The Art of Knowledge
Knowledge modeling for the preservation of institutional memory
This is another article off of Emerald. John W. Coffey and Robert R. Hoffman wrote this one. I'm also a big fan of this article because it sites R.C. Holmes twice.
- This article presents a description of an iterative approach to the elicitation and representation of organizational knowledge called PreSERVe, which stands for prepare, scope, elicit, render, and verify. The method involves an initial process of preparing for knowledge elicitation, followed by an iterative process of assessing the scope of the endeavor, knowledge elicitation and rendering, and, verification.
That sounds great, "How to extract knowledge." I really hope this talks about a really awesome system.
- What ensued were empirical evaluations of alternative knowledge elicitation (KE) methods and attempts to provide software support for the knowledge acquisition (KA) process. It was recognized that the acquisition of expert knowledge was a process having a number of steps, including knowledge elicitation, knowledge representation, implementation, and then validation or verification. Coinciding with these developments was a growing recognition that the technologies of KE and KA held promise for helping organizations cope with the loss of institutional knowledge. We now find ourselves in an era of rapidly shifting corporate and government workforces, and the preservation of institutional knowledge has become a widespread concern. Institutional memory loss is a significant problem that can impact an organization's ability to advance its mission successfully, its ability to avoid making the same mistakes it made in the past, and its ability to leverage the accomplishments of departing employees.
Ok so when people leave, yeah you lose all their knowledge and sometimes have to start back over at square one, or at the least go through a short delay. Did anyone else get really excited by the word "mission"? I also like when people say "missile" instead of "bullet". Anyway, I totally agree that trying to retain information that is inside humans is vital to any endeavor. It's vital from the apple picker in New Hampshire to the rocket scientist in Florida. The knowledge needs to be passed from one to the next, and sometimes without extensive training, or at least optimally, without extensive training or retooling.
- Various methods for attempting to retain Knowledge...
(1) Knowledge acquisition methods can consume large amounts of time of busy experts, and take them away from their main tasks. It is difficult to convince time-pressed employees to record information in an ongoing fashion, which mediates against the effectiveness of approaches that require ongoing collection of information.
(2) Exit interviews may yield useful information, but they are more likely to be brief and superficial or unworkable if the employee is exiting under less than desirable circumstances. When more exhaustive KE is attempted, as in the method of "oral histories" (Paris, 2001) large quantities of information can be acquired, but such information may not be in an easily used form (i.e. it may be extended text or video), making the needed information difficult and time-consuming to access.
(3) The search for information in large corporate archives is difficult for a host of reasons, many having to do with deficiencies in indexing. Attempts to automate this process have helped to spawn an entire field called data mining.
I have an idea. How about a system that people feel comfortable with and that they can interact with on an everything basis. Now obviously this won't be the only method for the retention of another's abilities but it's a start. I don't like how they envision a system for tracking people's skill sets as totally manual and time consuming. I think a system can be developed that would be intuitive to a person's interaction with their job and be able to mine skills and information wihtout being intrusive.
- The problem of preventing the loss of expert knowledge - knowledge preservation - is pervasive today and will only worsen as the post-war "baby boom" generation approaches retirement age. Knowledge elicitation and modeling can be combined to serve as one in a range of approaches to address this problem. This paper presents a description of the PreSERVe method (prepare, scope, elicit, render, and verify) of knowledge modeling. This method starts with an initial phase of preparing for a knowledge modeling effort, followed by an iterative process of examining scope, eliciting knowledge, rendering that knowledge in a computerized form, and verifying the knowledge with the expert.
- One unanticipated result of this effort was the discovery of how willing the experts were to discuss their work. These folks had worked in the area of launch vehicles for their entire careers and they were still excited about the opportunity to discuss what they knew and what they had learned. Their knowledge was available for the capture; it was only an issue of ensuring that someone would take the initiative to elicit it in a principled way.
This the diagram that they came up with after the process of trying to retain or extract or preserve information from a project.
When it first started off I really liked the tone of this article. It cut right to the point and made specific examples of processes that were well documented. But the more I read the more confused I got and the less I cared about it. So I went back and read it again and a little flag went up in the back of my head that basically said, "Nope...won't work". I think it could work if you have 25 days to really preserve information and be followed around by guys in white coats asking you questions all the time. That would work if the world had only one process. But unfortunately there are countless processes that can not be taken into consideration by a team of people over a month and then have very little conclusion when they are done. I think the idea is great, "How to extract knowledge from professionals?" That's a wonderful and noble goal. I don't think this way will work for anyone other than NASA though. Maybe vertical solutions are the way to go, I mean last night I read an entire article about how CRM is now going back to the Vertical solution with their products.