Every Signal Starts Out As Noise
During BlogWalk 1.0, a few of us concentrated on how blogs could serve as an early warning system to alert organisations to developments that require a response, i.e. as a business intelligence tool. (See the picture of the poster we made)
As one of the organisational attitudes I put forward as required to do this, was the notion "that every signal starts out as noise". It sounded cool at the time. What I meant was that you cannot know in advance what is useful information until you recognize it as such. Noise in this situation more or less equals unsorted data and information, and signals are data-patterns.
Information Overload
Why do we call information and data coming to us noise? Because we know not all that stuff is useful, we label the unuseful stuff as noise. And because of the tilted signal to noise ratio we perceive, i.e. what little we actually use from what comes at us, we say we suffer from information overload. I say that this is rubbish.
There is no such thing as information overload. It does not exist.
When trains were first introduced passengers suffered from jet-lag like symptoms, even at speeds as low as 20 km/h. Most likely because for the first time sensory input became asynchronous. What you heard and smelled (the train, people in the car with you) did not coincide with what you saw (the landscape passing by). We adapted, we have to do so now.
Different domains, different information needs
Dave Snowden of Cynefin, when he talks about complexity, divides the world in 4 basic realms. They range from the known, the knowable, through complexity to chaos. In chaos action precedes any order, in the known and knowable order, or routes to order, precede action. In complexity there is an intricate interplay between action and perceived order. Action is based on immediate observation, and shifts and changes with the surroundings, until patterns stabilize and migrate into the knowable or known realms. Only with hindsight is the causality chain apparent, and it does not hold any predictive power for the future in the complexity domain.
Each of these realms has its own information need, and hence a perception of what constitutes noise, i.e. unuseful information.
In the areas of the known and the knowable, where causality can be predicted, any information that forms a distraction is considered useless up front. In measurements we e.g. call them anomalies, and ignore them.
In complex surroundings there is however no way to know causality in advance, and hence there is no way to tell signal from noise. All stimuli are equal at the start. And you need as much stimuli as you can possibly get, in order to heighten the chance that you will stumble upon something that will form into patterns for you, that has meaning to you.
Knowledge Work Is Making Sense Out of Noise
Knowledge work, in my view, increasingly means shifting your activities from the known (routine practice) and the knowable (the classic approach of science), to the complex, where you have to choose your course of action by judging as much data as possible on the basis of experience, skills and attitude. It's in the complex realm where disruptive innovation is born, where continuous (action) learning is a prerequisite. Knowledge workers, ergo, need to be exposed to as much background noise as possible, to open up as much opportunities to respond as possible.
Adapting to 'Information Overload'
Information overload I say does not exist. It is based on the wrong assumptions, that try to apply the way we work in the known and knowable areas to the complex realm.
Those three assumptions are:
In the known and knowable areas, these assumptions make sure that you take the right decisions about causal relationships, and thus be able to control or to cope with your situation. This made perfect sense where information was limited or even scarce. There is fear hidden behind these assumptions, that culminate in the fear to make the wrong decision.
Those three fears are:
In complexity just as there is initially no distinction between signal and noise, there is initially no right or wrong decision. There is just decision, and the three assumptions don't hold up, and no longer improve your track record.
From a complexity perspective the answers to the three fears, and the three assumptions are:
Or when put into actions:
We've Dealt with Information Overload Since the Dawn of Humanity
It all boils down to trusting upon the notion in Open Space of "whatever happens is the only thing that could have". It is also basically how we humans have looked upon the earth for most of our existence. We always take in all our surroundings all the time. We focus on those things that seem most intriguing, and in the end act upon only those of the intriguing things that we think require a response. The old sense, fright, fight or flight sequence. We now have to learn to apply that to written and other new media. But 150 years of our industrial and a hundred more of our scientific paradigms (and educational methods!) stand somewhat in the way, until we learn to surpass them.
Cro Magnon, as information overloaded as we are, only media have changed
Blogs as Early Warning Systems
Let's bring this back to where I started. When organisations reach a certain size (albeit in number of people, the different geographical locations they occupy, or the different cultural backgrounds they encompass) natural communication flows are interrupted, and connection to the general background noise is lost, and there is a need for tools to fill that gap.
Blogs create and aggregate an enormous amount of often prefiltered background noise. Being exposed to the blogosphere enables companies to reconnect to their noisy surroundings. It requires however that they accept that their organisational structures are not all that make up reality and don't want information to flow along those structures only, and also accept that they will not know in advance what is useful information. All signals start out as noise, basically until someone decides it 's a signal. To disseminate blogging in an organisation some simple social network mapping might help establish who are the essential trusted people and hubs that could get blogging started. These maybe also are the people most likely to enjoy blogging, as they are already above average exposed to inputs from their surroundings. It's what makes them hubs in the first place.
Excerpt: Ton’s Interdependent Thoughts: Every Signal Starts Out As Noise This is an insightful, if long (for a weblog), examination of...
Weblog: cloudy, chance of sun breaks
Tracked: March 28, 2004 09:31 AM
Excerpt: Ton Zijlstra has done a eloquent blog, Every Signal Starts Out As Noise. He argues, provocatively, Why do we call information and data coming to us noise? Because we know not all that stuff is useful, we label the unuseful...
