Scientometrics, Knowledge Management, and Social Network Analysis

Here are some interesting SNA studies that I found:

Connecting the Dots: Tracking Two Identified Terrorists

Early in 2000, the CIA was informed of two terrorist suspects linked to al-Qaeda. Nawaf Alhazmi and Khalid Almihdhar were photographed attending a meeting of known terrorists in Malaysia. After the meeting they returned to Los Angeles, where they had already set up residence in late 1999. What do you do with these suspects? Arrest or deport them immediately? No, we need to use them to discover more of the al-Qaeda network.

527 Committee Donors (via ire)

In the 2004 presidential election “huge donations of a handful of wealthy liberals named Linda Pritzker, Stephen L. Bing, Peter B. Lewis and George Soros could determine the outcome. Together, they have given more than $26 million to help finance the most extensive get-out-the vote operation in history, the goal of which is to make John F. Kerry president.” These donations were made to 527 organizations. “Named after a section of the tax code, the 527 groups are doing much of the advertising and field work traditionally left to party organizations.” Included with this story is a diagram displaying contributions to Democratic 527s and a list of the biggest donors to these groups.

And finally, 17 ways to visualize the twitter universe via flowingdata.

I attended a Knowledge Management workshop held by the Singapore Civil Service College last Thursday and Friday. The trainer, Christopher Tan, mentioned that Knowledge Management has gotten a bad name among organizations as it was an over-used term by IT companies which were selling IT solutions. IT companies were selling packages of expensive solution which included so many modules that may not be needed by the organizations.

This also leads to the tendencies of organizations in seeing Knowledge Management as a subset of Information Technology Initiative, while in fact IT is only an enabler for Knowledge Management.

Another thing is that Knowledge Management is also often perceived as an initiative to “take away” jobs from people (i.e. if there is a proper knowledge management system, turnover of manpower effects will be very little). This, I think, is true.

I think this may be some of the reasons why Knowledge Management practicioners are not very willing to call what they are doing as Knowledge Management as mentioned by Graham.

According to Carla O’Dell (of APQC), there are 5 stages in a Knowledge Management Roadmap, namely:

1. Getting Started

1.1. Define KM in terms people can relate to

1.2. Identify others to join the cause

1.3. Look for windows of opportunity

1.4. Capitalize on the technology

1.5. Create a compelling picture

1.6. Know your own corporate history

2. Explore and Experiment

2.1. Form a cross functional KM task force

2.2. Select pilots or identify current grass roots efforts

2.3. Find resources to support the pilots

3. Pilots and KM Initiatives

3.1. Fund the pilots

3.2. Develop methodologies

3.3. Capture lessons learned

3.4. Land the results

4. Expand and Support

4.1. Develop an expansion strategy

4.2. Allocate resources

4.3. Communicate and market the strategy

4.4. Manage growth and control chaos

5. Institutionalize KM

5.1. Embed KM in the business model

5.2. Realign the organization structure and budget

5.3. Monitor the health of KM

5.4. Align Rewards and Performance Evaluation

5.5. Balance a Common Framework with Local Control

5.6. Continue the Journey

Do note that stages 4 and 5 are only possible when pilot results (stage 3) are compelling.

The following is taken from Susan Hanley:

What is a taxonomy?
A taxonomy is a collection of relevant topics and subtopics arranged in a hierarchical or networked structure. A library card catalog is a classic example of a taxonomy. The hierarchical structure on Yahoo is another example.

Why is a taxonomy important?
In a portal or content management system, an effective taxonomy helps users to navigate to documents in which they are interested without having to do a search (although, in practice, most studies seem to show that users use a combination of taxonomy navigation and search when both are available). Taxonomies also allow users to see documents in a context, which helps the user assess whether a document is relevant for what they are trying to accomplish.

How do I get started building a taxonomy?
Three key skills are required to effectively build taxonomies. The first, and probably most important, is content organizational skills – a combination of data modeling and library science. The second is some knowledge of the domain to be modeled. The third is knowledge of the end user of the applications that will leverage the taxonomy. In general, when there are trade-offs to be made in taxonomy design, design for the end user of the content, not the contributor.

During visits to other organizations before I started the Knowledge Management Initiative in my organization, I learned that many of the taxonomies developed by external consultants or taxonomy specialists have been very under-utilized, if not forgotten. The taxonomies were undoubtedly looking good, but apparently it is too complex or too alien for the users. So, in developing the taxonomy for my organization, I tried to get as much feedbacks as possible from the people in my organization, making sure that it is not only complete but more importantly understanable and usable.

