[Note: This is my third post in a series I’ve been writing on this topic; the two previous posts immediately precede this one]. What I want to do here is touch on a few subjects that came up in earlier posts, where I didn’t really have the time or space to express what I meant adequately. My intention in sharing all this is to give you as much depth as I can into my thoughts on the use of tools like Twitter and Chatter.
Connected versus channelled
Some of you may have noticed that, in previous posts, I appear to make a big thing of wanting to place filters at the point of receipt rather than at the point of dissemination, at the “subscriber” level rather than at the “publisher” level. This is no random thought, it represents something I have believed in ever since I took up blogging: you will find it a recurrent theme in the kernel for this blog. There are a number of reasons for it, and I’m going to try and articulate them as succintly as I dare.
Michael Polanyi, in helping us understand what he meant by “tacit knowledge”, is reputed to have said something along the lines of “there are things we know we know, things we know we don’t know, things we don’t know we know and things we don’t know we don’t know”. That fourth bit, the things we don’t know we don’t know, has always intrigued me. As a result, I used to walk around telling myself: “filter on the way out, not on the way in. Let everything come in, you don’t know what you don’t know.” What I was trying to do was to minimise the building of anchors and frames that would constrain or corrupt what was allowed to enter my head, what Einstein called “common sense: the collection of prejudices collected by age eighteen“.
When I see words like “connected” and “channelled” they conjure up different meanings, heavily laden with my own prejudices, despite all my efforts to avoid such prejudices. “Channelled” suggests a one-way street, a broadcast model, a structure where I am a recipient of a signal with all the choices made by the sender of the signal. “Connected”, on the other hand, has a sense of being two-way, interactive, with some sort of parity or equality between the things that are connected.
There’s also something else, something darker, harder to put my finger on, evoking a deep sense of distrust. And it’s rooted in some modern variant of Say’s Law: Supply creates its own demand. What do I mean? Well, let’s take terrorism laws. Come, perform an experiment with me. Open a separate tab or window in your browser, bring up Google and enter the term “UK terror laws used to snoop”. Just look at what you get. Here’s a sample list of the things that local councils have used terror laws for checking whether:
- nurseries were selling plants unlawfully
- a child lived in a school catchment area
- fishermen were gathering shellfish illegally
- alcohol was being sold to under aged
- benefit claims were fraudulent
- people’s dog’s were fouling
- people were littering
- cows were meandering
- calls were made to 900 number phone lines
It’s a much much longer list, with over 470 councils invoking the laws over 10,000 times in a nine year period. Why do they do this? Because they can.
Coming from a family of journalists, and having lived as an adult through the “Emergency” years in India, and having been on the receiving end of some of the power that such states wield, I’ve felt more strongly about such misuse than most.
With all this in mind, I want to remain connected, not channelled. I want to be able to choose what I can know about, learn about, be told about. I don’t want to block out what I don’t know. I don’t want the technology to have tools for censorship built in, which in effect is what happens when filters are designed into publishers. It is too easy to game the publisher end of the market, far harder to game the subscriber end.
So I try and avoid filtering at source. I have no problem with tags, with providing people the metadata that simplifies filtering at subscriber level. But the mechanisms for tagging at source should be designed in a way that they can’t become choke points used by the unprincipled.
Avoiding echo chambers, groupthink and herd behaviour
When social networks are used to share information upon which decisions may be made, you will always hear someone bring up the echo-chamber risk. After all, if you put a bunch of like-minded people together, you will get repeated assertions of the same thing. Or so the theory goes.
Wrong. Now this is not deep research, but anecdotally the results have been positive enough for me to want to assert this. Social networks bring together people who have a few common interests, rather than people who hold common views about those interests, or who replicate those interests. My twitter followers are not clones of me. Very few of them are into chillies and capsaicin in a big way; very few have the same “retarded hippie” tastes in music I do; very few are as crazy about cooking (and eating) as I am; very few are Indian and 53; very few go to church every Sunday. Some do. But not all.
Social networks create value because people in the networks come together, drawn by what they have in common, but creating value because of what they don’t have in common.
There have been a number of discussions recently about the “dangers” of direct democracy: how could we possibly run anything, manage anything, lead anything, based on the statistically expressed will of the Great Unwashed?
Surely what will happen is that people will keep on asking for faster horses.
Perhaps.
Maybe.
But who are we to decide that everyone else is wrong?
The tools we have today allow for greater dissemination of information than we’ve ever had before. Attempts to control, suppress or subvert the free passage of information are becoming harder and harder to pull off, there’s a Wikileaks waiting to happen in every command-and-control centralised hierarchical set-up. These tools are becoming ubiquitous, affordable, effective, and the empowerment of the edge continues apace. Snap polls are no longer about random sampling, not when there’s a Facebook around. [Incidentally, don’t underestimate the value of having good polling mechanisms in systems like Twitter and Chatter].
Democratisation does not yield dummification. Except perhaps in the eyes of elitist experts.
Signals, not trails: improving our work lives
Some of the comments I’ve received, some of the references I’ve been pointed towards, have a tendency to veer towards a trail-like analogy for lifestreaming and workstreaming. This is possibly due to my use of the pheromone analogy. If that has happened I am sorry, that was not my intention. If anything, my use of the wikipedia article in the first post was an attempt to avoid just that, by showing that the pheromone classification went way beyond the concept of trail.
