Pseudonyms Drive Community

This will not be surprising to the AVC community but will certainly be a shock to the "real names" crowd. Disqus has shared some research they have done on three kinds of commenters; real names, pseudonyms, and anonymous commenters. Click on this link and check it out.

But if you are not going to do that, I will summarize the data:

– Pseudonyms lead to higher quality comments


– Pseudonyms are more engaged and active

Comments per user

Of course, nobody in the AVC community will be surprised. Kid Mercury, Fake Grimlock, JLM, LE, Panterosa, Prokofy, etc, etc, etc, etc, etc. We have tons of pseudonyms in action here and they enrich and enliven the community.

Props to Disqus for putting out this data. There are so many misinformed and uninformed people working in social media, even at some of the top platform companies. Hopefully this will cause everyone to think a bit more before forcing the real names paradigm down our throats.


Comments (Archived):

  1. JimHirshfield

    Real-namers come on! Let’s get our sentiment up for crying out loud! (Damn, was that too negative?)

  2. jason wright

    Is there a correlation within an identity group? Is there a Pseudo tribe? 

  3. Carl J. Mistlebauer

    Actually, Carl J. Mistlebauer, is a pseudonym, so does that make my comments “HIGH QUALITY” (as per the comment, “Pseudonyms lead to higher quality comments”) now?

    1. fredwilson

      is it?

      1. Carl J. Mistlebauer

        Sometimes even I ask that question….Having spent the last 5 days trying to prove to Google that I am my account owner and that they need to let me access my email, I have decided that even though I can prove that I am who I said I am (so let me access email, blogger, analytics, and Google+) they obviously believe I am a pseudonym….So, I guess I can now be “Carl Mistlebauer” and start all over.

      2. Carl J. Mistlebauer

        If I say it does do my comments move to the “high quality” column?Personality, outside of Facebook, which obviously is a family and friends networking platform I think the issue of “names” is actually a stupid one.Whatever someone calls themselves and for whatever reason they use to select their identity has nothing at all to do with the quality of their comments.I would like to see a study done of how many “pseudonyms” are individuals and how many individuals maintain more than one “pseudonym”

        1. Mark Essel

          Right on Carl!

        2. tgodin

          I agree too.   There are instances – like Facebook – where to me there is real value in knowing that I am interacting specifically with the person I think I am.   I wouldn’t be likely to share (much) with people I knew only through nicknames.   And there are instances – like here – where real identity is less important.Since I think this debate on AVC started with Google + requiring real names, do you think that G+ is more like FB – where real identity is important – or another beast entirely? 

      3. jason wright

        Not just a pseudonym. There’s more to it than that. Got it yet?A clue. It’s staring you right in the face.

  4. LIAD

    Disqus use logging in through FB as a proxy for ‘Real Names’. Those of us who use our real names but don’t auth through FB are put in the wrong bucket and count erroneously as pseudonyms.  Are ‘likes’ really a good proxy for quality? Likes = High quality comment or Likes = Preaching to the converted?

    1. falicon

      Likes, stars, and any other rating system can mean a million different things to a million different people…I know people that use the stars in iTunes to group songs together…they don’t care about the rating, they just use it as a hack for playlist options (play me all the 1 star songs — doesn’t mean they don’t like the songs, just was a way to group)…I know people that use facebook likes as a way to bookmark things (because the likes get put into their facebook stream)…I guess you could say that’s an endorsement, but it could also just be a way to remember you want to actually read something later…Give people a system and they’ll find ways to hack and evolve it for their own needs…because their world is about them and they don’t really care what or how you ‘intended’ the system to be used… 🙂

      1. LIAD

        Agreed. What does that mean for the research findings?

        1. falicon

          Good question…prob. just means that the groups don’t nec. mean what the research thinks they mean…though I think the overall data and conclusions are probably still pretty good (ie. the findings agree with my gut). 🙂

          1. Mark Essel

            Uh oh, experimenter’s bias 😉

    2. whitneymcn

      I have the same hesitation about the bucketing for this: the fact that @fredwilson:twitter  himself would be counted as using a pseudonym nudges me into “I do not think that word means what you think it means” territory. To be fair, though, I can’t think of an approach to this sort of bucketing that would necessarily be more reliable.On the likes question, since the results are presented as “the most important” commenters for a community rather than the “best” commenters in some abstract sense, I think that likes are a pretty effective signal. Whether or not one thinks a particular community is worthwhile as a forum for discussion and debate feels like a separate question.

      1. fredwilson

        @fredwilson is a pseudonymat least i think of it that way

        1. whitneymcn

          I can see @fredwilson:twitter feeling (and being) different from Fred Wilson speaking “ex cathedra,” or as a a particular aspect of yourself, but I think that’s not quite the same as it being a pseudonym.In the naming and presentation of the account you’re being very clear and open about the fact that @fredwilson:disqus is the Fred Wilson that’s a VC in NYC, and to me that’s the key. 

        2. Laura Yecies

          I guess it depends on what you mean by pseudonym.  We know who @fredwilson is but that is not the case with many pseudonyms which are created to conceal identity. 

    3. Rohan

      Oh is that so? The numbers begin to make sense now.

  5. RichardF

    I saw this yesterday, it’s not a surprise as you say.  Although I’d be interested to know what they classify as a pseudonym.  Would they classify my username for instance, which doesn’t include my full name.Also would be interesting to know the ratio of troll comments to anonymous users ratio.

  6. John Best

    I’m the de-facto user of the startup’s corporate Twitter (and thereby Disqus etc. account). I like to think that people will interact with a brand as much as a person if they can see that that interaction has value. All of the founders have their own accounts, but it just feels like at this stage, putting the business name out there, and establishing some values for it will be more useful than a real name attached to the business. I know down the line (fingers crossed) as the business grows, I’ll have to look at that again. I’ve met people in real life I’ve known by pseudonyms for years prior. The fact I didn’t know their real name impacted my impression of them far less than their conduct online. I suppose we’re at a point where internet anonymity, and a need for unique usernames allows us to choose an *identity*. Forum names, gamer tags, Twitter handles, all are chosen by the user to embody some aspect of themselves they want to promote. I’m not surprised that those accounts are more likely to be active.I’m going to pass on the whole issue of trolling etc. because I don’t think it particularly relevant here, but I’ll acknowledge its a tangental issue.

  7. jason wright

    At one point I wondered if FAKE GRIMLOCK was Fred in disguise.

    1. Kasi Viswanathan Agilandam

      U sure now? 🙂

      1. jason wright

        No 🙂

    2. fredwilson


      1. jason wright

        Upper case. That’s a familiar footprint, a dyNO print.

    3. aminTorres

      ha! it can’t be Fred but I am sure Fred knows who he is… maybe not 🙂 btw, there is a Quora question dedicated to this which I thought it was hilarious the other they I ran into it:…

      1. fredwilson

        i don’t know who he is

  8. ShanaC

    I think we need to define a pseudonym better.I question the data because it is based on facebook logins as real names.  That may not be technically correct.Further, what about the transfer from pseudoanonymous to real names (that happened to me, I recall)

  9. Kasi Viswanathan Agilandam

    It is interesting to note … Whether psuedo, ano or real … the % of bad remains the same. After all there are not many bad people around the world :-)It very well correlates with my psychologist friend’s theory … he always tells…. the world is working because 90% are not bad.

  10. jason wright

    Real identity paradigm could not be enforced down our throats. Do I have to show my passport to get a user name?

    1. Carl J. Mistlebauer

      Nope your real identity, at least to the worldwide web, is determined by your cell phone number and your email address.Of course you have to have multiple email addresses and obviously now, cell phone numbers (in case your primary phone falls in a fish pond).  

      1. Miljenko Hatlak

        What percent of accounts at Facebook, Google, and other “identity services” are created for this purpose?

  11. Jan Schultink

    There is a glitch in the data visualization: 61% and 51% seem equally large.

  12. Miljenko Hatlak

    One such important features as identifying yourself  to some web service, forum or website is usually technically solved by most unintelligent solutions. Throughout web history, we all witnessed change of input/ registration forms. Years ago, in web prehistory, we saw entry forms that were asking to much detail, what changed to minimalistic username/ pass fields. As a result of that, we are in constant confusion what to enter, since identification requires unique names, and we never know what characters are allowed and how your profile will be handled later on. And in reality any identification process is boring procedure whether it is site/service registration or customs checkout. Furthermore, we are usually asked to provide our email to handle identification process (hope someday someone will realize this through simple twit out).

  13. aminTorres

    I love this.Whatever the reason is: you just don’t want to have a public internet life.You dont want/care/have time for the attention or the responsibility… whatever really.The important thing is that as long as the comments and thoughts are shared it is far better than not shared.  And, I think because of this detachment people tend to be more honest less reserved which one can argue that this may lead to more substantial opinions.Do you use a pseudo name ever Fred? just curious.  

    1. fredwilson

      I use @fredwilson and my avatar everywhere. I think of it as my internet brand and it is a pseudonym in my mind

      1. falicon

        That is how I use @falicon as well.

      2. Carl J. Mistlebauer

        So, does that mean your “internet brand” is just a figment of your imagination?I have never met you, but based upon your comments and those of others, you seem like a nice guy, a compassionate and curious person (maybe a little weak on your sense of humor, but we can work on that), so is that your internet brand, a pseudonym of your mind, or is it who you really are?I like FAKE GRIMLOCK and Rohan, does that mean I would not like them if I found out that while using their real names they are totally different?Its kind of like the Clark Kent and Superman issue….one with glasses and the other as a superhero…were they truly different personalities?

        1. falicon

          It’s certainly a lens into a larger personality…I’m sure there are things that Fred thinks, does, and says that aren’t put into his @fredwilson:disqus  brand (though he’s prob. more open with his brand than most)…just as I’m sure there are things he does or talks about online that he probably doesn’t spend much energy on offline…I don’t think it’s as much about a Clark Kent/Superman situation as it is about context and situations…FG and Rohan are interesting people online because you are engaging with them around topics/content you find interesting…run into them in real life at the DMV and you are likely not to find them very interesting or have something in common to talk about off the bat…doesn’t mean they are actually different, just means the situation to engage is different, and so you get a different part of the personality…

          1. panterosa,

            I’m definitely the person to hang out with at the DMV, or other boring place. 

