot of people left, and they seem to me to still reflect a pretty diverse opinion. 

      And you know what? It’s still a personal blog. It’s gonna reflect the ideology of the people who write it. It has a culture. If you come in here really hostile to that culture, you’re probably not gonna be real popular. I have seen plenty of people express ideas contrary to ultra-liberal dogma and not be punished or antagonized in the slightest. (Admittedly, if you come in here regurgitating equally fanatical stuff from Newsmax or Lew Rockwell, and do it rudely and confrontationally… yeah, you’re gonna get slapped… and I’m 100% fine with that.)

      Hell, I’ve openly called Antinous a shithead in a reply to him and he has not yet seen fit to ban me. I do have to admit I respect that. :)

  14. I figure that in a few years, everyone I know will know everything I ever wrote on the Interwebs, anonymous or not, due to fancy natural language processing algorithms that will match posts to me based on IP proximity and then based on the language usage structure of the posts themselves.  So, since I’m going to be fucked, I just speak my mind and act like who I am in real life anyway and try not to overblow myself.  Cuz, you know, jerkoffs are the ones who like to blow themselves too much.  The rest of us only blow ourselves a little bit.

  15. *blink*
    I get to say things as TAC that I can’t say as myself, not because I don’t believe in my words but because the repercussions of my some of my words could cause a world of problems for meatspace me.
    This, believe it or not, is a nym I use.  I have several nyms, there are other nyms but this is MY nym.  I made this point during the Google+ wars, just because I am a nym does not invalidate what I have to contribute.  I should be judged on what I contribute and not the fact I happen to be doing it using a nym.
    Some people use nyms for evil, to assume all nyms are evil is stupid. (and qualifies you to be a congressman)

    This nym is a full identity for “me”.  Another nym I use is a full identity for “me.”
    These identities fit different things, different groups, different needs.
    This nym is sorta political, sorta freespeech, very anti-copyright troll, and very anti-stupid.  Another nym I use doesn’t talk so much about trolls, because the community it belongs to isn’t that interested in troll discussion.
    There are pieces of me in each nym, the same mind drives them, but how and what I choose to expose under each nym is shaped by my concern for meatspace me as well as keeping my nym true to itself.

    Not everyone does that, but if you remove the obvious trolls trolling under nyms I expect you will find people using nyms feel more free sharing things and sharing more leads to more understanding and greater context.  Someone might feel better sharing a story of childhood abuse to add context knowing they will not have to deal with people’s reactions to them being altered in meatspace.

    But then… I’m just a nym… I could be a very well composed troll…But you’ll have to judge my words to decide on my character.

    If you discount me for not using my meatspace name, that is your choice… but how much are you missing out on by doing so.

    I don’t blame you if you think nyms are evil, look at the worst nyms we have out there today.  Senator/Representative, they will say ANYTHING to win the discussion even if deep down they know its made up or trolling the audience to win.  They will even change their opinion mid-sentence if they think that move wins.  I at least stand by my statements, and admit when I was mistaken.  When is the last time a congresscritter said oops I was wrong?

    1. I don’t blame you if you think nyms are evil

      TAC, did you read the post? I don’t think nyms are evil. In fact, I’m inclined to believe that nyms are excellent commenters … or, at least, no worse or better than commenters who use their real names. 

      I just think the methodology Disqus used is so flawed that I can’t believe its conclusions, even if they back up my beliefs. 

      You’re arguing against anti-nym forces that I don’t think are particularly present in this forum. 

      1. I was using the “royal” you, (like the “royal” we).  Not specifically referring to you yourself Maggie.  There are people out there that cling to the idea that everything in the world would be better if we all did everything in our own names.  Was using you in the context of the individual person reading my response, people have different opinions on the subject.  I find it helps to validate that you understand their position even if you disagree.

        While the Disqus methodology might be flawed, they do only have x data available to them to work with.  I’m a member of several different websites using the TAC nym (some Disqus some not), and I have often seen both sides of it… the troll nyms and the nyms who have something to share they might not otherwise do.  I find that the communities adapt to dealing with whatever flavor of real/nym they encounter, based more on the content than the name.

        I think there is inherent value to having nyms participating, and I would suspect that a larger study would backup the idea that people freed from the “constraint” of others knowing they said that always leads to better commenting.  It would also show a large number of people proving that you can be an ass online easily as well, but I doubt many of them commit to a nym for very long.

        I also think using FB as a source for a data point is not very fair either, one only needs to read LameBook for a little while to see why.

        I think the biggest problem with a finding like this is – we want it to be true because we have seen the “real world” results, but how does one quantify if a post is worthwhile?  What metrics really matter?

