Synthesizing Huber's Problem Solving and Kolb's Learning Cycle: A Balanced Approach to Technical Problem Solving

Article excerpt


One of the most important skills in Internet Protocol (IP) network design and configuration is subnetting, that is, dividing an IP network into smaller networks in a hierarchy, parts of which can be managed separately. Subnetting provides several advantages, such as improved security, ability to manage network resources locally, decreased overall need for unique IP addresses, decreased size of routing tables in core Internet routers, and reduction of broadcast traffic and thus improved utilization of available bandwidth. Without subnetting, it would be impossible to manage organizational IP networks of any significant size. See Kamis and Topi (2007) for a primer on subnetting.

Solving subnetting problems requires an in-depth understanding of IP addressing mechanisms. Once grasped, solving a subnetting problem is not difficult, but for less experienced network designers and administrators, it is often a challenge. As most instructors of networking courses can attest, subnetting and troubleshooting of subnetting problems are among of the most difficult challenges for individuals who are new to networking. (Cigas, 2003; Greca, Cook et al., 2004). A substantial stream of research indicates that intensive laboratory practice is necessary for individuals to solve networking problems in general and subnetting problems in particular (Cigas, 2003; Corbesero, 2003; Greca, Cook et al., 2004). With subnetting, as with programming, one cannot solve problems by reading a book or listening to lectures. Research has shown that an active or problem-based learning approach works better (Roussev and Rousseva, 2004; Whittington, 2004).

To investigate the solving of subnetting problems, we considered two distinct streams of literature: 1) Kolb's Experiential Learning Cycle (Kolb, 1976; Cook and Swain, 1993; Cornwell and Manfredo, 1994; de Ciantis and Kirton, 1996) with the abstract/concrete distinction (Reeves and Weisberg, 1993a; Reeves and Weisberg, 1993b), and 2) problem solving in general (Newell and Simon, 1972) with emphasis on Huber's approach to problem solving (Huber, 1980).

We chose Kolb's cycle because it is a comprehensive and influential model of experiential learning (Kolb, 1976) and because problem solving can be primarily mastered through experiential learning. Kolb's path-breaking work conceived of experiential learning as a four stage cycle, which, ideally, everyone would master. Within his seminal paper, setting that cycle as the ideal, Kolb claimed that few people could master the complete cycle. Instead, he argued that individuals tend to specialize in one of the four stages or styles of learning. Fortunately, organizations could manage an assortment of individuals, who, collectively, would constitute an organizationally balanced "meta-brain". Rather than follow Kolb's conclusions regarding learning styles, however, we focused on Kolb's distinction between abstract and concrete representations of problems and on an individual's movement between these representations as part of the learning process. Reeves and Weisberg have found that learning by analogy increases with the use of concrete rather than abstract representations (Reeves and Weisberg, 1993a; Reeves and Weisberg, 1993b). Sadoski et al (1993) conducted several experiments that showed that concreteness, i.e., the ease of imagery, had strongly positive impacts on comprehension and recall of information (Sadoski, Goetz et al., 1993).


Our domain of interest was network subnetting, an instance of technical problem solving. Problem solvers learn a set of rules and apply these to solve a network subnetting problem. Problem solving has been studied in linear, well-structured domains, such as chess (Chase and Simon, 1973; Charness, 1992) and in realistic, complex domains, such as the process control industry (Patrick, Gregov et al., 1999). Studies of problem solving can be descriptive (Srinivasan and Te'eni, 1995) or prescriptive (Patrick, Gregov et al. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.