          2. FAKE GRIMLOCK


          3. fredwilson

            yes, exactly

        2. FAKE GRIMLOCK


        3. fredwilson

          i think it is an expression of my imagination, not a figment

          1. Carl J. Mistlebauer

            Its actually your imagination attempting to ESCAPE from the confines of your MBA! 🙂

      3. John Best

        Exactly, its an identity (admittedly fairly close in name to your real life one) you’ve constructed *specifically* for the purpose of interacting online.

        1. fredwilson


        2. kenberger

          great, but it’s not a *pseudonym*, going by the definition in the article this post is about. It defines it as a fictitious name.  We all have our online *identity*, but fredwilson and kenberger with their respective likeness portraits attached should still be in the Real Names bucket for the purposes of this report (whose flaws are being exposed by this conversation).

      4. William Mougayar

        Hold that thought Fred. If @fredwilson:disqus is a pseudonym, then isn’t that a stretch of the use of that word? I never thought of wmougayar or wmoug as a pseudonym as such. It is your brand, as is AVC your brand, and USV is part of your brand. I see it as your social identity, but not a pseudonym. 

        1. fredwilson

          it’s not a stretch. its a rethinking of the word and what it really means

      5. kenberger

        The linked-to story with which you started this post starts with a clear definition of “pseudonym”. sorry Fred– yours is not one. No way uhuh fuhget it. You’ve simply modernized your real name with internet speak by removing the space and using lower case. Me too. Nothing pseudo there, and it seems clear that the referenced report is not talking about characters like you and me in this definition.I’d say the same goes for your avatar. It’s a bit of an abstract painting, not a photo, but it still is a fairly true representation of the real physical Fred Wilson.You’ve built a brand around this slightly tweaked “handle”, but I’d submit it’s merely a non-aberrative alias, far from a pseudonym.

        1. fredwilson

          then the same is true of JLM and LEwhere do you draw the line?

          1. kenberger

            I think the report is pretty, but flawed; it should allow a couple more lines to be drawn. Anon, real identity, and pseudonym don’t quite cover this universe.The big concept missing is pseudo-IDENTITY. The article states “a pseudonym is a fictitious name…” Using one’s initials is not quite fictitious, it’s an obfuscation that makes one partially anonymous, partially alternate identity, and partially real name (because it’s derivative of one’s real name).fredwilson and kenberger are pretty damn transparent and clear cut, even if we both think of them as avatar-like.

        2. fredwilson

          to me handles and pseudonyms are the same thing. at least that is what i’m thinking about when i use the word

  14. awaldstein

    I’m not blown away by this data.GRIMLOCK is a pseudonym; JLM isn’t really. Both are persistent identities but the difference is that GRIMLOCK has a mask; JLM is a transparent personality who has a nickname.That distinction I get and of course, the way I can relate to one over the other is clear. Both are useful and add to the community in a big way but if I can’t connect in the real world with you in anyway, the lines are drawn.They contribute to community but I don’t see them as a driver.Identities rule, names are simply handles.What am I missing?

    1. William Mougayar

      You’re correct that where the lines are drawn is a determining factor. For e.g. will @fakegrimlock:disqus  meet me next time I’m in Washington? And if so, will he make me swear to not say anything about him that others are dying to know about? Will he come with a dyno mask on his head, or will he reveal his real identity because we’re friends? And does that matter if we want to take our relationship to the next level? That could be a good test. 

      1. Ed Freyfogle

        fakegrimlock meets you all the time already. You just aren’t aware.Open you eyes and you shall see.

        1. William Mougayar

          Can you elaborate? You lost me. – posted via

          1. Ed Freyfogle

            @FAKEGRIMLOCK:twitter  isn’t one person. It’s a state of mind.

          2. FAKE GRIMLOCK


          3. Jason

            This time you’ve gone too far!

          4. Brandon Marker

            @FakeGrimlock:disqus might have made one of his funniest comments ever on this one.

      2. falicon

        No one knows who he is because everyone that meets him gets eaten…it’s nothing personal, it’s just the nature of the big metal dinosaur…

      3. Mark Essel

        I’d love to see FG with a dino mask.Of course you’ve revealed that he’s in Washington!I enjoy the hell out of the FG persona as a communication channel, I appreciate the person behind the pseudonym much more.

      4. Kasi Viswanathan Agilandam

        U r getting too real. Fakegrimlock says he is fake. Should i say FakeAgilandamKasiViswanathan???

      5. LE

        @FakeGrimlock:disqus is a fantasy. To me knowing the real person would kill part of the appeal. In the early 90’s I visited Apple in Cupertino. I expected this big deal campus like IBM headquarters in Armonk NY or GM in Detroit. What I got was a bunch of small building that could have been in my town.@JLM:disqus reminds me of “Dallas” but with a patriotic slant. Someone who is conservative and was in the military.  A real Texan. Not a vegan. No prius. Lives large.…If he looked like Steven Hawking (he doesn’t) it would be a total buzz kill.

        1. JLM

          Vegetarian, Indian word for “bad hunter”Vegan = really bad hunterShould I let you know that I was the President of the Austin Musical Theater?Of course, I think of myself as a financial guy and theatrical producer rather than a “talent” guy but it was one of the most interesting things I have ever done.It was an Actors Equity company and how I loved meeting those long legged dancers — the girls, mind you.

      6. Donna Brewington White

        “For e.g. will @FakeGrimlock:disqus   meet me next time I’m in Washington?”   My guess is that he will.  Especially if you’re buying.Okay back to work for me.  You people are too engaging.

      1. awaldstein

        I’m being a simpleton I guess.if all we are talking about is whether you need a visible and transparent fingerprint (aka real name) or simply a persistent identity (whatever you call it) to enable community, then really this discussion is kinda moot. Certainly levels of transparency engender levels of connection but this is not new news. Anyone who ran a BBS or chat room generations ago knows this.I need to refire up the La Pavoni then and have another double expresso.

        1. leigh

          agree that persistent identity is what enables community. Someone else said trust and consistency which i agree with. Knowing someone’s real name can be rather irrelevant in some cases — and some pple are better at making you believe in their creation (real or not) then others. – posted via

        2. fredwilson

          that is exactly right

    2. Donna Brewington White

      I’m with you on the identity thing.I don’t mind a nickname, but I’m not excited about connecting on any significant level with someone shrouded in mystery or using a pseudonym to game us.  The people with pseudonyms that I interact with online are consistent personalities and I experience some degree of authenticity and transparency from them.  When I interact with them outside of the pseudonym, they are the same people — okay, except that @FakeGrimlock:disqus is capable of complete sentences — at least in the minimal interaction I had with him (but the same brain) — and those using pseudonyms are just as consistent with their AVC persona as those using their real names.I’m just not sure the @Disqus:disqus research can be considered conclusive because there are so many factors involved in the question of “What’s in a name?” when this is translated into the context of online interaction and in particular online community. I guess I could have just liked your comment and left it at that but you are such a thought sparker.

      1. awaldstein

        Really well said Donna. 

        1. Donna Brewington White

          You mean the part about you being a thought sparker? 😉

          1. awaldstein

            Yes…I’ll buy the drinks next time you are in NY ;))Thinking Terroir in the E. Village. Original natural wine bar. Tiny. Wine Geeky. And they sell half glasses to encourage tasting

          2. Donna Brewington White

            I am SO there!But, isn’t it my turn to buy?  Then again, who’s counting?  😉

      2. Matt Straz

        Agree. I am not a fan of operating under a shroud.Couple of exceptions are if someone is under the boot of a repressive govt or for personal health reasons. Also, if it’s someone under the age of 18.Otherwise just be yourself.

        1. Donna Brewington White

          I think I am almost there Matt and it is making online engagement so much more meaningful and enjoyable.Although, I don’t necessarily think that everyone using a pseudonym is operating under a shroud.  And certainly, the exceptions you mention are wise.

        2. William Mougayar

          But the pseudonym usage sometimes also allows the person to be themselves. I’m sure there’s a 3rd bucket which covers people working for large corporations that aren’t allowed to comment openly, so they unleash themselves with a pseudonym. – posted via

          1. Matt Straz

            I loathe the comments section in Business Insider specifically because of the people who use the cloak of anonymity to say terrible things about the company they work for or compete against.It’s so cowardly.

          2. Mark Essel

            Free speech and anonymity are tied together though Matt. Some cannot speak without fear of retribution that will drastically affect their lives/livelihoods.

          3. William Mougayar

            Sometimes you don’t have a choice, for the same reasons that Mark mentioned.- posted via

          4. Matt Straz

            There’s a difference between free speech and the defamation and just plain mean spiritedness that anonymity has led to.Many publisher have recognized this and moved away from anonymous comment boards, despite the drop in page views.



        1. Douglas Crets

          Wait you don’t want “All In” at your poker night? What kind of poker player are you, anyway? 

        2. Mark Essel

          C’mon FG, everyone know’s you play Magic the Gathering.For a commander game night, count me in.

          1. FAKE GRIMLOCK


    3. Aaron Klein

      I think the point is that the “real identity” people in social media tend to lump anonymous and pseudonymous comments together as having the same set of problems.I do have to question the underlying data here – it seemed like you had to use Facebook to be counted as a “real name” and I assure you, if I was going to choose a pseudonym, I can be counted on for more creativity than “Aaron Klein.” :)But if the data is right, or close to accurate – it proves the point many have been making. Don’t assume psuedo identities create the problems that anonymous ones have (especially when you look at the comment problems you’ve had on political blogs with anonymous comments).

      1. awaldstein

        No question. I agree with you that pseudo and anonymous are different. But that pseudo is somehow a more quality contributor than real names raises more questions than provides answers to me.

        1. Aaron Klein

          Not to be too hard on my good friends at Disqus, but I think it points to a data methodology problem.I’d have gotten a database of common first names and tried to shift those to the “real names” column. There’s probably other low hanging fruit like that.But hey, we’re discussing it. So I guess it worked. 🙂

          1. awaldstein

            No one loves and pokes Disqus harder than this community I bet. And I would bet that there are very very few blog communities on a global basis that have his participation.Would love to see that data, to really understand the baselines on dynamics and of course (broken record rant here) how to search by topic by dynamic conversation. 

