Yefim,
That’s a good question. We recently did a project with Procter & Gamble’s R&D division to “extract knowledge” from departing knowledge workers, and capture it, really index it to a shared ontology. What we found were a couple of things:
· A great deal of the knowledge they were interested in was already captured (in documents) but barely findable by people in the same research group and impossible for people in other research groups to find (this was both a technology problem as well as a nomenclature problem)
· We did build a pretty good ontology that would allow one to express most of the specifics in a given domain using terms that would be shared throughout the R&D function. Our sponsor built a wiki around this ontology and we were able to demonstrate the small bit of real data that we had collected.
· In the process I came across and read quite a good book “Deep Smarts” by Dorothy Leonard and Walter Swap, They had several observations that pertain here:
o It may be that much knowledge is explicit, but accessing and using it well is very often tacit
o Tacit knowledge needs to be transferred via some sort of apprenticeship program
o One of the more productive ways to intentionally develop knowledge is through guided experiments and guided practice
· Another book that I just started is “Pragmatic Thinking and Learning: Refactor Your Wetware” where Andy Hunt retraces some of the Dreyfus brothers work on how people acquire skills. Seems to be pretty broadly applicable. This approach suggests that we have to go through five stages in skills acquisition as we move from novice, to advanced beginner to competent, to proficient to expert and the strategies to go from one level to the next vary greatly. In the lower levels explicit knowledge is going to be more useful, but in the higher levels it often gets in the way.
I like both books. I don’t think there is an easy answer. I think we need to make as much knowledge explicit as we can to help the great number of people move up from beginnerhood, but at the same time we need to recognize that the higher skill levels need tacit knowledge that they can only get through practice and tutoring.
Ciao
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Zhuk, Yefim
Sent: Thursday, May 26, 2011 8:11 AM
To: [ontolog-forum]
Subject: Re: [ontolog-forum] "tribal knowledge"
Mathew,
This is a good story (reminded me multiple “outsourcing” stories J).
Yes, it is great to have an expert who can clean up the mess.
The problem starts in between two experts when they participate in a complex process.
Very often there is a gap between what was delivered by one SME and what was needed/expected by another.
Some information is almost always omitted partially because it was a “tribal knowledge” of one and assumed to be known by others.
We constantly re-discover “stuff” in the long meetings and phone conversations. While it is part of a “normal” process, there are ways to significantly reduce this information gaps and the need for re-discovery.
These informational gaps are especially visible when we are working on ontological models. Quality of information and information completeness are pre-requisites for such work.
I wonder if there are any statistical data that would quantify percentage of the captured knowledge versus “tribal knowledge” in a corporate environment.
Thank you,
Yefim (Jeff)
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Matthew West
Sent: Thursday, May 26, 2011 5:05 AM
To: '[ontolog-forum] '
Subject: Re: [ontolog-forum] "tribal knowledge"
Dear Yefim,
Let me tell you a story about knowledge management.
Some managers were demanding that their knowledge management team captured all the knowledge about the business they were in (construction) so that the business could be largely deskilled, with unskilled labour following processes as defined in the manuals. The knowledge manager took them down to the local college and gave them several books on plastering, gave them the materials, showed them a blank wall, and asked them to plaster it. Not surprisingly after several hours all they produced was a mess. The knowledge manager returned with an expert plasterer who cleaned up the mess and had the wall perfectly plastered in about 5 minutes.
Regards
Matthew West
Information Junction
Tel: +44 1489 880185
Mobile: +44 750 3385279
Skype: dr.matthew.west
matthew.west@xxxxxxxxxxxxxxxxxxxxxxxxx
http://www.informationjunction.co.uk/
http://www.matthew-west.org.uk/
This email originates from Information Junction Ltd. Registered in England and Wales No. 6632177.
Registered office: 2 Brookside, Meadow Way, Letchworth Garden City, Hertfordshire, SG6 3JE.
From: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] On Behalf Of Zhuk, Yefim
Sent: 25 May 2011 19:12
To: [ontolog-forum]
Subject: [ontolog-forum] "tribal knowledge"
I’d like to ask for your help.
Maybe some of you can point me to some article/research that would quantify “average” percentage of information that is never captured, so called “tribal knowledge”.
One of the directions I am working on is related to improving information quality. I am using conversational approach and semantic support to making information more precise and closing the informational gaps. I’ll talk more about this at the SemTech.
I am looking for some statistical numbers and would highly appreciate your help.
Thank you,
Yefim (Jeff)
This E-Mail has been scanned for viruses.