Twitter is Not a Conversational Platform

Perhaps the most common reason given for joining the microsharing site Twitter is “participating in the conversation” or some version of that. I myself am guilty of using this explanation. But is Twitter truly a conversational platform? Here I argue that the underlying mechanics of Twitter more closely resemble the knowledge co-creation seen in wikis than the dynamics seen with conversational tools like instant messaging and interactions within online social networks.

Wikis are causally thought of as platforms for “collaborative” document creation. But on Wikipedia, while many people share knowledge to co-create pages, the process is not formally collaborative in the sense that contributors are not cooperating with each other ways that form group identity (to paraphrase Clay Shirky from his book Here Comes Everybody). To the contrary, passionate experts write the majority of text, and a long tail of other contributors offer relatively few, small edits. Many users contribute nothing. Through this process, Wikipedia pages often become valuable repositories of knowledge.

Brian Solis recently posited the dichotomy of whether Twitter is a conversational or broadcast platform. New data bears on this. According to a Harvard Business School study, about 10% of Twitter users contribute roughly 90% of its content. Anecdotally, these 10% are subject-matter experts, passionates, mavens, and thought leaders who break news, write strong opinions, and tell jokes. Like on Wikipedia, most users merely read this information, and a modest number of people in the long tail use the information in the form of re-tweets, comments, corrections, and alternative opinions or links.

So while an individual user may use Twitter primarily as a conversational tool or a broadcast medium, in its totality, Twitter operates a lot like a wiki: as a knowledge-sharing, co-creation platform that produces content and allows its consumption. Conversation is perhaps the most simple and obvious form of collaboration, but would anyone claim that Wikipedia is a conversational platform? Despite the presence of information sharing, co-creation of an end product, and even discussion pages, Wikipedians on the whole aren’t having conversations.

According to this argument, Twitter is no more a conversational platform than Wikipedia is. But is it a social networking platform? New HBS data showing that men have 15% more followers than women and being twice as likely to follow another man than a woman also bear on this to some extent. Authors Bill Heil and Mikolaj Piskorski state: “On a typical online social network, most of the activity is focused around women – men follow content produced by women they do and do not know, and women follow content produced by women they know. Generally, men receive comparatively little attention from other men or from women.”

As in the case of the conversational platform, it seems that Twitter is also no more a social network than Wikipedia is. Wikis have user accounts and discussion pages, and it is possible for relationships to form. Twitter has user handles and direct messaging, and relationships can form. But social relationships on Wikipedia and Twitter are not a prerequisite for satisfaction and success (inasmuch as that can be defined). For instance, the popular and useful account @BreakingNews has hundreds of thousands of followers but participants in effectively zero engagement. There are many Twitter users who contribute large amounts of useful information and engage in relatively little conversation. And it is not common for people to describe Wikipedia as a social network.

Andrew McAfee notes that two useful Twitter traits are its asynchronous and asymmetric nature. These two traits are also critical to Wikipedia, but importantly much less so within popular social networking platforms like Facebook and MySpace. Thus, entities that are clearly social networking platforms can be but are not necessarily knowledge co-creation platforms, and entities that are clearly asynchronous knowledge co-creation platforms can be but are not necessarily social networks.

If microsharing tools resemble wikis more than conversational tools and social networks, this has huge implications for how people and organizations approach use of this emerging technology. Solis suggests, I think rightly, that “sometimes it’s effective to…maintain a presence simply by reading, listening, and sharing relevant and timely information without having to directly respond to each and every tweet.” The strategy of being a “lethally generous” member of a community would seem to be more worthwhile in this context, contrasted with the individual-level customer service approach of (for example) @ComcastCares.

This framework for thinking about microsharing platforms as knowledge co-creation enablers also puts Nielsen’s recent data on Twitter’s “user retention and loyalty” in a new light. When the average user is a consumer of the content produced by subject-matter experts and passionate mavens, how much does it matter if the majority of use is infrequent spectating (particularly when the information is archived for asynchronous retrieval)? As Shirky recently noted in his talk at the IAC/ACT Management of Change Conference that I attended in Norfolk, VA, such an imbalance of contribution is not a condition of failure for the platform or its users.

Finally, if microsharing is equated with knowledge co-creation, rules for attribution becomes an important consideration. But while the wiki attribution process has generally been worked out, attribution on Twitter is like the wild west – there are no rules; only conventions that are commonly accepted in some circles but not others. In addition, it is relatively easy to cheat the system, hard to catch someone doing it, and difficult to determine what the consequences are of such behavior. This problem will be a lasting one, requiring careful consideration by not only the user community, but also Twitter itself.

tags: , , ,
  • Funny you should post this on the day I write about conversations on Twitter :-)

    Maybe Twitter isn’t a conversational platform in a formal or well-defined sense. But it certainly does facilitate conversations that would not otherwise happen.