        2. panterosa,

          Just depends on who’s talking, and about what, as to quality. Very elusive and subject to measure – all together too quantum.Compared to anonymous, at least with a pseudonym you have a person, or persona. I would think those two are real distinctions. For example, panterosa speaks with my same voice. I’d expect the same for JLM and LE. I met you in November and perhaps you can vouch for me sounding the same, as could other AVCer’s I’ve met. GRIMLOCK on the other hand might have more persona going, thru which his voice is channeled. SInce I have done artwork in this way as well, as a pseudonymic persona, I can tell you that bandwidth did broadcast views I am aligned with, but in a very vertical way, very focused. The other dimensions of that persona were vague, almost intentionally because it would simply have been to mundane to include those other details.I won’t bother asking FG whether he agrees or not, because in his case it’d spoil the mystique. 

          1. awaldstein

            True…But I associate (sorry, can’t help myself) the name and avatar with the linkage. I certainly remember and think of you when we converse here. I do wish that you were more linked up to your avatar so when I clicked on it I was hotlinked to your sites and a deeper sense of the person.The deeper that link the better for truer connections. 

          2. panterosa,

            I have sent you direct info/links to your email. Please confirm, via reply to that mail, your receipt, and we take it from there, off AVC, where I am still just a pink cartoon cat with a subversive sense of humor.With those I have met offline, I am comfortable to pursue the “truer connection” your reference. And for now on AVC in the general sense, I prefer to remain pseudonymic.- posted via

          3. ShanaC

            So maybe we’re looking for personas online as a guaranteor of quality.I’m still not totally sure how behavior online reflects into real life, and vis versa. On some level we’re all personas.  It isn’t like you can see my body language here.

          4. panterosa,

            Not exactly sure that the personas you reference are enough of one type or style as to guarantee anything.But we met in person so we have the luxury of a depth of conversation which perhaps skews the idea in general.It’s all a crapshoot. You just need to learn to shoot craps.- posted via

          5. ShanaC

            With loaded dice, and musical numbers in grand broadway tradition :0- posted via

        3. Pete Griffiths

          I think there is an important difference that common sense would suggest does result in different behavior and which may well result in different ‘quality’ (whoever you may measure that).If I have a persistent identity I have an interest in protecting it.  If that identity is in any sense ‘ranked’ this interest is enhanced.  This is true even if it is a pseudo identity.  Such an identity represents an investment which is to be protected.True anonymity, by way of contrast, has no such investment.  This suggests the likelihood of different behavior.  True anonymity facilitates trolling.  What would you care if you have nothing to protect. Pseudo anonymity inhibits trolling.  I still have a reputation to protect.So even though ‘that pseudo is somehow a more quality contributor than real names raises more questions…’ is a fair comment if you focus on how on earth to define and measure quality, casual empiricism would suggest that behavior is indeed different, and different in a way that matters to a community.

          1. awaldstein

            Agree. Diversity no matter how you define it adds to the fabric and value of community. That community value and quality is really the only one that matters. 

      2. Donna Brewington White

        “Don’t assume psuedo identities create the problems that anonymous ones have…”Good distinction.

    4. Pete Griffiths

      IMHO you are missing the fact that there are important social communities (facebook, google+) who INSIST on name = real identity.

      1. awaldstein

        Hmmm..I’m a very strong user of FB, less of G+ and I understand your point.But…the discussion is not whether FB or G+ is right for their approach but whether community on the open web is possibly with a layered identity structure. And my belief is that certainly from a community perspective, a layered approach works.I choose to have an overt real name, transparent identity poise. And I’ll agree that the depth of relationship and friendship is correlative to the amount of transparency. But as a fabric of community dynamics, all the levels contribute and work.IMO only of course. 

        1. Pete Griffiths

          I agree.I think we can have multiple approaches and I don’t see that they are necessarily incompatible when viewed across the whole ‘open web.’  And I do feel that ‘all the levels contribute.’  The problem I have is when people argue that ‘real’ identity is the only way to go.  So I applaud Disqus for reminding us that alternatives have real value.

          1. awaldstein

            Agree…I must admit that I used to feel more strongly about the real name necessity.  My views have changed, a lot because of the interactions here at over time.I’m a big Disqus fan so glad as well to see them step up as well. Could have been a more focused discussion but over 200 comments obviously shows that it struck a chord. 

      2. fredwilson

        i don’t think either are communitiestwitter is a communityfacebook and google+ are utilities

        1. Mark Essel

          To me Twitter is a utility, tightly or loosely coupled groups that communicate over it are communities.The group of people (which is dynamic) define the community, not the medium with which they choose to connect.

          1. Dale Allyn

            I was thinking the same thing, Mark. I feel no sense of community with Twitter itself, although I do see groups creating some sense of community. This is or can be similar on FB and G+. I saw a more cohesive bond at flickr, with close communities in groups or series of groups, and a looser community overall, but still a noticeable flickr-wide connection. I’m sure that this is because there is a common shared theme throughout (photos / photography). 

        2. Pete Griffiths

          Hmmm.Not at all sure about that one. I respect your opinions so I’ll give it some thought but my instincts are rather the opposite.  Twitter is a push based distribution platform.  It certainly can and does support a networks of active users but who form clusters that could be called communities though so in this sense I agree.What I am finding harder to get my head around is why you feel that facebook and google+ aren’t communities.  It is pretty evident to me that there is a great deal of communication between groups on both platforms of a kind that looks and feels like a community to me.The word ‘utility’ is useful for indicating something’s ubiquity but I am not sure it is helpful to contrast something with a community.  Not in the case of social networks anyway.  

  15. Rohan

    I’d actually like to see the definition of Pseudonym. What/who is a pseudonym? The names you mentioned above are pseudonyms of different kinds.Grimlock, Profoky are very different from JLM, LE, Kid Mercury etc.

  16. markslater

    one of the fundamental principals of our service IS annonymity. In fact – we see pseudonyms as replacing phone numbers. we love pseudonyms.

    1. awaldstein

      Do you love pseudonyms or do you just realize, I think, correctly that identities can be pseudonyms as well?Who cares what the name is as long as you are interacting or transacting with an identity? I don’t want to sink into semantics and fear I may be.

      1. markslater

        for us arnold – in one particular use case – people are asking questions about a career decision and do not want the “hard sell” before they have actually decided to proceed – s they can ask questions under a pseudonym secure in the thought that they will not be identified, or hounded after the chat session.So we enable behavior that is sought but maybe avoided today by 1. the inefficiency of the phone, and 2 the need for identification.there are lots of uses cases out there where we are validating this right now.

    2. jason wright

      Enum will do that.

  17. William Mougayar

    Good to see GRIMLOCK featured in that graphic. But who is the pirate and wonder woman behind them? And loving the bags over the Anonymous folks. Was that a “douche bag” innuendo?

    1. Carl J. Mistlebauer

      Grimlock made the graphics, so who is the pirate and who is the superwomen?I think they should have just used question marks for the anonymous folks and then had another category for quality of comments from people with ugly avatars, that they could have shown with bags over their heads….Is there a relationship between avatars and comment quality?

      1. falicon

        Good point.  I would bet there actually is a connection between avatars and comment quality (as well as avatars and comment engagement)…would love to see research to back that up as well. 🙂



    2. panterosa,

      Bag on head = Douche bagI almost spit my tea out William.I’m sort of wondering about Wonder Woman…funny choice to put in there.

    3. Mark Essel

      Oh I missed that when I first read it (on Google+). Looks like Wonder woman, and a Pirate?

  18. William Mougayar

    The key take-away for me here is that Pseudonyms are better than Anonymous. So it’s about Pseudo vs. Anon, and vs. Facebook, to some extent. But I can’t wrap my head around the survey data that only 4% of commenters use their real identity. That seems like a pretty low number.  

    1. Rob Hunter

      It does make you wonder if the real-name metric is working properly.  I’m not convinced that I sign in with Facebook (I use my Twitter account, may have never shared FB with Disqus).  Even if I have I could have just as easily not.  By the same token, I think it’s 4% comment volume – FAKEGRIMLOCK definitely posts more than 5x what I do, which was the statement that they made.  I don’t think they said anything about the count of real commenters versus pseudonym commenters.Pseudo vs. Anon is definitely the takeaway, but in both I wonder if there’s a way for a community to encourage one over another (for another forum/commenting company, the message is clear – allow pseudonyms).

      1. William Mougayar

        I think the main ones that don’t allow pseudonyms are Facebook and Google+. If you step outside the AVC community for eg. into the NY Times or HuffPo commentspheres, you will see lots of pseudo names.E.g. 1400 Comments to “Patrick Swazye’s wife takes her ring off” http://www.huffingtonpost.c… It boggled my mind that 1400 people were interested in commenting on that. – posted via

  19. William Mougayar

    Lol…you’re on a roll Carl, or is it Carl really? 

    1. Mark Essel

      As far as pseudonyms go, I’ve never shown you my license William.We base our identity on trust and consistency.

      1. William Mougayar

        Me neither…maybe my real name is not William. But does it matter? – posted via

      2. leigh

        “base identity on trust and consistency” and that is exactly the point (regardless of research methodology)

        1. William Mougayar


  20. falicon

    Interesting data for sure…I’m working on another little hack that I’m just about to release around this topic so I’ve been thinking about it a lot as well…the thing that I find especially interesting is, I believe, this data is all based on ‘logged in’ scenarios…so the ‘service’ itself has some concept of identity for the user, it’s just the other users of the system that may or may not know more about each other…It would also be interesting to learn more about how they quantify ‘positive’, ‘negative’, and ‘neutral’…those are actually very subjective terms…

    1. Alan Minor

      It’s elucidated here. (Right under “PSEUDONYMS ARE THE MOST VALUABLE CONTRIBUTORS TO COMMUNITIES BECAUSE THEY CONTRIBUTE THE MOST HIGH QUANTITY AND QUALITY OF COMMENTS.”)Obviously, there are some potential issues with qualifying a signal as positive based on the number of replies it’s received.

      1. falicon

        Yeah – I still question the terms…also I don’t see a definition of ‘neutral’…so I guess we are to assume it means  comments that have no likes or replies…if that’s the case, I would suspect the number on neutral should be MUCH MUCH higher across the board (do you really believe that comments across the web – through disqus – when left by someone with a pseudonym only get 28% neutral reaction? Even in today’s connected world, evoking a reaction is harder than being ignored…I still suspect the majority of comments are actually ignored [if for no other reason than for time/volume factors])

        1. Alan Minor

          I agree.

