Dave,

I think most decision models are lacking in resolution on the nature of info and the kind of knowledge it generates - and hence the different ways different kind of info impact on "pathfinding" and execution. A linear model such as this would make sense if we are designing a computer agent to make decisions, or if we are formulating a descriptive theory of human decision making.

For humans, the way info is analysed is dependant on the "frame" and the "standing hypothesis" that is dynamically generated as info is received. The same piece of info may be given different weightage or even different treatment (confirmatory, collarary, disconfirmatory, etc) in different context. It is difficult to capture this phenomenon in a 2-D linear model.

So it depends on what the model is suppose to do:- frame programming efforts in a DSS; or prescribe on how humans make decisions, or...

Cheers

I have also had my hair-pulling rants about the data-information-knowledge step model, mostly because it assumes (or seems to represent) that data is a primitive of information is a primitive of knowledge.

Naturalistically speaking, it seems to me that data is information that is purposefully decomposed for a variety of information and knowledge-driven manipulations ie it's a sophisticated knowledge artefact. Data doesn't exist in the wild.

Information is knowledge that is filtered and abstracted from specific contexts to make it communicable - ie so that it can be liberated from the specific and can persist across time and space.

Of course, within society and organisations we don't start from a primitive state. We have a whole lot of baggage made up of the interplay of information and knowledge mostly, with occasional injections of data depending on our jobs.

However, I just don't get the linearity bit in this model or ones like it. I just don't experience linear DIK progressions and I rarely see anyone else work that way.

I might want to compile data on something. My purpose in doing so starts from a context of having knowledge and some knowledge objectives. I know I've got access to various information resources. So I'll analyse what my data model needs to be either implicitly or explicitly. This is now a knowledge artefact, which sets up the specifications for the data I will pull from information available to me. I plough through my information sources or other data sources available to me to compile data into my model. I'll play with it - using my knowledge - to generate new knowledge and probably some information to communicate to other people. Knowledge, information and data are all interacting with each other - indeed, feeding off each other.

Now I agree that options, pathfinding and action are healthier extensions of the model than wisdom or intelligence are , but I just don't see how the message of linear "progression" between elements, or the "A is a building block of B is a building block of C" can help anyone relate to knowledge and information use in the real world.

Maybe you can help me see?

Dave,

I always worry how these models might shape the world views of those new to KM and you quite rightly point out that the stages are much more complex than the model implies.

It does imply that sense making is always of information. Do we not make sense of our knowledge? Do we access our own sub-conscious thoughts, values and beliefs and that of our colleagues always as information that we must first make sense of, to turn it into knowledge?

It also suggests to me the old chestnut definition that knowledge is distinct from information and data whereas lately I have been trying to use your ASHEN approach to encompass them in order not to alienate IS and IT perspectives.

The feedback from the options must return as information to be made sense of, but I agree that it is a more useful model than most.

Cheers
Ron

1. Consider changing 'analysis' to 'analyses'.

The sooner the reminder to consider alternatives is built in, the better. Information is data in context, true, but what might happen is forcing the data to fit a wishful or preferred context (warped information or premature sense-making?). In hindsight we might say 'why didn't we see this coming, we had all the data?', (but our information/context was wrong). The terrorists, who share a different context, communicate effectively between message creator and message receiver. What is a weak signal for the analyst is a strong signal for the terrorist.

2.

I like the concept of the OODA loop, too. What I don't like are the limitations of the diagram and the use of the word 'loop', which both imply that it is O-O-D-A-O-O-D-A-etc, whereas, as I understand it, O-O is going on all the time like a nucleus of energy powering D-A as and when necessary. That's actually how I see it in my head, with O-O in the centre and D-A shooting off and coming back with results.

I can imagine your diagram transformed similarly, with D-I-K in the middle in a dynamic relationship, (a triple knot?), and multiple O-A coming off as fractal curves feeding back into D-I-K. Would underline the complexity aspect too and possibly be a good way to bring in the idea of microloops.

I just worry that when some people see a simple diagram for a complex process, they interpret it simply and try to use it that way.

I'm not sure either that we can separate path-finding from execution from sense-making, because I don't think that we really work that way. The difficulty of keeping various terms distinct while discussing matters like this is an indication of such. Sometimes I wish that the ability to hypertext conversations existed. When you say that you like the idea of asking what the information means, avoiding assumptions, I'm not sure I understand why except as a reality check, which you pick up in the next sentence about going back in context to the original raw data

I often make a distinction between sense-making and meaning-making because people can often make sense of something, i.e. comprehend it and even understand it, but they won't act until it means something to them.

Dave, I have the impression that you feel like being part of the IRAHS effort you have to adapt, at least a little bit?

Anyway.

I see two ways to deal with this model in a productive way. Let me first characterize it's shape as: five linear steps, plus closing the loop to a circle, plus feedback one step back - but not two or more steps. No random connections, no feed forward.

Applications:
1. You (who?) divide the world into cases where this makes sense, and where it does not, or not so much. This is not an all-purpose model, but it may have it's place when there is enough order. Metaphorically speaking: when in Manhattan, count the streets and avenues. When in Venice - well, huh, at least don't try to count the streets and avenues.
2. Recognize that for really huge systems, there are different people/institutions/actors involved. Thus part of the boxes and arrows are things done by different people at different times - and then it makes sense to get into this circular model as opposed to a fully connected network. Even better if experts in different professions can see their place in the model, and thus also get an idea about the whole, and their place in and contribution to it.

