Introduction: Background and influences
If you’ve been reading this blog for a while, then you’d have come across the name of Clay Spinuzzi. I’ve been following his work for about five years now, and had the privilege of meeting him for breakfast while vacationing in Austin in the summer of 2008. Clay introduced the term “ambient signalling” into my thought process, a key ingredient in the thought process that led to this post. It made me read his work on organisational genres and on participatory design, often within the perspective of networks.
A number of other people have also influenced me considerably when it comes to this particular post, and I’d like to declare my debt to them at the outset. Howard Rheingold (whom I was meant to meet earlier this week, but couldn’t, as a result of my unwisely choosing to tear ankle ligaments rather than fall down) really set the scene for me over two decades ago in his Whole Earth Review and WELL writings, followed by his excellent book on The Virtual Community in the early 1990s; one of his other books, Smart Mobs, which I read nearly a decade ago, was also a key influence.
The next breakthrough came when, as a subscriber to Chris “Rageboy” Locke’s Entropy Gradient Reversals, I was taken on my first ride on the Cluetrain. I can’t tell you how many times I’ve read The Cluetrain Manifesto, but it’s in double digits. [Disclosure: I met Chris soon after the book was published, Doc soon after that, David shortly after that and Rick a few years later; they remain good friends of mine, and I was deeply honoured when they asked me to contribute a guest chapter to the 10th Anniversary edition of the book.] Cluetrain really got me thinking about community interaction from a business perspective, beyond the fortress-like walls of the corporation.
Around the same time, I had the opportunity to read Amy Jo Kim’s excellent Community Building on the Web. Over the years I’ve bought at least five copies of the book, it’s one of those regularly borrowed, rarely returned; I had the opportunity to meet her at Supernova some years ago. The book sparked my personal interest in game design and game mechanics from a business perspective, something Amy Jo continues to work on with her usual flair.
Shortly after that, I read Steven Johnson’s Emergence, deepening my understanding of slime mould and ants and swarming, all against the context I’ve described above. I’ve had the opportunity to meet Steven since, and continue to follow his work as well; he really helped me understand something about headlessness, how everything is a node in the network.
The final, and critical, influence was that of Clay Shirky, whose blog I followed religiously over the years, even when his output shifted more to books and speeches. It was he who helped me bring all this together with the trenchancy of his analysis of how communities work. Insights from him on three aspects helped establish my thinking: to create sustainable commons material, the cost of repair should be at least as low as the cost of damage (the undo button in Wikipedia is an example); that there is no such thing as information overload, only filter failure (which helped me understand something about the recommending, curating, filtering roles of network members); and that in the age of knowledge workers, much good can be achieved by effective use of what Clay terms Cognitive Surplus, the title of his most recent book. I’ve known Clay for some time as well. We had a delightful lunch together in Davis this year, and I visited him at Tisch when I was last in New York, a wonderful set-up.
Why am I telling you all this? Two reasons. Firstly, to give thanks where thanks is due, to point out the people who have influenced and inspired me in this particular context. Secondly, as a consequence, to give you the opportunity to go deeper into the influences, research things for yourself. Some of you obviously know all this, have read all the books, met the authors; for you, this may seem onerous and repetitive, and you’ve probably skipped all this anyway. This introductory section is for the rest of the readers.
For a few years now, I’ve been looking at information systems and services as if they were biological in origin, serene in the attitude that, if the 20th century was meant to be the age of physics, the century we live in will be characterised as the age of biology. Using that perspective, it was only a matter of time before the “ambient signalling” spoken about by Clay Spinuzzi would start feeling like the pheromones laid down by ants as they go about their chores. Which, unsurprisingly, led me to Wikipedia, to start reading up on pheromones.
There I found the definition of “pheromone” to be: a secreted or excreted chemical factor that triggers a social response in members of the same species. Now that interested me greatly. A “social response” as opposed to any other kind of response. So I looked further, and Wikipedia informed me that Peter Karlson and Martin Luscher introduced the term “pheromone” in 1959, to influence specific behaviours from two or more “conspecifics”, members of the same species. The intended etymology of the neologism was itself of interest: they formed this word from the Greek roots pherein (to transport, to bear) and hormone (stimulus, impetus): so a pheromone became something that was a carrier of stimuli. Hmmm.
Types of pheromones
It turns out that there are many types of pheromones, classified in different ways. You have the concept of primer, releaser and information pheromones: primers kick off changes in development events, releasers make you change your behaviour, and information pheromones just tell you things. If you look further, you find far more detailed classifications of pheromones: aggregation, alarm, epideictic, signal, terrotorial, trail, sex, and so on. If you’re interested, please read the Wikipedia article on pheromones yourself, which gives you the basics on pheromone types.
Intriguingly, “there are physical limits on the practical size of organisms employing pheromones”.
Twitter, Chatter and their pub-sub nature
I’ve always thought of Twitter as a publish-subscribe mechanism, and, in similar vein, of Chatter as an enterprise bus with pub-sub built in. [Disclosure: I’m Chief Scientist at Salesforce, the people behind Chatter. It’s one of the key reasons I joined the company]. I’ve been lucky enough to work with people who’ve believed in bus-based architectures for some time now; and, ever since delving into EDI in the early 1980s, I’ve been a convert to recipient “beneficiary” driven transaction and messaging systems.