        2. Donna Brewington White

          Ha — speaking of pseudonyms — this whole time I have been “seeing” Kevin when I see your comments — just now noticed that you’ve actually changed your handle.  Although I know this has been your Twitter handle all along.   

          1. falicon

            Heh…yeah I changed it once I actually noticed it yesterday…online I generally go with Falicon (or falicon33 in the rare event I can’t get falicon for some reason)…but offline I usually go by Kevin (the books I’ve written and contributed to all use my real name as well).Beyond simple brand, I have the incentive of having a really common name…so without the falicon connection, it would be almost impossible to find me online (I’m an active online person but if you google Kevin Marshall — before they did all the personalization stuff — I barely make the top results)

          2. Donna Brewington White

            Ah, so you just now changed it and I WAS seeing Kevin until now.  Okay, not losing my mind.Try having a name like Donna White to get lost on the internet.  The “Brewington” helps to distinguish but seems weighty for online interaction.  I just haven’t decided what to do with it yet. 

    2. Mark Essel

      Yeah. it’s an interesting social challenge with limited information.Just caught a great video on Hadoop’s map/reduce this morning (home sick). Thought it was a reasonably good fit for KAI as you grow, especially for my daily recommendations. I may have suggested this before. If so, please disregard.The gentleman who gave the talk uses a few phrases a little too much for my liking, but he does a great job of describing the system and why it’s designed the way it is.

      1. sigmaalgebra

        He can’t settle on “elements” or “records” in a ‘list’.  Sets consist of ‘elements’ and have no sense of order.  Since lists do have a sense of order, really should not say that a list consists of ‘elements’.  ‘Records’ is a bit too close to some old usage, say, with magnetic tape.  So I would prefer to say that a list consists of ‘components’. I will save time and not look up the terminology Knuth used to define lists.Why make this point? When I see technical material that drifts away from precise definitions used consistently, I get ready for some wasted effort in making clear just what the heck the content is. So, one of my first filters on technical content is good use of definitions.Of course, we are talking about ‘big data’. So, notice the new Swiss Army knife with one blade with a chip with a solid state storage device with 1 TB. A package about the size of a 3.5″ hard disk might be able to have 160 TB?Soon a smart phone could have in it nearly all the movies made by Hollywood since 1912 or so? Let’s see: Two movies a week, 100 movies a year, 100 years, 10,000 movies, 2 GB per movie, 20 TB for the whole collection, 20 of those little chips in a smart phone?Rotating magnetic disk is now up to 4 TB per drive, but it appears that solid state drives (SSDs) will scream past that capacity quickly!Main memory is now down to $10 for 4 GB (Tiger Direct). So, soon $10,000 would pay for 4 TB of main memory? Glad we’ve gone to 64 bit addressing!Intel can put 1000 cores on a processor and have them work together effectively.Recently one wavelength on one optical fiber on one cable across the Atlantic carried over 1 Tbps. The bandwidth door for the Internet to “go international” (COB, FedEx, but not about the Internet) in a big way is wide open.SOPA-PIPA supporters: Wake up. F’get about SOPA-PIPA and let the US continue to lead in the Internet for the world! Why? One reason is, for some months already Google has been getting more revenue from outside the US than inside:…Net: We’re in line for a lot more in ‘big data’. And at anything like current CPM ad rates, the time a server farm needs to generate enough cash to double the size of the farm is falling from months to weeks to days — maybe hours? “Amazing times we’re living in” (‘Master and Commander’).

  21. Rob Hunter

    Policing things like real names is hard.  It gets into why there are lots of social behaviors that are hard to police – enforcement isn’t easy, but I think it’s determining the rules (and then being flexible with them as needed) is way tougher.You’ve got some guys in a room, and one says “well, we don’t want trolls or spam, trolls and spam don’t generally come from real-name users, everyone needs a real name”.  None of these guys have ever been trolls or spammers, and are painting with broad strokes across complicated issues – with the best of intentions.I think you get the same issues in politics – I read recently that when intelligently spoken to about Internet piracy legislation (this was pre-SOPA) pro-legislation Congress members agreed with the disenting opinion – but they’d never heard the disenting opinion before (they’d perceived that there were people trying to solve the copyright problems and there were pirates, totally unaware of any other party).You’ll always have trouble when you don’t understand the other side or emergent behavior, and emergent behavior is REALLY hard.

  22. jmcaddell

    Oh, no! I need to think of a pseudonym, quick! Any suggestions?:)

  23. Tom Labus

    Does having a pseudonym release the real person from fear of speaking directly?It’s pretty interesting what makes people comfortable enough to let it rip.

  24. Alex L

    I wish disqus would have talked about their install base demographic. My expectation is most of their blogs are tech related, which I believe makes for skewed results. I wrote a longer response here:

    1. fredwilson

      actually they have a ton of politics blogs, mom blogs, videogame blogs. i think tech blogs is a smaller part of their base than you might think

  25. jason wright

    Generally real names (given names) are chosen for us by others (parents et.c.)?Pseudonyms we choose ourselves (normally) , and they are selected from a much wider range of choices. The reasons for choosing pseudonyms can be varied.

  26. jason wright

    Time for clarification in this discussion to avoid confusion.Are we defining ‘real’ identity as our legally recognized name? Forget any distinction for a moment between on and off line.Dictionary definition of ‘pseud’ – sham, false, spurious, deceptively resembling, pretentious, et.c.Perhaps the motive for choosing a pseudonym is a more interesting one.

  27. JamesHRH

    I think the data also shows that people who create a persistent internet presence – Fred’s internet brand so to speak – just enjoy the activity more, find it more valuable and do it more.’The real name’ crowd is the issue of import here.

  28. andyswan

    Well according to this I am a pseudonym as well….which makes it pretty flawed in my eyes.We already knew facebook people are lame.

    1. Mark Essel

      Harsh!”We already knew facebook people are lame.”With the TechCrunch comment switch this became evident.

    2. ShanaC

      Hey.  Facebook is where my friends who are not here are at.  And I am ok with that 🙂

  29. christopolis

    If they worked for Ben Franklin they work for me 🙂

  30. Richard

    This seems like a weak attempt by disqus at data analysis. Both in a statistical sense (how reasonably sure can you be that the means of all three categories are not equal) and in a data reliability sense (how reasonably sure can you be that the data categories are accurate). Disqus could release the data set. I’d be happy to analyze further.

  31. Elia Freedman

    I don’t really understand why there is such an argument around this issue. My name is the name my parents gave me. My pseudo name, if I choose to use one, is the identity I gave myself. If I know you as your pseudo name or your name, it really doesn’t matter as long as my interactions with you are consistent or I can connect the real name with pseudo name for my interactions (which to me is no different than recognizing you via a phone number or email address instead of a name.)

  32. Mark Essel

    Here’s a first I caught it first on Disqus’ blog, then on (my default morning news).Great post by Daniel.

  33. Dale Allyn

    Interesting conversation, but the “research” is a bit flawed. I don’t doubt that the hypothesis will hold, but there needs to be a clearer definition of “pseudonym”. JLM and LE are known to the community (we know their real names and businesses). Others using a “handle” which would count as a “pseudonym” are directly traceable to their own real identity via their Twitter profile (linking to their company for example), or their blog, etc. Need to dig deeper using a better research model if the data are to be meaningful beyond convo fuel. 🙂 Kudos for digging in to the topic though, Disqus. 

  34. abhic

    This is a multi-faceted and mostly subjective debate. I use @abhic:disqus  as my internet persona and on most sites its my real name. Is it my ‘identity’ or my pseudonym? No clear agreement there. Without getting into the semantics of “What is identity?”, I believe that I have a couple of web identities that I use actively and the difference really is the community, not me. On, I am @abhic:twitter , on 4Chan I am _________. However, on both communities I am still all me.In my mind, this is most valuable to understand when you are building something anew. If you want to build the next Tumblr/Pinterest, ‘pseudonyms’ like the way Disqus understands it need to be allowed. If you are chasing LinkedIn/Google+, then ‘identities’ should be mandatory. But whatever you build, people will find ways to co-exist and sometimes in an almost schizophrenic way.

  35. MartinEdic

    I don’t see how a positive comment is a ‘better’ comment. I do not think these are legitimate conclusions to draw from this survey. Sentiment analysis is very tricky stuff, especially when it seems likely that anonymity would encourage a wider range of expression, both positive and negative.

    1. Cam MacRae

      Yes, you have to be very careful in your analysis: A plethora of Likes might reflect nothing more than groupthink.

  36. MartinEdic

    As I think this through and read through the comments I have to say that this may be the worst post Fred has done in quite awhile. It simply doesn’t hold up to any kind of scrutiny and seems written to support a favored hypothesis rather than an actual conclusion.

    1. Dan Bowen

      I have to agree, this “study” by Disqus would have gotten an F in the most basic “Lying with Statistics” class.  Adding any credibility by putting it on Fred’s blog seems out of place. The topic itself is incredibly personal anyway as it is individual interpretation that determines the value of any input, certainly not volume or group reaction.  After all, Obama posting on MSNBC will get great reactions, on Fox, not so much so.  So if he stays in the comfort of one crowd over the other is he more or less valuable? I look at Fake Grimlock as a great satirical release on many heated discussions…how seriously do I really take FG however is pretty limited…is his work quality?  Sure, but for me, quality humor more than much else.  I also have enjoyed some of kidmercury’s positions on the monetary situation in the US but after reading through his site, particularly the links to the 9/11 truth stuff, I limit what I view as quality from him. In the end, just because you post a bunch of stuff and that happens to be important to whatever demographic you are posting to doesn’t mean an hill of beans to the rest of the world. But then again, I’ve obviously spent far too much time trying to get my point of view out which, based on my review, doesn’t mean sh…

    2. fredwilson

      this is one of my strong held beliefs. i am simply promoting it.

  37. kenberger

    I didn’t quite get @FakeGrimlock:disqus, but saw this video and now it’s all clear:



      1. kenberger

        That video shows you could be a CTO: Chief Transformer Officer

  38. Dan Goldin

    I’d like to see the worst comments too – I’d imagine they would also come from the pseudonymed users.Pseudonyms may help you get more engagement under anonymity but they also encourage “keyboard courage” and may lead to some hateful comments.

