Seite - (000286) - in Knowledge and Networks

283 repeatedly collaborated in concrete projects in the past.4 In professional software engineering, particularly in the IT industry, it is standard practice to modularize common project outcomes and reuse them in other projects. In the software industry joint projects can become a key process for knowledge imitation because developed project solutions such as programs, code, or parts of programs and code—so-called code snipplets—can easily be reused in other projects. The imitation network was the most highly fragmented of all the activities. Slightly more than half of the members had already used concepts, plug-ins, or code sections from other network members for their own operating purposes. This imita- tion allowed the companies to save development time and to make solutions to problems available in the company. Not only was friendly imitation as a network activity at a moderate level in the network, it was also the activity with the fewest relationships, the lowest density, and a comparatively low number of average rela- tionships per network member (Table 13.1). The size and similarity of the member companies were statistically unrelated to engagement in imitation among network participants. In particular, the exchange of knowledge and the cooperation on joint projects were the strongest network activities in Comra.de. They included the larg- est number of members, the greatest density, and the largest number of relationships. What was the position of the deviant firm that had repeatedly broken the conven- tions of the network? If the firm had really been sanctioned with disapproval and exclusion from the communication and cooperation relationships, that status would be reflected by a relatively peripheral or even isolated position in the network. Indeed, according to its own response and the responses of the other members to the items in the survey, the deviant firm was largely isolated from any activity. It did not lend any employees to other members, receive any solutions from other companies, 4 To rule out other explanatory factors, we included many additional variables, such as the entre- preneurs’ joint history, capital participations between member companies, and company prestige. Later analysis showed all these variables to be insignificant, however, so we do not address them in depth in this chapter. Table 13.1 Four forms of cooperation in Comra.de Variables Number of network components Network densitya Number of relationships Relationships per member (mean) Imitation 9 0.04 17 0.85 Knowledge exchange 7 0.10 38 1.90 Employee-lending 8 0.07 25 1.25 Project collaboration 4 0.09 35 1.75 aNetwork density is calculated by dividing the number of observed relations by the number of pos- sible relationships. Adapted from Glückler et al. (2012, p. 177). Reprinted with permission of Springer 13 Connectivity in Contiguity