Since then, on a the-physics-is-different basis, I’ve tried to bring in the time dimension as well. The signals we share as we workstream are separable by time, and each “layer” of time does not in any way corrupt other layers, contiguous or not. And I feel the very existence of these signal histories helps us improve our work lives dramatically.
How?
In four ways.
Firstly, they give us institutional memory as to what happened, what was done. This allows us to break away from blame cultures, move towards an environment of “We have not failed, we have found ten thousand ways that do not work”…. but with a difference. By being able to record the conditions under which something did not work, we learn something about the conditions under which something will work. And we can form the equivalent of seed-banks under the icecaps of organisations, storing the seeds we need for conditions that do not exist today, but could exist at a future date.
Secondly, they give us the ability to trend behaviours and forecast with somewhat more accuracy than has been the case in the past, based on data rather than political connections. It used to be said that history will always be written from the perspective of the hunter until lions learn to speak. Well, lions can speak. Now. Histories are less likely to be corrupt if they are constructed by bringing together squadrons of disparate tweetstreams. This sort of crowdsourcing of information has been happening for some time now; I could not hide my glee when I learnt that 18th century ships’ captain’s logs were being used to conduct climate change research. [And thank you, everyone involved in the project, for making sure the output was not behind a paywall, that it was searchable and retrievable. How wonderful.
Thirdly (and this may be my most controversial point) I think they make our work more interesting. Humour me on this. One of the most depressing things about the Industrial Revolution, assembly-line thinking and division of labour was the way human beings were somewhat dehumanised as a result, becoming narrow specialists good at doing mind-numbingly boring things well. Five or six years ago, I had the pleasure of listening to John Seely Brown and John Hagel at a Supernova conference (thank you Kevin Werbach) talking about motorcycle factories in China and how collaboration took place because people weren’t working sequentially. And it got me thinking.
It got me thinking about the new generation, and how they seemed comfortable multi-tasking, how they were being accused of being ADHD as if ADHD was an epidemic [if you have not watched Sir Ken Robinson’s talk on changing education paradigms, stop everything you’re doing and watch this 11 minute video. Then watch the whole thing, the hour long version, link provided below the summary. Thank you RSA!]
It got me thinking about knowledge workers and the lumpiness of knowledge work, the implications for the generation of cognitive surplus in the enterprise.
And it got me to a point where I saw the possibility that division of labour was a thing of the past. That for the millenial knowledge worker in a social network with workstreaming, switching costs were tending to zero.
More to chew on. I’ll be back. Comment away.
Point of receipt rather than point of dissemination, is reflected in the shifting architectures of the web, consumer and enterprise.
Earlier system architectures, often added SDK at the last stage, an attempt to increase adoption by data customization. Often with limited success, since the SDK is often filtered by business logic, some success by connecting 2 different SDKs (Frankensystems).
By provisioning an API closest to the data, and soonest, options for others to consume and build apps, the twitter effect sets up a viral network effect. The more data, the more users, the more value, repeat… (aka the new black gold of Facebook).
APIs allow us to add context, location filtering, and super personalization. Creating new markets for devices, iPhones, and waves of tablets to come.
–//—
Time shifting the pheromone scents, I see as replicating proteomics, unfolding natures proteins, and massive data for personalized medicine (which must be w/o filters to yield results)
I like the way you approached the direct democracy dilemma; that certainly made me grin. After we are done with our little extension of democracy, I’d certainly like to return to writing about it again. What you say about command and control structures always being one wikileak away from collapse is a well made and exploitable idea.
Another useful framing of “things we know we know” is “unconscious incompetence”, “conscious incompetence” and “conscious competence”.
Email is where institutional knowledge goes to die today. You could do worse than turn off a corporate email system and force true personal, private information back onto paper!
The frustrating thing about your blog in general and this series in particular is that it’s so inspiring, yet so difficult to imagine in translation to my day-to-day job; so thought-provoking, yet it’s so difficult to hold on to those thoughts when I’m behind the desk or in the lab (you mention the Cognitive Surplus – but what happens when I’m actually at work?)
Alas I can’t make bits for faster horses, but rather for cars. We’re industrial, we’re siloised and we have >10k employees globally. Whilst I’m sure our IT group can measure the sheer amount of data that is flowing within our company, the culture is not yet there to really make it coagulate (biology again) – and certainly not to have open discussions with customers. (Cluetrain – are we afraid? Yes, I’d say we are).
It’s not all doom and gloom; we’re trying, but it’s still emails and distributing on a perceived “need / safe to know” basis. My first an so far only attempt to generate a Sharepoint discussion forum fell flat on its clunky face. Whilst I’m not a cultural stakeholder in our company, I hope that by demonstrating the benefits of refreshing openness, of trending and filtering, I can garner some enthusiasm amongst my peers and then get some reverse percolation going towards those that make the decisions higher up…
The inspiration is there as long as I remember it is there, which is mostly evenings. But its residuals are present at work… Please don’t forget this as you continue developing your thoughts in plain view!
As someone who has always seen education as leaning how to learn, I think your filtering premise is exactly the right approach.
Indeed it highlights the increasing criticality of filtering as a modern-day skill and how we all need to cultivate an outlook of non-superficial generalism in order to maximise our interpretation of those informational pheromones.
I think you’re correct that value in social networks comes from the differences between individuals in them. It becomes a problem though when individuals choose not to express that individuality. Thats what makes some networks more effective than others.
Interesting blog post