I came across a post on Knowledge Futures quoting Dave Snowden, about knowledge as an interpreter in the abililty to turn data into information. And then using a sensemaking process (making sense of this information/understanding it) which can create new knowledge to you.
This throws the hierarchy view of Data-Information-Knowledge-Wisdom as separate steps or levels, into more of a flux environment.
From the post on the role knowledge plays in data and information:
“…knowledge enables me to interpret information. If I acquire knowledge of management accounting then a chart of accounts informs me, if I have no such knowledge then it is data. Knowledge management this has, as one of its primary tasks the creation of sufficient shared knowledge to enable the use of information.”
From the post on Information Management and Knowledge Management:
“I do see utility in understanding the different between what it means to manage knowledge and what it means to manage information. I normally do that with a metaphor of the difference between using a london taxi (knowledge) and a map (information) to get around London. The map is data which has been structured to inform and if I share sufficient context with the map maker then it informs me and I can take action on it. I can also get a taxi where not only has the taxi driver internalised the map, but lots of other things as well. There is for example evidence of significant changes in the Hippocampus in London Taxi Drivers as a result of the two plus years of training they go through. Compete with a taxi driver (as a map user with a hire car) and you will loose. The map may get you there, but the assumption of shared context can be dangerous. I once used a map in New York and almost got mugged for exactly that reason. Its like the point on french cuisine - you may have the recipe but that is just a starting point it is not complete of itself.”
An object like a map is more static (unless it’s a wiki or a blog) so it only has set information, and depending on your know-how of maps, you will be able to read the map, and create new know-how and use it to get around town…the map is only information, it’s up to you to create the know-how (meaning).
The taxi driver is a dynamic information base that continually learns about the area the map represents. She contains lots of informal information that is not normally represented in maps, as that’s not really a maps job or purpose.
What Dave Snowden hones in on is that the shared context with the map is assumed. Shared context is one of the most important aspects of successful information transfer, it’s assumed that you know the context of your activity eg. an understanding of the topic, an understanding of how your team deals with this topic, establishing aspects of the topic. The more existing know-how you have on the information and its context, the more chance you have of successfully acquiring new knowledge and taking action accordingly.
Knowledge as interpreter
I really like knowledge as the interpreter, similar to what was said on Anecdote a couple of years ago, including this amazing diagram.
“Knowledge acts as an interpretant to turn data into information. The information we notice (we don’t notice all information channelled toward us), might create some level of dissonance (its surprises us or we ask ourselves, “What’s the story here?”) and if we care about resolving this dissonance we create knowledge. Knowledge is created through a sensemaking process.
But data to one person is someone else’s information. A commodities trader might stare at a computer screen of numbers which would look to most people as raw data. To the commodity trader, however, slight changes in the numbers conveys messages which act as information they might convert to knowledge (via sensemaking) and take action. Consequently, context is a key ingredient acting as an underlay to all three concepts of data, information and knowledge.”
Dave Snowden also has a similar diagram.
[ADDED 1/7/08: Joe Firestone’s paper, Key Issues in Knowledge Management, also deconstructs the Knowledge pyramid. This paper goes into a lot of theory related to this blog post.]
My stream of consciousness
You use your current knowledge or understanding to see data as information (not sure if you are actually turning data into information)
If you don’t possess the knowledge then all you see is data.
If you do possess the knowledge, you then make sense of this information in the sensemaking process where you may gain new knowledge (understanding).
I like how Anecdote say that the information you see may create a “dissonance” (kind of like you understand the information, but how does it relate to the whole), this is done by a sensemaking process, and “if successful” you have gained some knowledge…and perhaps take some action.
In theory, next time you are in this same exact situation (ceteris paribus), the level of dissonance would be non-existent, meaning there is no need for the sensmaking process and no new knowledge is created.
If this happened all the time you may feel you need a new job that is more stimulating and challenging.
Where there’s dissonance, there’s learning to be done…and knowledge gained.