    1. LE

      “I’d like to see the worst comments too”There aren’t downvotes. 

  39. JaredMermey

    Quantifying quality is difficult with comments. I think we’d all agree that comment quality is a function of number of comments, identification type and sentiment among other inputs. But I also think this comment thread is illustrating the difficulty in defining the variables and that each variable would only have a minimal correlation to whatever we deem as quality. Maybe it is how my mind works but I’m think of a regression that would have a super low r^2 and hard to define dummy variables…

    1. sigmaalgebra

      Note that it is possible to have random variables U, V, and W so that U and W have correlation close to 0, V and W have correlation close to zero, but in ordinary linear regression variables U and V predict W almost exactly! So, that V and W are not correlated does not mean that V is irrelevant  in predicting W!

  40. Vinay Pai

    It really upsets me that this bit of poor research has been picked up so far and wide across the Internet and repeated by people over and over again without actually examining their claims.  It upsets me not because I have some sort of axe to grind against against Disqus or pseudonyms, but because of what it says about critical thinking vs. the Internet echo chamber.Firstly, their definition of “real names” is people authenticated via Facebook. It doesn’t take more than 3 seconds of thought to see why this is deeply flawed.  I, for example am using my real name but I just signed up for Disqus, without connecting it to Facebook or anything else, and I suspect that goes for plenty of other people.Their definition of “positive” is also pretty dubious. While “likes” may be a decent proxy for quality, “number of replies” is an extremely poor one. For example, anything sort of flamebait or egregious factual error is likely to produce a wave of reaction, but I wouldn’t call that a “positive” comment.I think the only conclusion really supported by their data is that people who take the time to sign up on Disqus are likely to use Disqus more often. That doesn’t make for as exciting a headline though.PS: Now I’m really nitpicking, but who taught these people to draw graphs? In the “quality signals” chart, why is the pink bar representing 51% the same length as the green one representing 61%?

    1. LE

      “Firstly, their definition of “real names” is people authenticated via Facebook.”My understanding is that they are defining a pseudonym as “a fictitious name chosen by a commenter”.   So then @wmoug:disqus or @ccrystle:disqus is an example of a real name.I agree with what you are saying about flamebait comments but I’m not sure it applies to this blog since the forks are generally welcome here (although disqus is not presenting research conclusions about this blog but all communities they manage as you know).   I agree with the nitpick. My nitpick is that the page should be text and not a graphic (better to be spidered).I don’t think they are presenting this as good research either. In any case the 4% “real identity” number seems almost to be an error to me.

      1. Vinay Pai

        I’m fairly certain they are only treating people who have connected their Facebook accounts as real names (which would explain the suspiciously low 4% number). Under “average comments by user by identity) they say “4.7 times more likely than users identifying with Facebook”. Again, in their “Conclusions” section they say “…real names (via Facebook)”.I totally agree with you in that the AVC community is way more respectful and welcoming than average.

  41. LE

    My suggestion is that @disqus survey people in all categories to find out why they do what they do in order to gain further insight. Why they choose the identity they did. Then release that information and conclusions formally (as a press release) in order to garner publicity. I can see that having nightly news potential. By survey I mean personally reach out. Not some no effort internet survey. 

    1. fredwilson

      good suggestion

  42. Douglas Crets

    The data analysis is overlooking something very very important. I think it’s also leading us down the wrong path. VOICE is more essential for substantiating a need to comment and an ability to draw quality commenting than pseudonyms. Pseudonyms alone do not lead to higher quality content in commenting, not by a long shot. that’s like saying that a vanity license plate leads to better driving. They do allow more people to be freer in their expression, I won’t deny that. What’s really happening, I think, is that in communities are run better with or without pseudonyms.Place a comment with or with a pseudonym on a “quality” news site like The Atlantic Monthly and you will be surprised about how egregious the commenting is in its balderdash and cynicism. Place a comment here on AVC, where you have quality from the top down and from the bottom up and you see a wildly different experience. Pseudonyms mean nothing. The tone of the first place blogger is what sets the interaction. Look at Quora, a non-blog site, but a site all the same that offers named and anonymous posting. The policing — I hate to use this word, but it rings true and Quora monitors less like editors and more like police — keeps quality in check. Before anyone goes off saying that I am sponsoring dictatorial control of a site, let me say this: I am not. What I am saying is HOW you interact as the first mover on a blog site with comments determines how your audience responds to you. MG Siegler, for instance, gets very upset about how people comment on his blog, to the point that he turns off his comments. But people react that way because of how he treats them. His “voice” on his blog completely demands the kind of attention I would give him, namely, a slap in the face. In short, the pseudonym defense is a canard, though a canard with merit, because I do buy into the fact that pseudonym commenting allows some people to be more free in their expression than they would with a named system.

    1. Douglas Crets

      Sorry for the grammar,  “What’s really happening is that in some communities, they are run better because of the voice of the author, not because there exists the ability to use pseudonyms.” that’s what i really mean to say. 

      1. William Mougayar

        Very true that the quality of comments also has a lot to do with the blog owner’s participation in the threads, case in point with this very community.

    2. ShanaC

      I think you’re pointing to communities as the silent partner in all of this, no?  If you don’t behave, the people around you don’t.There seems to be some sort of network effect of behavior, maybe?

      1. Douglas Crets

        I’m not pointing to communities as a silent partner. I am saying that a psuedonym has no basis on quality. It’s a way of grouping things as evidence, but I think it overlooks how people manage communities. 

        1. ShanaC

          True, and I am asking if the why has to do with the way communities structure themselves

          1. Douglas Crets

            I’m not sure I follow you, actually. There is no one “why” communities structure themselves. I assume you mean that communities get behind an idea or search for some progress together. I think that in blogging communities form asynchronously and without symmetry because people come to the page for different reasons. To me, the biggest measure of quality is whether a community or a group of pseudonyms or non-pseudonyms solve anything, or advance information. – posted via

    3. fredwilson

      my experience moderating this community and reading every comment placed on it for eight and a half years has told me that pseudonyms have way more voice than any other type of comment

      1. Douglas Crets

        Right. I get that. You moderate, though, and you read every comment. You are a community manager — you even do the job of connecting people based on their interests. It really has little to do with whether there are psuedonyms here. If you took away your participation, and you acted like The Atlantic Monthly, or if you were a faceless beast like The New York Times, you would see a totally different behavior. – posted via

  43. Douglas Crets

    I would also like to say that I believe a catalyst for the ultimate decline of mainstream media will be because people running mainstream sites think that adding comments, which provide metrics for advertising, seems like a one-stop solution. I think they need to invest in people — community managers — who are very good at literacy, writing, humanities, and communication, and who can listen for the emotional and psychological jobs that the audience needs done. 

  44. Colm Brophy

    It’s very interesting to see how anonymity has influenced sites like 4chan. While it doubtless generates a lot of negative things such as racism and misogyny, it’s also been the source of incredible creativity because it removes the fear of failure which is so prevalent in society and hinders creativity. Particular in the online world where our failures, when associated to our names (or even pseudonyms) can stay with us for a long time. It’s definitely a double edged sword in this respect.Christopher Poole has spoken about it a bit –…

  45. James Ferguson @kWIQly

    For the purpose of this comment I am both kWIQly or James Ferguson, but in no way anonymous.Consider this -I leave a building and am hit in the back by a snowball – Surprised I turn and see some children on their balcony laughing at me. – Fair enough. I leave a building and someone walks up to me and throws a snowball in my face – I throw one back – we become equals at some level – Great.Someone wearing a balaclava walks into my community and helps out – I wish they would show me their face – to share a smile.Someone wearing a balaclava walks into your community and destroys – They are anonymous and they are cowards. I remember the feeling being woken by IRA London bombings – anonymous war is deeply disgusting.This is why military personnel in conflict zones remove sunglasses to speak to civilians – it is disarming – In fact it is civil.And this is why a public whistleblower (who stands for right against greater power) deserves our highest regard.  Because even those within a community are sometimes anonymous and remain unspoken when it matters.

  46. matthughes

    To each her own…My wife is really sharp and has a lot of clever commentary that she could (should) share.But she’s reticent (mostly concerned about privacy) so her ideas remain perpetually banked.I’ve been encouraging her to create a worthy pseudonym and hit the Internet ground running.

    1. William Mougayar

      That’s a good and typical use case for it. I think political sites probably have a high number of pseudos because some users are afraid to reveal their political preferences in public, but they will be more open and honest if they can protect their real identity. 

  47. Ted Rheingold

    I really like their goal in this research and think there’s a lot to pull out that is valuable, even though there are tons of great crituqes below.My critique is that since this was just studying comments, the title shouldn’t have expanded the coverage to ‘Drives Community’. While comment participation is a great example of community it’s only one of dozens of online community patterns.

  48. Donna Brewington White

    I’ve tried to use a pseudonym.  But the real me keeps creeping through.  I give up.

  49. Adam Gering

    I’ve always been a big fan of pseudonymity (not to be confused with anonymity). In fact I used pseudonyms exclusively 1999-2011; my real name became invisible to Google and the Internet at large. Ironically it was a combination of factors including a USV Analyst job posting (in addition to other “networking” and fundraising) that encouraged me to drop the pseudonyms and start using my real name again. (The USV job posting suggested including links to online reputation (Twitter/Facebook/G+/etc) — although I didn’t apply, it made me re-evaluate).I think there’s an unrealized potential in expanding the use of pseudonyms, to include even payments, commerce and contracts.

  50. Otto

    Real names = Internet suburbs = less creativity= less ideas shared = boring.But anonymous trolling does suck and could bring down a good community, if that community is weak. Then again real name trolling sucks too and happens all the time on Facebook comments, but that’s a suburb for you.

    1. LE

      “Real names = Internet suburbs = less creativity= less ideas shared = boring.”I agree with that. But I also think that may correlate with age as well. Younger people appear to be less concerned with making an ass out of themselves or saying something they might regret. They are  less conscious of how what they say, and what they do, could have future negative consequences. One example: creating a paper trail that can be used against you if you ever decide to run for public office.  If you use a pseudonym you at least have plausible deniability and could claim that your online persona is a character and doesn’t represent what you really think (not saying it would work but at least it gives you a leg to stand on.)