  • Andrew – It’s not that conversation doesn’t occur, it’s that I don’t think that networks of conversations are the best way to understand the Twitter information ecosystem. A long-tail distribution of information sharing entities (accounts) co-creating knowledge is better, I think. I think that the dynamics of chat rooms, IM, Facebook, and so on are quite different. This affects both tactics and strategy for both organizations and individuals.

  • Richard Tryzno Ellsberry

    (Regarding issues of attribution on Twitter, they’re discouraged by the 140-char limit. Sometimes you can back-track through the RTs to an original source.)

    I believe very strongly in ‘Twitter groups’ being the way of the future, and by this I mean an integral Twitter ‘grouping mechanism’ as opposed to 3rd-party add-ons (eg, — appears to be gone).

    A good real-world testing ground would be my own hometown of Baltimore MD, which would benefit immensely from folks being able to carry on real conversations (I agree) in real time with each other, while being archived if necessary. Twitter groups would be a powerful & low-cost organizing tool for small non-profits and arts venues. Instead of today’s architecture of 1-to-millions, community groups would narrowcast a-few-to-a-few — within the larger Twitterverse. The possibilities are mind-boggling.

    Richard Tryzno Ellsberry

  • M.J.

    This confirms a lot of what I said here ( in response to this Scoble post ( about the character of Twitter. Facebook is largely about conversation and sharing with friends; Twitter seems to be about building on ideas in an entrepreneurial, collaborative way.

  • I agree that Twitter is not, actually, a conversational platform. But – it could be a way to get to know someone and start a conversation. Many pitches come to me for my online magazine because a PR person (or other individual) learned that I liked a certain product or service from my tweets… and continued the conversation outside of Twitter. Either to get to know me better, or to introduce themselves so that we could have a working relationship. So true conversation? No. Conversation starter? Definitely.

  • Kelcy

    I do believe that Twitter is a conversational platform. Conversation is after all the basis for knowledge sharing. Twitter facilitates the ability for conversational knowledge to extend beyond a group of people (whether large or small). Twitter allows us to include our personalities along with the knowledge – to share our humor, anger, compassion, love, and other passions that are a part of knowledge development which includes the networks that we create.

    Referencing the Harvard article on gender bias a proof of your argument is rather weak since both Zack Tumin and Drew Conway have questioned the validity of that particular study (

  • Kelcy

    How about calling Twitter a Conversation Knowledge Management Platform (CKM?)

  • It’s not the same thing as facebook (Though facebook has tried to copy it lately in their redesign of status updates), and it’s not the same thing as a chat room or instant message, but if people were just talking “at me” and there were no conversation taking place, I wouldn’t be there. Passive acceptance of broadcasted information is the past.

    I think Twitter is voyeuristic instant messaging.

  • I’ll disagree here – I do think Twitter can be a conversational platform, in that (unlike Wikipedia) it spurs and invites participation in larger conversations based on interests, opinions and personal life experience.

    To be more accurate, Twitter is a broadcast medium that can lead to conversation. This isn’t much different than text messaging – if you want to have a long and meaningful conversation, you make a phone call or send an e-mail. But a text is useful for quick status updates and pre-call communication.

    I appreciate the Wiki comparison, but I’d shelve Twitter under “conversation” before I’d shelve it under “wiki.” Personally, I shelve it under “microblog” – but then again I like micro-definitions. ;)

  • I think it’s both a broadcast and conversational platform – and it depends on how someone is currently choosing to use it.

    For example, I use Twitter primarily as a broadcast channel and that includes me searching for content that others have broadcast and for me doing my own broadcasting. I find most of my interesting blog posts and articles to read comes from monitoring my Twitter feed several times per day, and I try to reciprocate with my own feeds.

    However, I know people who do use Twitter for conversations but they are generally people who have passed more than 10,000 lifetime tweets. I can also see the benefits of a small company engaging in conversations. In any case, this type of engagement takes an investment of time that many users probably don’t have or haven’t made time for yet.