Over the years, as I’ve continued to play with Twitter and with Chatter, their innate strength-from-simplicity has become more apparent to me. Which is why I wrote posts describing Twitter as a “submarine in the ocean of the web” some years ago. I’ve been able to use Twitter to rescue hamsters lost down holes in floorboards, to get visas for foreign travel, to collect hand-me-down recipes for ragu, to acquire limited-release CDs in the city of origin, the list goes on and on. And, now that I’ve been using Chatter in anger for the past six months, I’ve been able to learn something about its differentiated value. How following “things” as well as people becomes valuable in a business context. How exception handling becoming the norm is no longer a frightening thought. How closed and open groups can overlap and coexist. How institutional memory is established and rekindled, how new forms of knowledge leadership emerge as a result.
Tweets as pheromones, and their Chatter equivalents
Nearly a year ago, I spent some time looking at why we share and what we share, using tools like Twitter. More recently, I took this further, in a three-part post looking at social objects and their role in such communal enterprise activities, which you can read here, here and here. And a few weeks ago, I tried to put all this in the context of why sharing is important in the enterprise, a theme I looked at tongue-in-cheek here.
Which brings me to the nub of this post.
“Birds do it, bees do it, even educated fleas do it”
Like falling in love, providing signals meant for sharing is normal and natural. We have to start thinking of tweets as the knowledge worker’s pheromones. Signalling. Alerting. Marking out “territory”. Warning off. Pointing towards food or shelter. Looking for relationship. Sometimes preparatory, sometimes catalytic, sometimes just plain old informative.
But always social, always designed to share.
Sometimes only visible to your “conspecifics”, to those belonging to your own species.
Sometimes visible to all.
Sometimes reinforced by repeated overlays and relays.
But always always social, always always designed to share.
More later, as I extend this theme into compound and multi-authored shared signals, and of the immense value in being able to record, replay, aggregate, analyse.
That’s for my next post….that’s if I find the comments and feedback such that writing a next post becomes worth the while. [Reminds me of an apocryphal Churchill-Bernard Shaw story. Shaw is meant to have sent Churchill a pair of tickets to the opening night of one of his plays, saying “bring a friend… if you have one”. Churchill is meant to have replied, returning the tickets, “can’t make opening night. will make second. if you have one.”
18 thoughts on “Thinking about Twitter and Chatter: the knowledge worker’s pheromones”
Another great post. And there are other cool metaphors operating down at the cellular level with the receptor and the way it communicates with networks of other cells within an organism. Cell, organ, organism, organization. It’s turtles all the way down….
Not to argue with the beauty of Twitter’s simplicity, but lately I’ve been wanting the ability to segment my tweets out to followers based on their differing interests – kind of like the way people categorize blog posts. Tags themselves don’t quite do it because while they help others find tweets of interest, they don’t filter out tweets that aren’t interesting to my followers.
Here’s an idea I’d love to get your take on since you’re thinking about Twitter and Chatter right now:
Insightful post JP. You are onto something with your tweets=pheromones analogy. As an advocate of social media within the enterprise for several years now, I have been calling blogs, comments, tweets etc as Digital Breadcrumbs – the digital equivalent of the breadcrumbs from hansel-gretel nursery tale. Based on your blog post pheromones seem more apt. I’d also include facebook likes, retweet, blog posts, comments and other such signals as various aspects of the pheromone complex. Thoughts?
Associated random thoughts:
Pheromones > Social — Google has recruited Ed Chi an expert on Information Scent — understanding how users navigate and understand information environments such as the Web.
Biology > Pub-Sub — I’m fascinated by the Transcription Process of DNA and its analogy with information interfaces. Pub-Sub is the closest information interface I’ve seen matching the Transcription process of nature. That is information is transposed from one actor to another without binding the interface and with automatic error correction and mutation. (“reinforced by repeated overlays and relays”)
Chatter > Disappointment — I signed up for a free Chatter account, but deeply let down I couldn’t communicate with anyone but myself. Looking forward to some mutation.
Signalling is something that I’m trying to understand from the molecular level on up – http://www.bio-balance.com/Ref.htm
This is really a fascinating post. Thanks to @gideonro for pointing me to you on Twitter.
Do you have a Twitter handle so that I can get updates?
Like this take JP:
“Like falling in love, providing signals meant for sharing is normal and natural. We have to start thinking of tweets as the knowledge worker’s pheromones. Signalling. Alerting. Marking out “territory”. Warning off. Pointing towards food or shelter. Looking for relationship. Sometimes preparatory, sometimes catalytic, sometimes just plain old informative.”
Dave Winer describes it as: “narrating your work”. I like your amplification here.
the debate on tweet-everything-in-one-place versus have-different-tweet-identities has raged for some time. I have never liked the idea of segmenting my twitter-id, it seems an artificial route. similarly I can’t see the tweet-to-list working, because the lists are not standardised, they overlap, there is much to clean up. following a list still makes sense. tweeting to a list… I am not convinced. in a perfect world I would want the publisher to use universally accepted hashtags to segment tweets, and subscribers should be able to follow a person while explicitly blocking tweets for a particular hashtag for that person. you shouldn’t block the hashtag universally. so you should be able to say “i’ll follow JP for food and for tech but not for news or music” while you still follow others for music and news.
there is more to it than just the trails. see my next post.
sorry about the disappointment. watch this space. thanks for the pointers on Ed Chi.
interesting. thanks for keeping me posted. see what you think of my follow-up post, it starts deviating from the biological alone….
thanks for the feedback. i’m @jobsworth on twitter, I will ping you there as well.
let me know what you think of my follow-up post, I’ve tried to reflect where my head is at on this
Thanks for your thoughts on the Tweet-to-List idea, JP. You may well be right on some of the implementation challenges, but it’s interesting to note that Facebook just announced today the functional equivalent of what I’m talking about with its new “Send” button:
Gideon, glad you and FB think similarly. I’m still uncomfortable with the segmentation at publisher level. Simplifies the tools for central censorship. The more control we build into the core, the more we create a choke point for misuse