      1. Douglas Crets

        Can’t really agree with what seems to be a biased attitude about what a suburb is compared to the city. I have lived and I have worked in suburbs and cities in many countries around the world, and I think that what this kind of bias overlooks is that at a granular level, every community has smoething of value and that value is derived from very specific and contextual issues involving that community. If you are looking for drama and excitement, and you think that’s what media, and even blogging like that at AVC is supposed to deliver, then I think you’re being shortsighted.  Just think of the inherent flaw in the logic: some of the people commenting at this “city” blog are living in a suburb. This is just really poor metaphor for how Internet communities work and live. It might be more apt to say that the Internet is like the world and that pseudonym or real name or anonymous, something is always going on behind closed doors. Some communities are better at revealing what is going on behind those doors. Some are not. But differentiating a real name system from a pseudonym system is, like I have said before, like saying that driving improves with a vanity license plate. You are ignoring what is actually happening behind the wheel, or in the mind of the person behind the wheel. 

        1. LE

          “like saying that driving improves with a vanity license plate”There is improvement in behavior if people are aware that someone knows who they are. Which is not the same as saying skill level changes. While a vanity plate might not be the best example I can assure you that a truck driver driving a rig clearly labeled with the company name and phone number vs. a white panel truck will be more careful in some ways. In general.  Just like a teenager will be more careful if a parent’s friend is around.There are also companies that have monetized driver reporting. Surely you’ve seen the “How’s my Driving” stickers.http://www.1800howsmydrivin…The claim according to the site “an insurance company study found 23% less crashes and a 52% dollar savings with a driver monitoring program in place”. (That’s the entire system of course of which the “call me” is one piece.)

          1. Douglas Crets

            But it feels to me that people are associating quality with “good behavior.” That is not exactly quality. You can have a pseudonym, a real name or anonymous and the way those are labeled will not force someone to act a certain way. That’s the point I am trying to make. A pseudonym system is not what creates a community and drives its quality. It may be a facet, but I don’t think that the research presented here comes close to making that case. i think that’s what the marketing team at Disqus wants people to think. Community and behavior in a community are independent of quality. Quality is just another layer to describing what goes on in a community. 4Chan could be argued by some to be lacking in a whole lot of quality, and types of quality, but it is a community and a strong one, and it behaves poorly in some cases. But there are perfectly good examples where people using their real names act poorly in a blogging community. n that case, how could it be argued that the presence of a real name system led to it. You wouldn’t make that argument. I don’t think you could make the reverse argument either. The psuedonym vs. real name argument is false logic meant to push another agenda. It’s interesting, but I don’t think it’s a driving force in defining WHY people act the way they do. 

          2. LE

            “I don’t think that the research presented here comes close to making that case.”No question about that. Even the highest level research is called out on methodology. Obviously this doesn’t even rise to reasonable credibility as has been pointed out by many comments.  But of course from a business perspective I say “go with it and do what you can to get publicity for your company”.  After all we aren’t talking about pharmaceutical research where there are lives at stake.”i think that’s what the marketing team at Disqus wants people to think.”Agreed.  I thought it was a smart use of a minimal amount of data. 

          3. Douglas Crets

            In a business world increasingly shaped by online reputation and authenticity in relationships, I think that moves like this will draw suspicion. It is best to keep it as real as possible. Not sure marketing gimmickery falls into the realm of authentic moves that strengthen relationships. Sure, it causes discussion, but let’s get something really rigorous going.I think that companies are going to have to hire more and more community managers who can understand people, deliver years of experience, and who understand the world in a literary, humanitarian, and human-focused way. – posted via

          4. Douglas Crets

            I just need to correct the grammar in this. It should be 23% fewer crashes.- posted via

      2. Otto

        Good points. How many people are going to run for public office though? These days there seems to be no shame with politicians anyway. I would rather vote for someone who tells the truth while using a pseudonym, rather than somone who lies using their real name.

      3. ShanaC

        Maybe it will (finally) make for more transparent public officials?  I would like that.

    2. fredwilson

      facebook is so suburbia

      1. Douglas Crets

        There’s some really sweet shit that goes down in suburbia. You should try it out man. – posted via

  51. Clay

    I run a pseudonymous community, and was explaining the concept to a friend the other day and describing the vibrant/candid/thoughtful/helpful interactions we see on the site. her reaction was “funny how removing people’s identities can make them more human.” so true.

    1. fredwilson


  52. Dmitry Shapiro

    TOTALLY agree!!!  Fred, this is exactly what I showed you that we we are doing at, and our members are having the kinds of conversations that you would NEVER find on Facebook.  While we are still relatively small, the amount of content and quality of conversations that our members are generating is awesome!    Here is Mark Suster’s take on it after our conversation – http://www.bothsidesoftheta…And the original video –…

  53. laurie kalmanson

    consistent “real” personas free the people behind them to be who they really are, and there’s the proof

  54. vruz

    A new lesson for Google, despite their attempts to match credit cards with their G+ identity turf, despite their user’s wishes.

  55. Dave W Baldwin

    Without going into the apparently flawed method used, I think it is important to strongly point out you can’t have great community without both participants.The psuedos can make big claims hiding.  The real names can put a coin next to opinion by psuedo that the real name wouldn’t publicly say.At the same time, if you have a forum consisting 100% psuedos, it would die.  That’s because it is the reals that help put life into an avatar.

  56. David Fleck

    Thank you everyone for your opinions about a topic that we all find very important with where the internet goes.  Our founder, Daniel Ha, has posted additional thoughts here: I work at Disqus and I wanted to weigh in and address some of thethings I’m seeing here. Why we did this?  We have noticed that the conversation right now in the marketplace focuses almost exclusively on real identities (e.g. Facebook) vs. anonymity.  However, we know that a lot of our community chooses to participate in a different way…via a pseudonym.  So, we had a hypothesis that this middle ground in terms of identity was not being included in the appropriate discussions but may be very important to communities.  We decided to dig into our data to see what it told us. MethodologyThere are invariably tradeoffs when doing research. One can always poke holes inany research methodology. However, we were careful to not misrepresent what our data showed us and we want to be clear about our methodology so that you have the full picture. (1) We wanted to break down user types into anonymous, real identity and pseudonyms.  So, how did we define the user types? Anonymous – guests that are completely unverified. Real identity – those that have chosen to identify with Facebook when posting because Facebook makes best efforts to enforce real names and it is tied to the Facebook profile. Pseudonym – registered users that choose whatever name they’d like to call themselves. (2) Next, we wanted to find proxies for how important (a subjective word, but one worth exploring) each identity type is to communities. We broke down importance into (a) quantity and (b) quality of comments.  Quantity is fairly self-explanatory, but quality again is a subjective word that we tried to make objective by looking at discrete data points.Things that signal high quality comments are things such as likes and replies to a comment.  For example, if the goal of a community is robust discussion and engagement amongst its members – our POV – then replies show that the original comment initiated this community conversation and engagement.Things that signal low quality comments are those that are deleted by the moderator, marked as spam, or flagged by the community. ConclusionsWhat we discovered surprised us.  Conventional wisdom in the marketplace stated that Facebook users should have the highest quality comments and also that the quantity of comments by identity type should be similar. However, we found that pseudonyms ranked higher than the other identity types on both measures…pseudonyms had the highest quality comments and also the most comments per user (and overall). There you have it.  By commenting on AVC you are already a Disqus user.  For that we thank you.  We also promise to take the learnings from this research into account as we continue to innovate on our product.

    1. Pete Griffiths

      I think this is absolutely fair.  You are not writing a doctorate, you are exploring something and have made your methodology plain.  People can quibble with it but it seems to me that the very high level point that you set out to address is indeed addressed – does ‘real’ identity rule the roost?  Apparently not.  At least, not blindingly obviously.  And that is an important result.  It may be wrong, and further research may show that it is, but it is a good first cut at an important question.  Why important?  Because social networks that insist on ‘real’ identity make unqualified statements about the benefits of real identity and without some kind of evidence those whose intuitions suggest different conclusions have nothing with which to challenge such representations.

  57. another cultural landslide

    Amen, Fred. (No, we’re not a total pseudonym – but we believe in it.)

  58. another cultural landslide

    Reading through the comments, I’m kind of taken aback at how many people are slamming Fred about this issue… and reading between the lines in these comments, one can see this is still all about Real vs. Pseud.The whole Real Name issue is about one thing, and one thing only: gathering data – making the user The Product. The more correlatable data gathered, the higher the quality of the Product. The latest search integration push at G+, and the implementation of Timeline at Facebook are further refinements of delivering this quality Product. But I’m a People. Not a Product.And a community is Not a Product. It’s People. (I know, like Soylent Green. Beat somebody to that one…)This is very important for startups & marketers to understand: do not lose your humanity in assessing your userbase – because it could come back & bite you in the ass.An example: I was on G+ very, very early in the beta; wen & I always share our online identities as one (because, well, that’s just the way we are in real life). We created an anothercultland account as well.At first, we were excited at what G+ could be; and then, three weeks later, our main account was suspended because both our names were in the account name. They wanted the account to be one name only – or even better, two separate accounts. To hell with that.We haven’t been back since then. (Strangely, the anothercultand account is still alive. Go figure. We still won’t use it.)I’m also beginning to sense a slowdown in posting at Facebook since the implementation of Timeline. While it might provide a better data pool, for a large number of users we’ve talked to, it has become a deterrent to using Facebook. (Confusion & what not. Oh, and that little privacy bugaboo.)Never, ever let a person feel like a product. No one wants to be cattle.A community works because People are allowed to be who they want to be. That builds involvement.Be glad you have giant robot monsters, panthers and aliens in your midst. Life would be awfully dull without them.kirk (& wendy), aka another cultural landslide

  59. Pete Griffiths

    Yes – it’s a very interesting piece of work.  I’m a big believer in pseudoidentity and found these results fascinating.

  60. obscurelyfamous

    There’s a really good discussion here, so thanks for that. About the data testing methodology… this data was to guide our own product decisions. Then we wanted to shine it up with some fun art. It’s definitely not an independent study and the data is Disqus-specific… that said, we’re confident in the patterns that we saw and the product will show how it’s meaningful to us. 