    Tom Humbarger

  • I am a newcomer to ‘Twitter’ and must say I like its simplicity without the usual kettle of fish used on most other sites (Facebook etc)…People as you know are entitled to post as much or as little as they like and everybody there is and will be on a very different level of education/experience and I think it is a very quick and easy way to post without being toggled into staying online for hours as you would do with other sites such as Facebook…It uses minimal downloading for wireless dongle users like myself…I feel I am not as exposed on Twitter as I am on Facebook and think that after Facebook came around (Which I loved in comparism to Myspace) I didn’t think Facebook could be beaten and though different kinds of sites entirely, I do actually like Twitter! As much as/differently/more than Facebook! Must go! It’s time to Twitter and talk talk talk! :D xx

  • Well put. What the alpha 10% really don’t realize is that most people don’t really feel compelled to speak up. 90% are perfectly content as voyeurs.

    More than that, most of the 10% have 97% of the followers. So they will NOT REPLY TO YOU partially for fear that they will loose their followers.

    full findings at my posterous.

  • twitter is a living breathing collective knowledge brain, we push and pull on it with our comments, searches and links. its smart but can be deceived, luckily it seems to learn quickly and correct itself.

  • Mark,

    Once again you have made me pause and think. So thanks.

    Personally I try to have fun with Twitter and I try to learn from the many tid-bits and trinkets folks post there. Some of that is conversational, so I might take a little exception to your title, but most of the value I gain is from the key point of your article: it is the knowledge co-creation. The folks I follow on Twitter make me smarter and I appreciate that. I wish I could give back and I try but I get far more value from Twitter than I put in.


  • I have asserted for some time that what’s making twitter is the important and/or interesting people on it, and the fact that the 140-character funnel simultaneously makes it easier for them to participate (by reducing the chance they’ll get flooded), and increases the odds that you can whisper something in their ear and they’ll actually hear it.

    The local aspect of it is rather useful too; I have mourned the almost complete lack of geographic locality of reference on the Internet both at the technical peering and audience levels, for almost as long as it has been since the Net almost completely replaced BBSs.

    Nice to know I’m not the only one thinking in that general direction.

  • Todd Weidman


    This should be required reading for all the corporate “social experts” who see the twitter platform as conversational in nature.

    In response to a Tim O’Reilly blog post here about a month back (Goodreads vs Twitter) I gave my opinion on the Facebook vs Twitter argument:

    “My objectives for each platform are, for now, at the opposite ends of the social spectrum (I will answer the question of why I use each).

    Facebook to me is highly personal. The symmetric approach is exactly what is needed in order to protect certain thoughts, pictures, etc that I share. My relationships with my FB friends allow me to discuss the highs and lows of life – as my friends have the context to understand what I’m describing and living. I’m not ready to share much of that yet with the twitterverse.

    Additionally, with respect to urgency and relevance, my updates – and the updates from my friends – can be read at leisure. Important from a personal context, but not important from a professional context.

    Twitter is a different story. I use it much like I traverse through the unstructured postings in Usenet groups. The micro-focus on my interests allows me to immediately identify those users from whom I can learn. I use it to “go to school”. I also use it to “teach” others – those that follow me – when I have something relevant to add to the conversation, via broadcast or via dm. I have found the links, reports, research, etc that are distributed via twitter to be invaluable. I use the smart apps to cut through the noise and have a focused “classroom” experience. The context is professional – career building -special interests – that I can control.”

    Knowledge sharing is what twitter is all about – right now. The building of relationships is an outcome from the knowledge building – like minds finding each other and discussing possibilities.

    We can think of three levels of engagement:

    1. If people want to learn, listen in to the chatter.

    2. If people want to share, contribute to the dialogue.

    3. If someone shares something intriguing, and you believe there is a benefit to a coordinated conversation with that person, twitter reduces the friction in making that offering. You just need to be sure you have something to give back – your own contribution.

    I think that third level of engagement is still evolving – it happens today, but I’m not clear I can see the outcome, aside from the wealth of knowledge that is shared. Co-creation to me is some deliverable – content, product, business plan, etc.

    I know we’re headed in that direction with this micro-sharing medium, as I’m sure many other folks smarter than I are figuring this out right now. I’d be interested in hearing how this is evolving.


  • Thanks, Mark for your analysis. I appreciate your bringing together a lot of the “literature” that has been floating around the past few weeks.

    Your argument is strong against Twitter as a conversational platform. Your description reminds me of the many list serves that I have been on over the years. There is always a small group of people who actively participate, and everyone else soaks it in. So is Twitter just a shortwave list serve?

    I would offer, tho, that the people conversating think that it’s conversational–and so do many of the lurkers. I don’t doubt that they’re wrong, they just don’t know it.