  16. I like the pseudonymous ability to like my own post, then to weakly disagree with my post under another name, only to come back as a third version of myself and gang up on my disagreeing self.  It is all so validating.

  17. As a guy who’s been managing online communities for a dozen years, I find the idea that anonymous commenters produce better posts suspect.  (Perhaps in very specialized communities, like a medical forum, that might be true.)  My observation suggests that anonymous posters can and do post stuff that’s every bit as insightful as those using real names.  But, essentially 100% of flamers, trolls, and spammers post anonymously, which is bound to bring the average quality down.

    Even normally polite and respectful people can get carried away when they are anonymous. 

    1. Anonymous != Pseudononymous, I thought that was the point of the original post?  (Point being that forum participants who consistently use a particular nick / handle / avatar / persona / whatever build up a social history and contextual identity associated therewith, as opposed to the total anonymity of a one-off fly-by…)

  18. I’ve got a couple of different nyms that I’ve been banging around the net for the last 10 years or so.  I don’t have so much comment history here because Disqus annoyed me for a very long time.  (Too! Many! Logins!)

    That having been said, I’m seeing a certain parallel in which FaceBookLogin is to web forums as AOL was to USENET – namely, the easy-to-use vehicle by which a whole lot of new people could very easily access a forum which had previously required a certain degree of technological aptitude to get into…  

    I leave the rest as an exercise for the reader.  

  19. My pseudonym is a nickname that I have actually been called (with an addition of my geographic location because someone else already had my nickname).  I like to use it as my log in on Disqus to avoid diluting my FB identity with actual intelligence or valuable opinions.  *duckface!*

  20. This makes me want to write well. This ‘nym’ is, when pronounced a certain way, how my birth name sounds. All artwork I make is also presented with this referent. I like liminal spaces!

  21. If the criteria you are using for “real name” is “logged in through facebook” then that’s not accurate at all.  I know a ton of people that don’t use their real name on facebook.

    What you have here is data suggesting that people that log in through facebook write less well “liked” responses.  You could posit that this means the quality of posters that come to us through facebook or would like their posts associated with their facebook is lesser.

    In general I would say I support that hypothesis.

  22. Wow, quality is measured by likes and replies. Guess that means we should junk the Oscars and stick with the Peoples’ Choice awards. Big Macs are better for you than spinach. Snooki is a better actress than Meryl Streep.

    1. In Snooki’s defense, she does have everyone believing she is human and not the love child of an oompa loompa and Elvira – Mistress of the Dark.

  23. I thought the DISQUS article was insightful and at least showed a trend. Many comments associated with using one’s FaceBook account stated they didn’t like SPAMing their FB “friends” with comments posted all over the Web. I agree with that logic and wouldn’t think of signing in with any of my FaceBook accounts even though they are pseudonyms too.

    I tend to comment a fair amount, spreading my comments across my different DISQUS accounts. As to likes, I’m doing OK, but that may be a factor of being high in the Top 10 commenters on a few sites.  This account (which is about a month old) only has 452 comments and 353 likes.  My favored Account (not this one) has 16,394 comments and 7,008 likes.

    What you can do (and I don’t) is go to a national level “special interest” site and preach to the choir. I have seen commenters with hundreds of Likes for a single comment.

  24. I really just wanted to leave a pseudonymous comment on a post about pseudonymous comments because I thought that was pretty meta.

    …and “Blake Morgan” is definitely my pseudonymous name. LOL wut.

    I guess if you sign in with twitter here, it exposes your real name and not your twitter handle as I expected. huh.


  25. Is this a relavent enough thread to express my distaste for Disqus? 

    Sure it may appear palatable to some when put next to Farcebook logins, but it’s still akin to saying you’d rather eat hardened pig stool than sow diarrhea.  For the record, I use ‘nyms on both.

  26. Oh, it’s all about like-whoring? 
    I give “likes” generously like I would smiles and nods in a real conversation. 
    That’s how, it feels like, I get them. 

    On the subject:
    I think there are a great many semi-lurkers that only say something when they really, really feel they have to say something worthwhile. 

    Anyhow, the real high quality comments are those that are made days late, comments that nobody even reads. 
    Like this one. 

  27. “Real identity,” in this case, means “logged in through Facebook.’

    I had to laugh at this.  I use the handle GrumpySteen all over the internet and it’s as real and identifiable as my real name as a result.  I also have a Facebook account which I use to get free stuff.  Absolutely everything in the Facebook account other than my gender is a blatant lie.  If I logged on through Facebook, it would be because I didn’t want to reveal my identity.

Comments are closed.