    1. Carl J. Mistlebauer

      So, you are doing away with registering/logging in via Facebook since Facebook folks use their real names and are really horrible at interaction and community?  You could have gone to their neighborhood and driven around and noticed that!Most communication on Facebook is via games, pleading for some sort of gift, sharing fortune cookies, or some other gimmick.  We need to create a Disqus gated community to keep the Facebook riff raff out of civilized society.In case you cannot tell, that is my personality peeking out from behind my comments and I use my name that was on my birth certificate.I think the idea that pseudonyms as some sort of “freedom” or “expression of personality” is quite a leap; kind of like believing that we are freeing the drama queen or superhero inside of all of us…The reality is disqus and its community of blogs are intent on communicating, expressing, and exploring while Facebook is basically a place for people to hang out and wait for something to happen.AVC is a bar or coffee house while Facebook is a bus stop….

      1. obscurelyfamous

        Nope, we’re not doing away with supporting Faebook logins. I believe that authentication and identification are different things! Facebook is a fine authentication method to trust (and it’s awesome at it), but I don’t love a world where I would have to express myself with the person I am on Facebook. Facebook is a fantastic utility — they’ve convinced me that it’s a canonical representation of myself and my Facebook identity has a near-transactional quality to it. But a social network isn’t a community. Where’s the personality?You’re right that your real name can have personality to it. I would never argue otherwise. But the big difference is the (perhaps subconscious) control and connotations associated with it is different from what’s associated to something with as “real” as your Facebook name.Highlighting pseudonyms, though important, is admittedly simplified. It’s more of a persona, and a persona can use the same name as the one of your birth certificate.

        1. Carl J. Mistlebauer

          Real science vs. pseudoscience; Personality is an essence and you cannot create a tool that will reveal the essence of a person.One study showed that 89% of Facebook have met and or know the people they are friends with in real life thus if that holds true for you, then you ARE who you are on Facebook; and if you truly feel that “…I don’t love a world where I would have to express myself with the person I am on Facebook…” then become the person you want to be, don’t attempt to create an internet world where you can escape from who you really are.That’s not changing the world, but rather escapism to an alternate reality.I am not attacking you, but do think about where you are trying to head, and what type of world you are attempting to create; the ask yourself if this is where you want to head.If it is true that 89% of all facebook users recreate their real world communities on facebook then the issue of “control” is societal and you are attempting to create an escape mechanism or a fantasy world, where Donald or Carl can be whomever they want to be, for a very short period of time.I am a really simple guy and probably not all that bright, but if you truly want to “change the world” and want to do so within the concept of community, then find a way to measure “real” life vs. “websphere” life. Find a way to measure things like: 1. Does the concept of internet community lead to more actual real world interactions of the community participants? 2. Does the use of pseudonymns within internet communities improve the mental health of the participants in real life? You could start with things like anti depressant use. 3. How many people comment and or visit after work hours or on weekends? True communities are personal and after work and weekends are when we are communities in real life.Concepts such as tribes and communities involve physical interaction and most importantly human touch, the success of the web is where it improves upon and or stimulates that, that is why the true success of the web is reflected in things like Occupy and the Arab Spring.The idea that the quality of an online community is directly related to being able to escape who we really are behind persona’s is definitely thinking in the wrong direction.

  61. Engagio is made of people!

    The interesting thing is that I have been posting pseudonymously since 2006 yet recently I have been banned from leaving comments on this site. Go figure! All hail Wm. the Conqueror!

    1. fredwilson

      that should not have happened. what is the userid that you want to use here?

    2. William Mougayar

      Can you please remove Engagio from your user name or delete it. That is the name of my company, and YOU-in no way, shape or form represent it. If you don’t do it immediately, I will take the recourse to ask Disqus to remove it because you are falsely impersonating a brand that is not yours. 

      1. jason wright

        William, time to apply to register a trademark.  

    3. ShanaC

      I apologize about what happened.

    4. William Mougayar

      The other interesting thing is that you have been posting pseudonymously using different pseudonyms and changing them very often and posting very few comments with each.That doesn’t really fit the typical pseudonym persona, but rather is another behavior of its own. Why are you doing that?

  62. Tom_Nocera

    So those of us engaging in public discourse without the cloak of a monkier just need to up the level of our game.  Yawn. 

  63. Sean Black

    Hey Fred, I am now getting your posts via FeedBlitz one day later. I got this one at 3:32am today/January 13 (see attached), but read it early yesterday morning. This started happening a few days ago. Sean

    1. fredwilson

      That’s bad. I’m on it

    2. William Mougayar

      Actually I used to get the same thing as well. The alerts were delayed by almost a day. Then I stopped them and I go to the site directly every morning, because Fred is quite regular in his postings. 

      1. Laura Yecies

        Ditto for me (and Gotham Gal’s feedblitz) – this has been true for me since I subscribed a couple of years ago. 

  64. JohnEBoyd

    Real identities are needed in a lot of real-world situations, but anonymous and/or pseudonyms work well in many, but not all, online situations.  I watched the movie Catfish a few months ago.  So easy to fake who or what you are onlineOne additional benefit of pseudonyms is “branding”.  That is, a pseudonym user builds a following or goodwill (or “badwill”) over time.   Others within the community get to “know” the user.However a “pseudonym” with just a fictitious name is basically the same as anonymous unless the user can control the name over time and preferably provides additional verified information with the pseudonym. Airbnb has been using some interesting tools for verified profiles.  For example, you can input your phone number and they call to verify its a real phone and then displays the last two digits and indicates verified. By itself not that helpful since even crooks have phone numbers, but in the right direction.  Have a great weekend all.

  65. etudor66

    One problem with a pseudonym crafted for Disqus is that while you can create a one-time pseudonym identity on Disqus, it might be taken on another website.Big providers of identity such as Facebook and Google+ don’t provide anonymity via pseudonyms at their core, in fact they actively oppose the use of pseudonyms on their service.We’ve created a new digital identity that uses a pseudonym at its core, and lets you stay anonymous when interacting with people you’ve just met.  You can check it out at

    1. Tyler Hayes

      “One problem with a pseudonym crafted for Disqus is that while you can create a one-time pseudonym identity on Disqus, it might be taken on another website.”True. For what it’s worth we try to minimize this by offering the extra “Full Name” field in the Profile tab at That way if your desired username is already taken, or you want to change how your name appears next to your comments, you can still show the name you want next to your comments. @ryanvalentin:disqus wrote a blog post about this yesterday too: Tip of the Week: Personalize Your Profile.

  66. Eric Brooke

    Its easier to offend a cartoon character online than a real person face to face.  In market research the honesty you get back is higher online than on the phone which is higher than face to face MR.

  67. Eric Brooke

    I agree..I think we have many parts of our identities online, what I put on Facebook is different to what I put on Google+ and different again to Twitter. I use the channels for what they are best at.  I think in reality we are all using pseudonyms, just like we do in real life, we change depending who we are talking to and in which context e.g. work or play.

  68. Youssef Rahoui

    The study seems to be somehow controversial. Anyway, IMO, I would not handle the issue from the quality POV (very hard to assess) but from the freedom one. We should be able to use the Internet with our real name, pseudos or anonymously and no company should have the right to go against that.

  69. Carmen Benitez

    Yeah. Penn Olsen or Amanda Chapel? All very real and relevant pseudonyms.

  70. paramendra

    It is not either or. I prefer to use my real first name everywhere, on all platforms. And with my name chances are it has not been taken already. But there are plenty who would rather use assumed names. Those are the two most engaged groups of people, looks like. 

  71. abhic

    @fredwilson:disqus I don’t know if you are aware but there is an engaged (read – heated) debate going on in over at HackerNews about the Commenting & Community –…Its an interesting insight into the progressive devolution of a community.Even Paul Graham is on record there to say, “I hadn’t looked at the comments on the story about the girl who died till now. I am so embarrassed for this community. I feel like this is the worst I’ve ever seen people behave on this site.”It would be cool to have you jump in there.

  72. Russell Himelein

    maybe if everyone started developing blogs under Pseudonyms the noise that is the majority of blogosphere would start to disappear and we would have quality like AVC everywhere.

  73. saeedomar

    Interesting findings. I’ve written on pseudonyms use in the Middle East, which is quite an interesting behavior. read it here:… 

  74. leigh

    “enragement” new favorite word 🙂

  75. kidmercury

    absolutely, beat me to it……not to mention there are issues surrounding determining whose name is real and whose is fake. plenty of facebookers using fake names. likewise i wouldn’t be surprised if many disqus IDs were fake names that sounded real. i’m sure it would not be very difficult to concoct a study that showed the opposite. i favor pseudonyms and think they have a big place on the web. i’d really love to see a service that let me switch back and forth.  

  76. whitneymcn

    I think it’s worthwhile to note that Disqus presented the results as identifying the “most important” commenters for a community, rather than the “best” commenters in some abstract sense. As I said in another comment, I think that likes and replies are a reasonable measure there — whether or not one feels that any particular community is worthwhile as a forum for discussion and debate seems like a separate issue. Whether or not you agree with them, the people getting a lot of replies are “important” to the community on a blog.But overall, yeah…I think that the substance of this research doesn’t quite fit into the frame that’s used to present it. It’s interesting, but I’m not sure it definitively says anything about a relationship between “real identity” and quality of comments.

  77. sigmaalgebra

    Easy patch up:  Just replace ‘quality’ with ‘quantity’ in the conclusions!  Then the result is still interesting and a little less easy to criticize!With so much data available, might pay closer attention to some of the old work some of the social scientists had to do to make good sense when they constructed ‘measures’. In physical science, that time, temperature, and velocity were good as ‘measures’ was fairly easy; in social science, many of the constructed ‘measures’, e.g., prestige ranking, influence ranking, are difficult to justify, and ‘quality’ in blog posts will be similar.Construction of Measures 101 will discuss ‘reliability’ and ‘validity’.Until we get past this 101 level, we will continue to construct measures difficult to justify.

  78. CJ

    Conflict breeds discourse. So maybe quality could be measured as something that promotes engagement.  Therefore controversial comments, even if they appear to lack quality on the surface, serve to promote greater engagement in the community and so could be viewed as quality comments. 

  79. Cam MacRae

    I wish they’d published their methodology because my enragement would be lessened if the number of times a comment is replied to was a factor in some composite measure of quality.

  80. fredwilson

    the reason you are called an “instigator” is a lot of people reply to your comments

  81. JLM

    No way! FG is WAY too sensible. Sorry could not resist.

  82. Kasi Viswanathan Agilandam

    Loved it.Blame it on the lame guy.