  • My view is that Twitter’s viral appeal stems from its low entry threshold to group membership allowing people the sense of “in-ness” by being essentially annonomous members of conversational groups without having to speak. Mechanically it is Instant Messaging with a more permanent and lower energy method of group formation. It blends features from IM, (micro) blogging, RSS News Feeds, Social Networks and email email into a single simple interface that does not trigger much technophobian or social resistance – ergo easy entry “conversation”/group in-ness = rapid growth.

  • You can (and do) have conversations on Twitter – but they’re stilted. I find Plurk a better platform for conversations – mainly due to the layout & the drop-down comment box. I describe Twitter as being for statements and Plurk for conversations.

  • Mansih

    New HBS data showing that men have 15% more followers than women and being twice as likely to follow another man than a woman also bear on this to some extent. Authors Bill Heil and Mikolaj Piskorski state: “On a typical online social network, most of the activity is focused around women – men follow content produced by women they do and do not know, and women follow content produced by women they know. Generally, men receive comparatively little attention from other men or from women.”
    So what is it? Men have more followers or women?

    Not sure if the arguments are really valid. Twitter enables far more conversations than without it. More on comments in the post

  • As at least one comment was implying, it depends on what lens you look at Twitter through.

    Between what you and Andrew Maynard wrote today, along with others who try to define it, Twitter starts to look like the descriptions of the elephant from the five blind men:

    … “It’s a social network”
    … “It’s a micro-blog & personal branding tool”
    … “It’s a world wide chat room”
    … “It’s an unstructured social wiki.”
    … “It’s a dessert topping!”

  • Twitter isn’t either a conversational or a broadcast platform, it upends the dichotomy and all these variously Aristotelian definitions. If you throw away the definitions, however, and take a prototype approach, while Twitter looks like both, it simply doesn’t look like a Wiki very much.

    Much of the problem is the idea of “content.” Twitter can act like Wikipedia, but it’s not generating “content” any more than a really good extended party involves “content.”

    More generally, these definitional arguments run up against the basic poverty of arguing over terms and definitions. It isn’t a large cat. It isn’t a striped bear. Some day we’re realize it’s a tiger.

  • The Harvard study seems to have garnered all sorts of attention, but perhaps instead of just blindly taking its outputs as correct, one might wish to consider whether there are any flaws in the methodology and the conclusions that have been drawn… If so, spend a few minutes here

    for an assessment on the limitations of the study.

    I love Paul Worsham’s comment – he is right, and coincidently this is a point that Clay Shirky does make in “Here Comes Everybody” – platforms get used in all sorts of ways by the people who coalesce around that platform (Read the bit about twitter use in Egypt). I’ve seen a whole bunch of posts on “the categorisation of twitter” and they all seem to fall into the same format: twitter is/isn’t [insert metaphor here] and therefore this means [insert point of view here].

    We should be guided by the words of William Gibson; “The street finds its own use for things, uses the manufacturers never imagined.”

    As publishers, we should understand the assorted uses that our various communities are making of this (and other platforms). We should then act on that knowledge accordingly.

  • This article has nothing to do with whether or not Twitter can be “used to have a conversation” or “used as a social network” but rather with understanding the patterns by which entities (accounts) interact within a complex system. The way in which information is shared and knowledge is created has a lot in common with the *patterns* seen in a wiki. While the discussion page of a wiki (for instance) can be used to have conversations or as a social network, too, it is not really useful to describe a wiki as these things. So too, perhaps with unregulated microsharing.

    Reposting this great link by M.J. which hinted at this earlier ->

    David Smith: Obviously the HBS work is not a full-blown scientific study. It’s a jumping off point for discussion. It’s pretty obvious that the Twitter ecosystem does not behave like a typical social network, if for no other reason than the relationships are largely assymmetric.

    Baylink – The nearly complete lack of geographic references w/ tweets is an abomination. And tools like TwitterLocal are inadequate. Politicians campaigning, not to mention many other groups, would find a “location filter” extraordinarily useful.

    Kelcy – Based on Shirky’s definitions, conversation is a form of collaboration, and collaboration is synchronized and forms group identities (paraphrasing). Sharing and co-creation is a simpler form of organizing without organizations, and the latter seems to be more in line with what occurs on Twitter (which does not stop people who meet on Twitter from forming group identities through collaboration in other information-sharing environments).

  • to me Twitter is mostly broadcasting platform, FriendFeed – conversational. Simply even possibility of normal conversation didn’t realize in Simply even Twitter’s concept (140 characters) against the normal conversation.

    Here (in commenting section to this post) there is a conversation. Can you do the same on Twitter?