Is it possible to never experience dissonance because you have reached nirvana?
I personally don’t think a highly evolved spiritual person, living in the now, means you have finished learning.
Experience as Interpreter
The above describes that you need knowledge in order to have the opportunity to create (discover/acquire) new knowledge.
But this can’t be right.
As a baby I may burn myself by touching a flame, as I have no knowledge that the flame will harm me.
How does this work, was the flame data to me, as I didn’t have the knowledge to be informed that it will burn me if I touch it?
And have I created new knowledge (not to put my hand near a flame again), all this without having knowledge to act as an interpreter to turn data into informatiom.
Since I could not turn data into information, then how could I possibly go forward in the cycle to use sensemaking to create new knowledge.
In this example rather than requiring knowledge to interpret data to information, an “experience” has become the interpreter.
I’d like to read about knowledge from an infant psychology perspective (if there is such a thing). Because if you have knowledge of nothing how do you ever start?
Etienne Wenger briefly mentions the need for social learning theory to connect with developmental theory.
[ADDED 1/7/08: Joe Firestone’s paper helps out here, on page 16 he says:
“…we are born with genetically encoded knowledge that enables us to interact with the external world and to learn…”
He desribes this as world 1 knowledge:
“…encoded structures in physical systems (such as genetic encoding in DNA) that allow those objects to adapt to an environment”]
If a piece of data on a screen is flashing, “anyone” can notice that something has just happened to the data, it’s flashing, and if you possess the knowledge about what this signal means then it informs you, ie, it’s information to you…I guess it becomes knowledge if you can successfully interpret the meaning of the information.
Even though you know what this information means there is still a level of dissonance as you don’t yet know why it occured, you try to make sense of it and work out what’s going on, once you work it out, you have created new knowledge.
Structured to inform
Dave Snowden says “The map is data which has been structured to inform…”
“We have a mess of unstructured data to which we apply structure or interpretation in order to inform others, we put the data in context”
“If I structure data through process of abstraction and possibly codification then I create messages with which I seek to inform someone else. If that person understands the message they are informed; however if there is no shared context between message creator and message receiver then we are left with data, no information is created.”
A whole heap of random names means not much, but if the list is titled “customers aged 12-20″, then this becomes a structured message intended to inform.
If a person has shared context they will receive the message signal as information, ie. that these random names are people in an age bracket.
This does not imply that structured data equals information.
The data is structured or in an organised format, and whoever understands this organised data receives it as information.
Knowledge is the tool to achieve understanding and interpreting this process, but not only that, your knowledge has to have a shared context to receive the intending meaning.
Without the title these random names are just data, more precisely unstructured data.
According to the model in Dave Snowden’s post once you go through the sensemaking process, ie. you understand what the information means, eg. making inferences from comparing two lists you may work out that, “customers aged 12-20 tend to have more overdue books than customers aged 21-35″, then you can go through a path-finding process and take action.
The action you choose from your path-finding process could be sending a reminder out more often to customers aged 12-20.
If someone gave me a project management schedule (eg. MS Project) it would be data to me as I don’t understand the technique and the symbols.
But if I was proficient in MS Project then it would be information as I can understand (interpret) this software.
But if I don’t share context with the author (know the person, history of the project, etc) I may have a hard time understanding (sensemaking) this information completely in order for it to become knowledge for me.
Knowledge is not a thing or object, it is what a person uses and creates.
Dave mentions that the knowledge management is about providing or creating conditions for shared context.
Shared context enables you to get to first base, which is being informed…to be able to use information, so we can take it to the second base in the sensemaking process.
I’m using a step process to explain, but really this happens more fluid or in a flux…all the steps mentioned could be happening at the same time.
Rather than the loop of Data - Information - Knowledge - Options - Actions being the main components, I’d rather think of what happens (based on Dave Snowdens diagram) as Analysing - Sensemaking - Pathfinding - Executing (ASPE)
To be continued…