Dave Snowden

Good judgement comes from experience. Experience comes from bad judgement

RSS Feed

Now the origin of this model is Russell Ackoff, one of the giants of systems thinking (not to be confused with complexity) who created five categories of the human mind.

  • Data, seen as symbols (IT sees this as an abstraction of signals)
  • Information, seen as useful data answering who, what, where and when questions
  • Knowledge, answering how questions through the application of both data and information
  • Understanding, as the appreciation of why
  • Wisdom, or evaluated understanding

Now while this still has a semblance of hierarchy it critically focuses on different purposes rather than abstractions. It also includes understanding, the omission of which by most IT orientated advocates of DIKW is at least ironic.

 

 

 

 

 

 

Bellinger et al both summarise the Ackoff model, but also seek to simplify it to DIKW as per the adjacent illustration which I take from their paper. Readers of this blog may remember mby earlier publication of a John Poindexter model which went Data to Information to Knowledge to Options to Action

The Bellinger model becomes meaningful when you look at the definition of the transitions between the stages, the dynamic rather than the static aspects of the model. Understanding relationships, patterns and then principles makes a sort of sense. The Poindexter model also looked at the dynamics, namely analysis, sense-making, path-finding and execution. I said at the time that I felt there was some millage here, but not in a linear model, and that combining some of these ideas with the OODA loop could make sense.

I am still working on that, and incorporation of the Ackoff original. If nothing else as a transitionary model between the world of systems thinking and complexity. I am still not there, but there are two things of which I am convinced.

Neither Ackoff or Poindexter make the mistake of seeing each layer as a progressive set of abstractions. Instead they see different types of process in play. Understanding dynamics rather than static models is key, but the IT community has a preference of things being in their place and not out of place.
The pretension of wisdom management should be rigorously attacked whenever it rears its ugly head. If that was not bad enough the ActKM list also produced a proposal for enlightenment as the next stage on from wisdom. This is pretension combined with pomposity. However to move to the transcendent state of enlightenment may be away of avoiding learning from the consistent failure of the DIKW model in practice.
In the earlier Turing Test exchanges on ActKM on this subject I finally, after due warning resorted to poetry. The failure of the protagonist to appreciate this use proved that it was a computer (or as Patrick Lambe suggested a human trying to be a computer) on the other end of the exchange. More enlightened members of the listserv understood the reference to the greatest of the Vogon threats. My sign off on that exchange was to suggest that the protagonists major intestine should revolt before he said any more. You need to be a Hitchhiker aficionado to get this, and to understand my final request for him to cease:

Or I will rend thee in the gobberwarts with my blurglecruncheon,
See if I don't!

Top