  • Saying that Twitter is used for publishing not conversation is like saying cars are used for commuting not recreation. And saying that wikis are about publishing because Wikipedia is publishing, is saying that music is about arena rock and not singalongs.

    With Twitter, the same medium supports both publishing and conversation. Some people – especially the most famous people – use it for mostly broadcast and not much response. Other people use much more heavily for interaction.

    Wikis can be used for publishing as wikipedia is – or for collaboration – as are many smaller wikis used for groups and organizations.

    A given tool can have multiple uses, within the same community or as used by different communities.

    Also, the 90/10 participant ratio in Twitter isn’t by itself a sign of the publishing/conversation split. That number is probably at least as much about the traditional “lurkers to participants” ratio in public online communities, and less about what the participants do when they get there.

  • Just because Twitter doesn’t retain a convenient historical thread of the statements, it can be used to converse in the vocal style rather than the written style. A “real” conversation is two or more people talking to each other. Written conversations and particularly threaded conversations are relatively new to the human experience. They are a record of a conversation, they are not _the_ conversation. Unless a conversation is recorded, the only way a newcomer or participant can revisit a statement is to ask someone to repeat it. Therefore the recording and threading of past statements is not a prerequisite for the existence of a “conversation”.

  • Interesting article, Mark. There’s been a fair amount of “what-is-Twitter,-REALLY” discussion recently in the Twitter- and blogosphere. As the slicing and dicing unfolds, I’m not sure we’re getting any closer to the truth. What Twitter IS is perhaps indefinable. Twitter is somewhat of a chameleon: it’s whatever a user/reader/contributor WANTS it to be. That’s strictly IMHO; but it gets me through the “minefield” of trying to define the indefinable. (Not sure why Justice Stewart’s “I’ll know it when I see it” comes to mind when contemplating this, but that’s another issue.)

    @sengseng asks: “Is Twitter a platform for collaboration.” Discussion among/between you, @timoireilly, @amcafee (& others) ensued. Split decision. My thought: Twitter doesn’t necessarily provide a collaborative product; but it does provide/enable a collaborative process. Wikis provide a collaborative product that is enabled (via the wiki itself) by a collaborative process.

    As for “is Twitter a social networking platform?” Presenting statistics of male/female follow/follower behavior, lashed up with contributory/discussion patterns on wikis doesn’t necessarily mean Twitter OR wikis aren’t social. (From my own experience in my last job, there was a rich social network that formed around wikis, blogs, and chat. I left government service before Twitter or other microblogging platforms gained widespread use in the organization.)

    It may be difficult to characterize what Twitter is or isn’t. That may be cause for some consternation for “web- or enterprise 2.0” consultants. But maybe Twitter is such a game-changer that maybe it’ll be okay for consultants to paint Twitter in a somewhat Impressionistic manner.

  • I have a big problem with using the Harvard study to characterize twitter as a one-to-many broadcast network. I don’t think those 90% who don’t tweet are “listeners”. Over and over again, I see people who hear about twitter, get an account, follow one or two people they already know. They don’t know how to find other people to follow, and a tweet from someone they already know once every week or two doesn’t make it work the trouble to log in. So their account basically goes dormant.

    I think most who are actively listening are also actively talking. For the most part, these are not one-on-one personal conversations, but more like a cloud of interaction, in which bits of information or personal news are tossed out into the waters; the give-and-take of it makes people feel “in touch” with their chosen group. I don’t think “knowledge building” is an apt description for much of it. It’s more like the mingling one might experience at a large convention.

  • twitter provides many options. I agree

  • A social service like Twitter is not going to instantly make introverts into extroverts. With 10% posting and 90% reading/listening – that sounds like most conferences I attend. Panelists speak and only a handful ask questions or challenge their views. Sort of sounds like Twitter; pick who you follow and create your own group of panelists.

    Information sharing. Instant updates. The ultimate conference room? Why not just leave it as a social service, because it quite simply is a service that caters towards social interactions. Why do things need to be labeled?

  • Carol

    Hi, thanks for posting this. I find it very interesting as I’m currently researching Information sharing on Twitter for a thesis. I see what you’re saying and I agree, its a microblogging site rather than a communication device, not that it cant be used for communication but its more like a public notice board than, say, a telephone.

  • I’d counterargue author Mark Drapeau notion: “
    I argue that the underlying mechanics of Twitter more closely resemble the knowledge co-creation seen in wikis than the dynamics seen with conversational tools like instant messaging and interactions within online social networks.”