  83. William Mougayar

    The Eagle has landed..that’s the first thought on my mind when you pop in and start rifle shooting our comments 🙂

  84. Rohan

    Missing a nice long JLM story! It’s been a while! 🙂 



  86. JLM

    @ccrystle:disqus How could you possibly say that Obama’s record on taxes and jobs is better than Texas?Texas has created a third of all the net new jobs in the entire country during the last 3 years.Obama has an abysmal record on both.Fight fair and accurate.As for secession, Texas has no desire to secede and never has.  Though we do have the right to subdivide into 4 states and thereby get 8 Senators and a platoon or two of new Reps.  I like THAT idea though not really practical. 

  87. LE

    “a service that let me switch back and forth”You’ve developed a brand. I’d stick with that.

  88. CJ

    Mine is a pen name, I originally wrote fake but I don’t think that’s correct as this is my internet persona and if I ever do publish any work beyond blog posts, this will be my pen name.  So it doesn’t have as much to do with anonymity as it does keeping my ‘real’ life and internet/writing life separate and there only because, like most aspiring authors, I prefer a pen name to my own.  I wouldn’t have any issue sharing my real name here, it’s just that I used this one when I signed up for Disqus.  *shrug*

  89. panterosa,

    And quality according to whom?? It’s all so mushy.

  90. Mark Essel

    Sentiment analysis is one of those problems companies are working to commoditize (Mechanical Turk, Humanoid). But quality analysis is altogether a different beast.The definition of quality is subjective to the individual, and community where the judgement is made. Hard problem Charlie.

  91. Donna Brewington White

    I like that you wore your pink panther t-shirt to the meetup revealing the person behind the pseudonym.  

  92. ShanaC

    is it quality because you learn from it or because you like it or both?Hard to say.

  93. panterosa,

    It was funny for me to see who got it at the first event – mostly those who did were regulars, and then they came to the drinks after, where others recognized it.

  94. Mark Essel

    I skipped out before drinks, and missed the revealing shirt. This is fun not knowing.I can’t resist guessing based on communication patterns. The heavy use of question marks, which sites @panterosa comments on.

  95. Douglas Crets

    I said something like this this morning. The conclusions of this analysis are just half-baked. I did not mention the FredWilson investment. But I also don’t think that bias alone led to Wilson promoting the idea. His interest in this topic seemed to come before he invested. In any case, I really don’t think pseudonyms have the strongest correlation to quality. I think quality and community are contextual.

  96. LE

    “P.S. I see that Union Square Ventures is an investor, so I believe that Fred’s assessment of this issue is biased.”USV investments are no secret. Someone who is biased doesn’t open themselves up to criticism that will happen on a blog like this.  Number one rated comment (right now) is by @ccrystle:disqus and starts with “Number of times a comment is replied to? Really? That’s a signal of quality? “Edit: Sorry I shouldn’t have used the word “bias”. What I mean to say is “trying to manipulate” in a non-transparent manner.

  97. LE

    “Fred’s assessment of this issue is biased.  I’m disappointed, because I’ve been a long time subscriber of MBA Mondays.”I’m curious what the lesson you learned from this is?

  98. fredwilson

    this blog is biased Grace. full stop. if you think you are reading unbiased opinions here, please get rid of that idea immediately

  99. LE

    “Right. There’s little correlation between the use of pseudonyms and quality of discussion, if any”What research or body of evidence are you basing that statement on? 

  100. Douglas Crets

    It might be more accurate to say self-interested in the topic for the benefit of the attention it brings to that which he is invested in. All that aside, though, there needs to be better quality in drawing some of these conclusions about what the data means. I find it difficult to believe these conclusions because they are so isolated from the whole ecosystem of how blogging communities work. – posted via

  101. awaldstein

    I don’t know.Shock jocks became shock bloggers became shock commenters.Wearing a loud shirt just to get noticed and create some conversation works but quality? Depends how you measure it;)

  102. awaldstein

    Good, then you buy the wine next time for being contrary;)

  103. Cam MacRae

    no problem – just doing my bit to improve quality. (next time i’ll reply in english.)

  104. sigmaalgebra

    In some cases in practice there are ways to make good quantitative measures of qualitative situations, but these ways can be difficult to design.In my project I make good quantitative measures out of ‘meaning’, but my techniques, while more general than my project, still have relatively narrow applicability and are nothing like general enough for, say, the Disqus effort.For Disqus, hmm ….Fast, off the top of my head, wild guess number 1: Build a directed graph of arcs and nodes. For each Discus user, have a node. If user X likes a post by user Y, then have an arc from the node of user X to the node of user Y.Assume that each user has a level of ‘quality’ in their own posts, in what they like, and in the quality of the likes they receive.Now see if that graph, as given, can be used to determine unique quality numbers, say, from 0 to 100.If so, then look at reliability (essentially standard deviation) and validity (essentially bias in expectation).Another approach to such things is factor analysis: That is, use principal components, that is, use the theorem that every symmetric, positive definite matrix A can be written in the form of QDQ’ where Q is orthogonal, Q’ is its transpose, and D is diagonal with positive main diagonal.Then each column of Q is an eigenvector of A. and essentially defines a new measure, and its importance is given by its eigenvalue which is on the main diagonal of D.Then by qualitative means, determine which of the ‘principal components’ corresponds to ‘quality’. Now have a numeric measure of quality.That’s basically how IQ is done, but got to be pretty dumb to believe it’s very good!

  105. ShanaC

    We need better methodologies of doing so online.

  106. ShanaC

    I’m still trying to decide what “engagement” means.

  107. ShanaC

    Shock jocks also allow our culture to expand by making things less shocking.

  108. awaldstein

    Hey…I’m all about shaking things up.But if you are going to quantify quality, then be straight about how you think about it is all I’m saying.And sure…it’s ok if you are just about the headlines but then don’t expect a true response. Expect a reaction. 

  109. ShanaC

    @awaldstein:disqus That’s just it, can we quantify quality knowing that?

  110. panterosa,

    The explanation? Of who I was? Of the reference? Please elaborate Charlie…- posted via

  111. fredwilson

    wow. i had no idea. what’s your real name?

  112. fredwilson

    Edgy smacks drive community when they cone from trusted sources

  113. CJ

    Well the picture is actually me and I use ‘Malcolm Lloyd’ in all of my online postings so it’s really an alter-ego at this point so it ‘feels’ like me.  In any case,  My real name is Cory Wesley. *Edit – it was harder than I thought to write that. 🙂 I guess anonymity does give one a sense of security.

  114. Laura Yecies

    Why do you care? If pseudonyms are a good thing. (sorry – couldn’t help myself on that comment)Of course it’s likely that pseudonyms get more comments because they can hide behind the pseudonym to generate controversy without consequence if they want to.  I certainly don’t associate that with quality.I think pseudonyms should be allowed to exist – I want to know that they are pseudonyms but once I know my respect for what they say diminishes greatly – thats a personal thing – i’m not suggesting it for others but it is true for me.

  115. Ro Gupta

    Hey Charlie – quality inherently is impossible to define to satisfy all people and contexts. In this case, signals used were based both on data availability and b/c we see that high levels of likes and replies tend to correlate with actual 2-way discussion vs. just 1-way comments. So it’s less about evaluating the content of the comments, which is difficult to measure accurately, objectively or directly. More about indicating how conversational and vibrant a given community is based on those two user actions.That said, we’d like to do deeper dives in the future to inform product dev so would love to hear your thoughts on other ways to approach. Part of the issue may just be the varying connotations of the word ‘quality’ vs. thinking about ‘healthy’ or ‘successful’ communities instead.

  116. Donna Brewington White

    Hmmm… do you like to code?

  117. CJ

    I’m thinking he only asked because I indicated that I wouldn’t have a problem sharing my real name here.  

  118. fredwilson

    got it. thanks.

  119. James Ferguson @kWIQly

    Fred – If you suppose that all opinions are subjective and thus biased, it may be in danger of belittling the truth (this is not an accusation but an argument – if you care to follow this)…This is a deep issue so allow me a longish post -I accept your implicit position that an observation though it may be factually correct is, in the very process of publication by an observer, driven by motive. Surely the case for all commentators here ? (rhetoric  – but answer if you wish).So use of a real name no more exposes our motives than the use of a pseudonym or anonymity, the difference lies in accountability.And in accountability – there is an implicit accountant here we are talking judgement !Human motives remain anonymous (you may protest your motives but you cannot lay them on the table).So judgement of motive is always left in the hands of the listener.This is why it is important that there is always a listener ( for a theist this is the role of God himself, for the Humanist / Agnostic I am unsure ).This allows inherent and personal value in external truth (regardless of subjective inner motives or corruption).There has been great debate (Luther and Calvin) in distinguishing salvation through belief or through works.  Consider that faith requires the an act of both nomenclature and authority  (eg that  Jews and Christians are identifiably  “washed” in My Name” mikveh or baptism), thereafter, even as mere statement it refers  in(deed) to an act. (regardless of the physical manifestation)In other words when you speak you either “do truthfulness” or you “do falsehood”.Scripturally Buddhist, Christian, Jews and Muslim (alphabetic list 🙂 and no doubt others are warned of the consequence of doing otherwise –  the  Truth is revered above all else as synonymous with God.This is very clearly shown in James 5:12 –  “Above all, my brothers, do not swear–not by heaven or by earth or by anything else. Let your ‘Yes’ be yes, and your ‘No,’ no, or you will be condemned.” This emphasizes that Truth (capitalised) is not personal but objective and conceptually stands above all else.The teaching (which you will respect regardless of your theology) – rhetorical again 🙂 is that value is in what you say, and what you do and this is not only consistent with but IS what you are. So to conclude to be judged by words and deeds is all that is available to any member of this forum. regardless of how they may name themselves.So  we can name ourselves as we will James Ferguson or @kWIQly:twitter matter little, but remaining anonymous our comments can only be read in the context of the mass.By scriptural definition. Truthlessness is Godlessness because it cannot be identified and salvation depends on the naming ! Wow – Heavy never knew I was a preacher :)Wow-squared – If a Humanist reveres Truth as an absolute is he axiomatically Theist ?Fred – If you got this far I must now apologise for the rants emotions and name calling that may follow on such a subject – I meant to achieve non nocere – but that’s another post altogether  🙂