Finally a comment about OODA: yet it may be better, closer to "the truth", but it may be too complicated, and more difficult to explain, remember, and to get from understanding to action. The Poindexter model is simple enough, "lies to children"-style.

Hi Dave,

In this context you (and others) may be interested in a piece by Sonali Ojha entitled 'Pathfinding' - as a 'gateway' - in Fieldnotes 12 from the Shambhala Institute

http://www.shambhalainstitute.org/#

Dave

Dave,

Ik think the first step is wrong: information doesn't emerge when context is added to data. The whole concept of data is flawed. It is in IT invention that had nothing to do with how we humans inform ourselves. So PLEASE do not change your original definitions.

Dave,
In contrast to Alex's post, I think the diagram may be a bit too complicated!! There are redundant steps in it.

I was struggling with the diagram, and felt sure I had an improved version. It is the first nice Spring day I've been home this year so I decided to take a nice bicycle ride along the Potomac before posting. Thinking things through I realized I had the same problem I've always had with the DIKW model. As you have said, you dislike the "wisdom" part, since it really does appear to be simply a grading or ranking of knowledge, externally labeling some knowledge that is privileged over other.

My problem is with the "information" part. Is information partially digested data? Or is it simply a qualitative difference between rough, hard to digest data items and stuff that is easier to consume?

So let's start with this - Knowledge is the stuff between my ears and consists of patterns of patterns that are associated by patterns, etc. For example, there are a bunch of patterns associated with 'Dave Snowden' including a name pattern, a relation to the 'Welshman' pattern, etc. When I get a new data item, an input from my sensory system, I map it according to the pattern matching software in my head. This is the process whereby I am creating knowledge, call it Learning, or SenseMaking (I like the multi-ontology ideas that suggest we aren't stuck in a single way of making sense of things but there are multiple ways we Make Sense). The knowledge may be simply recognizing the gentleman offering me a beer as Dave Snowden, but there are other patterns being activated, modified, created, destroyed in the SenseMaking process, such as his facial expression, the way he offers the beer, all of the socially complex patterns that are really hard to desribe but capture the richness of knowledge.

Data Items consist of observables. Some are natural, others are created. Some are simple, such as the crosswalk sign on the bicycle path that tells me in 15 seconds the signal will change (which, when you think about it isn't all that simple!). Others may be bundled into complicated artifacts, things such as Newton's "Principia". Neither is more or less "information". I use patterns I have to process and empattern Data Items I observe (I never have developed the "Understand something written by James Joyce" pattern!).

When we speak, write, etc. we are creating data items that are our attempts to capture and communicate sets of patterns we have. But, as you have mentioned often, something is lost in the process. We create Data Items, but not Knowledge. "Actions" are our interactions with the external world, our creation of data items that consist of speech, writing, facial expressions, driving a car, etc.

So I guess my problem with the diagram is that it confuses several things that seem to be similar things. I propose that The Diagram might be something like Data Items, which via Learning or SenseMaking becomes Knowledge, which is used, stealing John's term since I can't come up with another, via Path Finding to select and perform Actions, which modify the external world, hence creating more Data Items.

Of course, this is actually a continuous process versus a discrete one as suggested by the diagram, but that's a different issue!

Wayne

The linkage of sense making and path finding is crucial for any social service and public service organization. I would like to add a vote of confidence for this diagram, as it provides a better starting point than most to articulate the peculiar issues that all government organizations face.

Government organizations are inherently different than businesses, as Jim Collins explains exceptionally well in his "Good to Great and the Social Sectors." Fundamentally, government organizations are overly constrained, a condition I paraphrase as "everyone can say no and no one can say yes." It sounds simple, but emphatically it is not. As a long time Government employee, I should know better by now. But I find that I am forced to relearn this lesson with a brand new vocabulary every few weeks. I should know better, especially since my job title includes "policy specialist" in it. I could complain, but when I do my boss reminds me that "translation" is not in my job description. The best advice I get is that I should be talking to the other department that is paid to deal with foreign currencies.

From the inside of an organization, whether formal or informal, sense-making and path-finding are co-equal and co-evolving on any timescale that exceeds a few minutes, or the average meeting time. You can't distinguish the two forms from beyond the organization's boundaries, whether in a local group or a large extended community of practice. Bringing in a consultant or starting a knowledge management program perturbs the evolution, and while you gain clarity in one, you lose focus on the other.

Clearly both Mr. Poindexter and Snow have wrestled with these permutations and have tried innumerable options and actions. Despite the Heisenberg-like principle they have shared knowledge and wisdom to arrive at the diagram. I don’t have any knowledge on how they reached their conclusions, although I suspect the Cynefin framework was involved.

It seems that you may be struggling to describe emergent meaning in the intelligence community, or in translation, the co-evolution of information and knowledge in an organization of organizations. If this diagram is a viable starting point, would adding a parallel path-finding and sense-making set of boxes on the inner Iteration side of the diagram adequately create a picture of the desired process?

If the picture, which is the instrument of analysis, becomes clearer, then the boundary names will become suddenly obvious through other sense-making or path-finding tools such as Nonaka’s SMEI framework. In my experience the Cynefin framework has been a useful Rosetta stone for sorting through these complex theories of organizational functions.