    Twitter as anew webbeast platform or digital form has a potential to development being a proteyan and chameleonic re-creative process of thinking or twitting vs webmonster wikis’s established patterns of writing which in principle has a noncreative essense with dominant plagiarizing style of mirror text writing and not thinking vs twitting as twinning thinking instrument nature.

    Thus, I’d argue with author’s notion that ” ..Twitter operates a lot like a wiki: as a knowledge-sharing, co-creation platform that produces content and allows its consumption” I’d contrast the dinamics of twitter with wiki’s stalled metrics of anonimous writing as plagiarized editing or (self) censoring pattern of thinking and I agree with contrasting views posted by Zoe Winters: ‘I think Twitter is voyeuristic instant messaging’ along with alternative view as counterpoint by steve blumberg:’ twitter is a living breathing collective knowledge brain’ which in my view epitomises twitter’s (re)creative cognitive aspects of competitive and entertaining nature.

  • Before you put too much stock in that Harvard study, ask yourself: how did they identify gender? Did someone (a person) look at each twitter account and make a judgement? After all, some people don’t share a name, just an ID. Some names are androgynous (if that word can be used to describe a name!). Some people use pseudonyms (and some simply lie).

    OTOH, you have have seen their methodology. I have not — the initial release on the Harvard site (two blog posts) provided no details.

    Turning to your point, I think that you are missing intent. The intent of Wikipedia is to create an artifact (dynamic thought it may be) that captures the essence of a subject. The are mechanisms for conversation of a sort (The Talk Page) but the conversation is task-oriented.

    Twitter, on the other hand, is very different. It’s more like the telephone than wiki software. Some people use the technology to have conversations with a small group of people. Some use it to share ideas with lots of other people. The end result may be knowledge sharing or knowledge aggregation, but that’s not the *intent*.

    Finally, I’d really REALLY like to see the same network analysis on the use of phones or faxes or email or even the US postal service … my guess is that the 90-10 or 90-9-1 rule works _there_ just like it does in other spaces.

  • A few recent twists in my work on Twitter shed some light on the unpredictable roads I’ve been able to discover with Twitter, and show how hard it is to “describe the elephant” — Starting with Swine Flu, in April (incidentally, I think, an excellent example of the unfolding evolution of Twitter being defined by events – Who could have imagined the way swine Flu would play out on Twitter). In my tweeting on Swine Flu, I looked especially for news from Mexico, where the disease was much more serious than in the US. In doing this, I made many good contacts with Twitterers from Latin America and other parts of the world. … Then, after the Swine Flu stream had past, I pursued a subject that had been brewing in my mind, having to do with Salman Rushdie envisioning the Web in one of his stories … and, lo and behold, some of the contacts I had made around the world in working on Swine Flu were also interested in this seemingly very different subject, but apparently the Rushdie idea resonated with them, I think because of their cultural mileau. Another subject I’ve been tweeting and blogging on recently is eBooks in the Third World, and this, quite understandably, also resonates with the contacts made around Swine Flu.

    So, my point is … to paraphrase the visionary words of Rushdie — Twitter is a very big ocean, that contains many different streams, and the streams can change as they flow, starting out as collaboration, morphing into conversation, going from one subject to another …

  • Tom Waters: I think that the 90% or so that don’t tweet much can be considered “sporadic spectators” of Twitter, in the sense that they may read things from time to time, perhaps long after they were written. But someone who has signed up for an account, even if they don’t use it, is at least peripherally aware of Twitter, and thus they may go back to it for information at some point. I don’t think this is too different than someone who edits Wikipedia once about their hometown, doesn’t edit anything else for a year, but looks stuff up once in a while and basically knows what Wikipedia is.

  • Rachel

    I see Twitter as context. It’s like dipping my toe into the ocean of current information — I get a sense of what people are talking about, and a list of things that might interest me. I notice I get news through Twitter first (usually NPR, but various sources), see clear emergence of new trends or ideas, and can follow individuals for specific updates in fields I’m interested in. I have a separate group for friends, for the same sort of context (like “everyone’s talking about Tien’s wedding today” or “how many of us got laid off this week?”) It’s like a broad outline of what’s happening in the world, with links and blogs to fill in some of the details I’m interested in.

  • I’m partial to the term “social presence.” While it’s possible to have conversations via Twitter, I find its most interesting effect is to enhance the feeling of connection to my community (the people I follow) by having constant exposure to their ideas. This makes it more likely that I will engage some of them in conversations (typically through other channels than Twitter) because I already have a sense of what they’re interested in or what they’re thinking about at the moment.