Weblog: The Ourhouse Weblog
Tracked: March 28, 2004 03:33 PM
Excerpt: How can we can we get good connections between different online places (email lists, forums, blogs and so on), work out which tools do what best, deal with so much communication ... and also think on what personality type and
Weblog: Designing for Civil Society
Tracked: March 28, 2004 11:26 PM
Excerpt: In einer Diskussion über Weblogs als Frühwarnungssysteme in Unternehmen meinte Ton Ziljstra an einer Stelle: "Every signal starts out as noise". Es klang ziemlich cool und ich hatte so eine...
Weblog: Das E-Business Weblog
Tracked: March 29, 2004 01:07 AM
Excerpt: Long sequence that started with John Moore's comment on Open Access, and ended up with Ton Zijlstra's Piece on Every Signal Starts Out As Noise from which he points to this poster from Blogwalk 1.0
Weblog: A Networked World
Tracked: March 29, 2004 02:44 PM
Excerpt: I may be in trouble. Ton Zijlstra has a recent piece on signal vs. noise in which he says: Ton's Interdependent Thoughts: Every Signal Starts Out As Noise Why do we call information and data coming to us noise? Because...
Weblog: Knowledge Jolt with Jack
Tracked: March 29, 2004 11:43 PM
Excerpt: Ton has written an extremely clear, thoughtful and original piece on blogs, information overload and signal to noise ratios. It's really well worth a look. Briefly, his argument seems to be this: We have always had a lot of stuff...
Weblog: Monkeymagic
Tracked: April 2, 2004 03:38 PM
Excerpt: Reading Stephanie Carlin's overview
Weblog: Mathemagenic
Tracked: July 23, 2004 04:44 PM
Excerpt: Interesting observation @ Every Signal Starts Out As Noise - Ton Zijlstra "Why do we call information and data coming to us noise? Because we know not all that stuff is useful, we label the unuseful stuff as noise. And because of the tilted signal to n...
Weblog: soulsoup
Tracked: August 24, 2004 06:32 AM
Excerpt: On information overload vs cognitive overload
Weblog: Monkeymagic
Tracked: November 22, 2004 03:01 PM
Also see Sebastian Fiedler.
Posted by: Ton Zijlstra at March 28, 2004 09:55 AMfound your article rather "striking" when i found it via martins e-business blog. i have to admit that my worldview (or paradigm) was a strict "scientific" one for a long time, focusing on cause and effect and the models connecting them. i still believe that this is how the world is actually working. but i don't believe, that i have to (or can) base my decisions on knowing and applying al the relevant models when coping with real world complexity. (took me quite a time actually, to reach this conclusion.)
i still am somewhat suspicous with regard to intuition and subconcious enlightenment - no one is going to make an esoteric of me for quite some time. but i think you made a very good point selecting the example "blogosphere" for how we cope with information overload. i myself feel this overload often when sailing the blogosphere (and, as a newbie, my blogroll only contains some 20 or 30 items). and i too have felt the fears you described (missing something important etc.) but ever so often i get in some kind of "flow", skimming lightly over the waves, diving into the depths only when there's that special kind of spark in the back of my brain (or in the front lobes). (i reckon you know Csikszentmihalyi's flow-concept ...)
to reach this state and be able to optimally absorb and use the information and opinions flowing through the blogosphere i think it is important for me and all of us contributing to this overwhelmingly large network, to efficiently utilise the tools at hand. personally, i find it a big "flow-stopper" for example, when i only see a to short or no excerpt at all in my news aggregator and have to actually go the bloggers website, only to find that some keyword sounded interesting but the arcticle is not. do not misunderstand me: i have no problem to visit the "real blog" AFTER i have the feeling that the article is relevant. but i find it unneccessarily tedious to do this only for checking the relevancy. this is a problem especially with some of the big us-based blogs like slashdot and the rss-feeds of the real-world papers (for obvious reasons. they need the adviews ...) martin also tends to have this to-short excerpts in his blog. that there's another way even with those primary aggregating blogs can be seen with cory doctorow (of boing boing fame). his excerpts are short, concise - but no substitute for following his links. and that's fine with me, cause it keeps me in the flow.
i think (being an ashamed egocentric most of the time), that this is not a personal preference, but an important aspect of the "blogosphere as a group-mind", especially when using blogs for knowledge management. processing is much faster when it is not hampered by additional - and unneccessary - time consuming steps.
Posted by: markus at March 29, 2004 10:07 AMThanks Markus! For your extensive commenting.
Meanwhile also see: SoulSoup
Posted by: Ton Zijlstra at March 29, 2004 02:01 PMAlso see Ralph Poole.
Posted by: Ton Zijlstra at April 1, 2004 05:47 PM
Weblog by 
Lilia Efimova *
Gary Murphy
Sebastian Fiedler *
Thomas Burg *
Terry Frazier
Dina Mehta *
Stuart Henshall *
Spike Hall
Phil Wolff *
Jim McGee
David Gurteen *
Elmine Wijnia *
Julian Elvé
Denham Grey
Ian Glendinning
Paul Goodison
Postings that link here