Thursday 14 March 2013

Explicit knowledge is only valuable if it is accurate

 I have had a knowledge problem recently, which pointed out, in a very graphic way, the problems with inaccurate explicit knowledge.

Let me explain.

My house is situated on the side of a valley, which captures drainage from all the surrounding fields and feeds it into a small stream.  When we have very heavy rain, my garden can flood temporarily for a day or two.  In this last winter we had very heavy rain indeed; the wettest winter on record.  The garden flooded, but this time the flood did not subside.  Our lawn was covered in water for five or six weeks (see upper picture), and the flood took a very long time to subside.  It became obvious that the main drain, which carries water away under the road, had become blocked somewhere along its length.  We would need to dig the drain out, and unblock it.

So what is this got to do with knowledge? 

In essence, it was a knowledge problem, because I did not know where the drain went.  I could see one end of it, at an inspection chamber, and the other end in my neighbour’s garden 100 metres away on the other side of the road.  I had no idea where, in the intervening 100 metres, I should start digging.  The only clue I had was a small sink hole that had developed, taking a limited flow of water.

So, as any good knowledge manager should, I started looking for the local tacit knowledge.  The first thing I did was to phone my neighbour, who had sold me the house and who still farms the surrounding fields.  “I can’t remember where the drain goes”, he replied, “but I have got a map of the drains, if that will help”.  Even better!  Explicit knowledge! His tacit knowledge had faded, but he had retained the explicit knowledge.

The map (the second picture to the right) showed the drain heading westwards from the inspection chamber (marked 2 on the picture) to another inspection chamber, now buried (marked 1 on the second picture), then heading northwest for the road, and nowhere near my sinkhole (marked S  on the picture).

Armed with this explicit knowledge, I “knew” exactly where to dig. I located and unearthed  inspection chamber number 1, saw the drain heading north westwards as it should, and called in my local builder and his digger-driver to dig it up.

Of course the map was wrong. So much for explicit knowledge.

We had to cut down 10 trees and dig five deep trenches over a course of two days before we finally located the drain.  As the third picture shows, it left inspection chamber number 1 heading NW, but then took a sharp turn and headed northwards under the garden fence, right next to my sinkhole (the third picture).

So what is the lesson for knowledge management?

The lesson is that the explicit knowledge that was preserved was, as is so often the case in industry, the plan and not the reality.  The original design for the drains had been preserved, but nobody had documented and preserved the “as-built” design.  Picture number 2 shows the original design, while picture number 3 shows the as-built reality.  The lesson is that if we are to preserve explicit knowledge, we MUST preserve explicit knowledge that shows reality; that shows the as-built structure and not the planned structure.

The cost to me was an extra day and half of digger hire and labour rates, 10 dead trees, and a garden that looks like the Somme at the height of world war one (picture 4).  Lets say £500 in total. The cost in industry could be very much higher, if the explicit knowledge does not reflect reality.

However we have now located the drain, which was choked with willow roots, and cleared it out.  I have amended the plans to show where the drain really goes.  We will add additional inspection chambers to trace the course of the drain and to act as rodding points, and build them so that they do not get buried.  This way the knowledge of the drain's course will be captured not just on paper, but in stone and mortar.


  1. This is one of the reasons why when I work with organizations on their explicit/documented knowledge I encourage them to initiate a lifecycle process to go with it. The process covers everything from how the knowledge is captured in the first place to how it is maintained and retrieved, to what is done with it at end of life. A lot of organizations miss this step and 3-5 years into their KM program "can't find anything" and blame the technology, but it's not the technology, it's the missing lifecycle process that's the problem.

  2. Look on it as gaining firewood you otherwise would not have...and there may be re-enactor groups that will be happy to lease your recreation of the Somme on weekends...

    I think that the actual issue, Nick, is that you acted on unvalidated OIL, even if from an authoritative source and given with the best of intentions (one assumes - did you push him on the price?) to assist you in your quest. Perhaps some further inquiries may have been in order before getting the digger in?

  3. You're right Simon. There is also something about the precision of a map, that makes you think it is authoritative (and indeed, it was validated by leading me to the inspection chamber, but let me down immediately after). That's another lesson - just because its a large sheet of paper with authoritative-looking markings on it, doesn't mean it's correct. The question I should have asked was "is this the design map, or the as-built map".

  4. This comment has been removed by a blog administrator.

  5. "Only men's minds could have unmapped into abstraction such a territory."
    Norman MacCaig (14 November 1910 – 23 January 1996)

  6. That reminds me of a lesson learned I came across once: "The As-Built was not as it was built". :-)


All comments welcome! I review all comments to remove spam messages, so there may be a short wait until your comment is visible.