  • Shaquana Grant

    I love this article. Just
    one note though just
    because social networks
    usually work in one way
    ie men following women
    etc does not mean that if
    another network
    functions another way, it
    stops the network from
    being catagorized in
    anyway. Twitter is
    changing everything! Lol

  • Twitter is Not a Conversational Platform
    Twitter soon be over or just spam marketing website

  • Talking and sharing information do not necessarily constitute a conversation. I think of conversations having a number of attributes that Twitter doesn’t necessarily carry off that well:

    – A narrative thread that is flexible and changeable, but follows a familiar pattern. Twitter lacks that (or rather, the way that Twitter threads are typically represented make the narrative very difficult to follow unless you spend some time learning the lingo)
    – A coherent, relatively stable, and acknowledged set of participants. Twitter conversations jump around between participants, you never really know who you’re talking with, and random people can jump in at any point
    – Shared contribution. Twitter obviously allows anyone to participate, however, and Mark notes that 90% of Twitter content is produced by 10% of participants. This is more like someone holding court at a party with people standing around saying “hell yeah” or “ditto” rather than a real conversation.

  • Adam Richardson – Perhaps the best comment yet, really getting at the definitional nature of what “conversation” is. Part two of your comments really gets at why Twitter info-sharing and “talking” are not necessarily “conversations” – a lack of formation of group identity – it’s never really clear who’s part of a conversation and who isn’t, and so it’s hard to know who’s “collaborating.” It truly is in many cases like someone ‘alpha’ holding court with ‘betas’ chiming in later.

  • To twitter or not to twitter.

    First of all. From my friend congbo; T”witter is not about Twitter, it’s just one more opportunity, one more format, giving us a richer language in the conversations that technology, for now, are limiting”

    Second, twitter is populated with media/advertising/pr people who still behaves like they´re operating in the world where old communication models worked. To quote Kevin Rothermel;
    ““They collect followers on Twitter as proof of how brilliant they are at social media marketing, without realizing the irony that they are just turning their Twitter feed into a broadcast medium that reaches more people than they could possibly hope to have a “relationship” with.”

    We need to step down of our high horses and be more open for conversations. Participate, not solely broadcast ourselves. That said, it depends what your purpose with using Twitter is…

    More here:

    Joakim. Oslo. Cold. Summer?

  • I’m really fascinated by this discussion of the nature of twitter. From the start it has seemed to me like a massive discussion forum. But instead of following a topic or thread I am following a person. Well, I am following lots of people to see if they will say anything interesting and relevant to me. And 3/4 of those people are following me back to see if I say anything interesting. I know more or less who these people are. If I was following 10000 people rather than 400 then I might have more of a problem.
    82% of my tweets are conversation. They are in reply to the tweets of others. I don’t see how I could see twitter as anything other than a conversational tool.
    Sometimes I see a #tag from a conference in my stream and I search to see all tweets using that #tag. This is more akin to following a thread on a forum. And I might start following some of those using the #tag because they are people who I want to have in my network and who I want to converse with.
    I don’t feel that I use twitter as a broadcast tool, any more than if I was posting on a forum or in a Ning community. And I don’t follow those who might want to broadcast to me.
    Is twitter like a wiki? Well, no. Is it like a conversation? Yes.

    (family doctor and university lecturer interested in medical education)

  • And how do we know the 10% who do tweet are broadcasting to the 90% who don’t? The 10% could be talking to each other. The 90% might have chosen to use twitter as an alternative to an RSS feed and are happy to be broadcast to by an even smaller percentage, maybe 1%.

  • I think it takes Time to understand what´s really going on on twitter. Asynchronous and asymmetric nature of twitter? Thats right! Nice Post!

  • Twitter launched initially as a “microblogging platform” if memory serves, so it makes it makes sense if its still somewhat of a broadcast platform despite that its obviously evolved into a communications platform today. It’s understandable that it’d be true to its core roots — MySpace is still rooted in bands/music users even though usage and function on the site shifted since its original inception.

    It just seems like a natural evolution to me.

    More disturbing in my opinion is how people fail to recognize that social networks are just message boards, in their core functionality, and there’s nothing new about them.

  • some thoughts on taking this idea to mobile devices The twitter phone .. Of OpenGardens, co-creation of content and where #cnnfail is more important than CNN itself ..

  • No, wikis are ridiculous, a collectived format in which the shill is that the many democratically participate and yet in reality, only the few code it, frame it, control it, fill it with content.

    Twitter is completely different, and yes, Twitter is a conversation.

    Perhaps it isn’t a conversation to YOU because as a cheeky geek, you imagine that you should just dispense pearls of wisdom and everyone should follow you in awe.