During KM Singapore 2008, I met Graham who stressed the need to keep the taxonomy simple, he also shared that it is not even necessary to call taxonomy as “taxonomy”, why not simply call it as it is: folder structure. Of course taxonomy is not only a folder structure, but at least this will prevent people from considering it as too complex of a thing.

I earlier mentioned about Carloz Mendez presentation on open source knowledge management tools.

One of the most common reasons why organizations are reluctant to embrace open source software are lack of support and continuity as compared to proprietary software.

The 451 group has just published a report supporting the notion about the lack of continuity of pure open source software business model. One of their conclusion is that there are few vendors generating revenue from open source software that are following a pure open source approach when it comes to developing all of their code in the open and licensing all of their software under open source licenses. 

This throws some light to the “lack of support and continuity” reason by organizations. Why is this so? Because many organizations outsource their IT-related development to vendor and the potential vendors will very likey be offering proprietary software since it is more profitable for them.

More write-ups at readwriteweb.

I found some useful resources that teach network analysis on the web, here they are:

• Steve Borgatti’s course
• Methods for Social Network Analysis by Tom A.B. Snijders
• Social Network Analysis Graduate Course by Barry Wellman
• Organizational Network Analysis by Rob Cross
• Program on Networked Governance (Compilation by JFK School of Government)
• Compilation by KM4DEV (Knowledge Management for Development Site)
  

Knowledge can be categorised as explicit knowledge and implicit knowledge. So, when we hear the word “knowledge management system”, we’ll expect to see the system to be able to capture both the explicit and the implicit knowledge. However, this is not necessary the case.

Carloz Mendez drawed the diagram of what tools can be used to capture both type of knowledge as shown below:

Rather than suggesting certain proprietary knowledge management system, Mendez has listed some open source tools that can be used as knowledge management system as can be seen in his presentation.

In 2000, Ponzi and Koenig wrote a paper titled Knowledge management: another management fad? with the following abstract.

Abstract

Knowledge management is a subject of a growth body of literature. While capturing the interest of practitioners and scholars in the mid-1990s, knowledge management remains a broadly defined concept with faddish characteristics. Based on annual counts of article retrieved from Science Citation Index, Social Science Citation Index, and ABI Inform referring to three previous recognized management fad, this paper introduces empirical evidence that proposes that a typical management movement generally reveals itself as a fad in approximately five years. In applying this approach and assumption to the case of knowledge management, the findings suggest that knowledge management is at least living longer than typical fads and perhaps is in the process of establishing itself as a new aspect of management. To further the understanding of knowledge management’s development, its interdisciplinary activity and breadth are reported and briefly discussed.

Ponzi and Koenig used article counting technique where the article counts were retrieved from Science Citation Index, Social Science Citation Index and ABI Inform. They counted the number of articles from 1991 to 2001 with the following result.

So, I was curious to know how the knowledge management field has evolved in the past 5 years from 2002 to 2006. Hence I retrieved article counts from both Science Citation Index and Social Science Citation Index (not ABI Inform). The result is as follows:

So, it can be seen that Ponzi and Koenig’s finding still holds now. However, it will be interesting to see whether the number of articles on knowledge management will stabilize around 500 (Article counts in 2007 as per 28 June 2007 is 153).

It’s simply a way to represent relationships using nodes and ties. The nodes represent the actors, and the ties representing the type and strength of the relationships among the actors.

Network analysis has been growing to the terms where there are businesses concentrating on providing network analysis, such as this one.

Social Network Analysis can be used in knowledge management initiative, specifically if you are in the cartographic school of knowledge management. SNA can be used to know who are the key people in the organization and how important are them for the organization.

According to Boisot, 3 states where required in the structuring and sharing of knowledge:

1. Codification: The minimum number of bits of data that need to be processed in order to distinguish between categories when categorizing an event
2. Abstraction: The minimum number of categories required to apprehend an event for a particular purpose
3. Diffusion: The number of data processing agents in a target population that have access to a given item of data within a certain time frame.   

In order to cultivate a knowledge management initiative, it is important not only to make sure that there is a good platform to codify (store) and diffuse (share) knowledge, but also to make sure that the knowledge stored can be comprehended by other people that have access to the knowledge. IN other words, the knowledge need to be abstracted. One of the things that can be done is to give proper metadata for each knowledge shared, or to build a taxonomy and tag the knowledge based on the taxonomy. One blog that I find useful in keeping uptodate with how taxonomy is being implemented is taxonomy2watch.