    Fortunately we can step past you now and talk among ourselves. Make the technology — and then get out of the way, please.

  • eddie

    With you can’t just look at the main feed, but also need to look at the editors’ feeds. They do interact greatly with other people. You need to look at,, and

  • Hi Mark,
    Your post seems to pick stats that back up his assertion that twitter is more like wikipedia, instead conversational. I think microblogging is a lot like blogging but smaller:

    * 10% of the people may do it the most but if they are talking to each other often, even asynchronously, who cares? it’s still conversational

    See danah boyd’s recent research on this

    Where she talks about “36% of tweets mention a user in the form ‘@user’; 86% of tweets with @user begin with @user and are presumably a directed @reply.”

    They got this from a random sample of approximately 720k tweets in several five minute periods.

    Other studies show similar stats. In other words, people do appear to be talking with each other.

    * twitter doesn’t keep your info longterm.. which makes it not at all like wikipedia

    * searches can aggregate a topic, but going in reverse chron view makes it’s also not like a wiki, but like a conversation where people jump in and out.

    I think danah and others are right on this.. twitter is conversational, people drift in and out of conversations, and they spend time directing their words to specific people because of the @’s in their tweets.

    My own experience with twitter is a place where people do speak directly with others on topic. I am often in topic focused conversations, often once or twice a day several days a week.

    It’s not wikipedia, in our conversations. We are are not shaping and honing a single article to some consensus.

    We are often disagreeing and sharing the disagreements to bring out different viewpoints. No one is right.. it’s a conversation with different opinions.


  • Andrew Cherwenka

    Great post Mark. If the argument is about nailing Twitter down to one definition then we won’t get there just yet. It’s a wiki, an aggregator, a chat room, a broadcasting platform, an immediate news source. It uttered its first word just 3 years ago. Who knows what it’ll grow up to be?

    From a marketing or promotional standpoint I’ll say it’s also a conversational platform. If somebody’s talking about their lousy or great JetBlue flight and @jetblue responds, it’s more than just a 1:1 service call. It’s a publicly displayed marketing impression consumed by many.

    (Responding to your tweet on how my post – – went astray. BTW why didn’t I know about when i was manually counting @’s?)

  • I find it interesting that everybody is trying to define Twitter, but nobody has bothered to define conversation. According to Merriam-Webster, the definition is (1) oral exchange of sentiments, observations, opinions, or ideas (2) an instance of such exchange (3) an informal discussion of an issue by representatives of governments, institutions, or groups (4) an exchange similar to conversation”

    Except for the “oral” requirement in (1), which I think we can discard due to (4) I think Twitter qualifies. But I also agree that it is more than *just* a conversation.

    As Bob Gourley said, interacting with others on Twitter make him smarter. The same can be said of conversations, as long as you’re having conversations with the right people. The beauty of Twitter is that it helps us find the right people, or vice versa.

    As Kelcy noted, we could consider it conversation knowledge management (albeit with an expiration date, since Twitter search only finds tweets posted after a certain time).

  • Someone wrote a post here about how “of course Twitter is a conversation.” It isn’t a conversation just because some people have conversations on it.

    Many seem to define “conversation” in a very casual manner, as in, when people send information to other people, that’s a conversation. I define it along the lines of what Clay Shirky wrote in his book, that a conversation meaningfully occurs when a defined group of two or more cooperates with each other in ways that form some group identity. True, meaningful conversations don’t happen when no one knows who’s in some undefined, asynchronous mess of information. Info is shared, perhaps knowledge is created, but very little real conversation happens.

    When I write to @andersoncooper and say “Great show today!” and he doesn’t notice, that’s not a conversation. When a bot account writes @cheeky_geeky and auto-replies “Thank you for tweeting about vaccuums! Would you like more information?” that’s not a conversation either. When someone says RT @scobleizer OMG he is giving a killer keynote! that’s not a conversation. And when @juliaallison twitpics herself with the note “Am I hot in this dress? @ me!!!” and while everyone is writing her she’s on a date, that’s not a conversation.

    Saying that 35-40% of tweets have an @username in them means “conclusive proof of the conversational use of Twitter” is ridiculous. And my new post at Radar gets at the nature of what bots and brands are doing with all these @ replies – not really having conversations, are they?

  • Phil Baumann

    Mark –

    I’ll add one small comment: I think the best description of the exchanges on Twitter is to call them “chats”. Conversations may take place on Twitter, but they’re more like chats – and I’d be hard pressed to refer to Twitter as a Conversational Platform.

    Chatterbox Platform may be more apt. :)