Brokerage, Boundary Spanning, and Leadership in Open Innovation Communities

From P2P Foundation
Jump to: navigation, search

Article: Brokerage, Boundary Spanning, and Leadership in Open Innovation Communities. By Lee Fleming (Harvard Business School), David M. Waguespack (University of Maryland)


Contact via And



“What types of human and social capital identify the emergence of leaders of open innovation communities? Consistent with the norms of an engineering culture, we find that future leaders must first make strong technical contributions. Beyond technical contributions, they must then integrate their communities in order to mobilize volunteers and avoid the ever-present danger of forking and balkanization. This is enabled by two correlated but distinct social positions: social brokerage and boundary spanning between technological areas. An inherent lack of trust associated with brokerage positions can be overcome through physical interaction. Boundary spanners do not suffer this handicap and are much more likely than brokers to advance to leadership. The research separates the influence of human and social capital on promotion, and highlights previously unexamined differences between brokerage- and boundary-spanning positions. Longitudinal analyses of careers within the Internet Engineering Task Force community from 1986–2002 support the arguments.”


“We define an open innovation community as a group of unpaid volunteers who work informally, attempt to keep their processes of innovation public and available to any qualified contributor, and seek to distribute their work at no charge.”

Open Innovation Communities

The authors start by reminding us of the advantages and achievements of open innovation communities:

“They have dramatically changed our conceptions of how innovation can and should be managed and have prompted calls for new theories of innovation (von Hippel and von Krogh 2003). Open community development methods appear superior to proprietary efforts on some measures (Kogut and Metiu 2001, Mockus et al. 2002). Open source operating systems challenge the world’s most powerful software firms, and proponents of community innovation are extending the model into new contexts such as gene transfer technology (Broothaerts et al. 2005), medical innovation, crime solving, textbook and encyclopedia publishing, education, space exploration (Goetz 2003), and communities in developing countries, for which software customization in local languages remains cost prohibitive (Economist 2003). Open communities have spawned some of the most important technological breakthroughs of our era, including Web browsers, e-mail, and the Web itself. The very protocols that enable different Internet technologies to work together emerged from innovation within a voluntary, nonproprietary, and open innovation community.”

On Leadership

Leadership in Open Innovation Communities

By Lee Fleming (Harvard Business School), David M. Waguespack (University of Maryland):

Rivlin (2003) illustrates how Linus Torvalds (the original author of LINUX) realizes that his authority is technically derived, tenuous, and constantly in need of collective reaffirmation:

His hold over Linux is based more on loyalty than legalities. He owns the rights to the name and nothing else. Theoretically, someone could appropriate every last line of his OS [operating system] and rename it Sally. “I can’t afford to make too many stupid mistakes,” Torvalds says, “because then people watching will say, hey, maybe we can find someone better. I don’t have any authority over Linux other than this notion that I know what I’m doing.” He jokingly refers to himself as “Linux’s hood ornament,” and he’s anything but an autocrat. His power is based on nothing more than the collective respect of his cohorts.”

How Leaders are chosen at IETF:

“One of the principal differences between the IETF and many other standards organizations is that the IETF is very much a bottom-up organization. It is quite rare for the executive leaders within the IETF, the IESG (Internet Engineering Steering Group) or the IAB (Internet Architecture Board) members, to create a working group on their own to work on some problem that is felt to be an important one. Almost all working groups are formed when a small group of interested individuals get together on their own and then propose a working group to an Area Director” (Bradner 1999, p. 49).

It is much the same with other open innovation communities.

Today, an open source software development project is typically initiated by an individual or a small group with an idea for something interesting they themselves want for an intellectual or personal or business reason” (von Hippel and von Krogh 2003, p. 211). This process for Linux was described thus: “ ‘The lieutenants get picked—but not by me,’ explains Torvalds. ‘Somebody who gets things done, and shows good taste—people just start sending them suggestions and patches’ ” (Hamm 2005, p. 66). See also Rosenkopf et al. (2001). (

Brokers vs. Boundary Spanners

Leadership is a crucial variable:

“Despite their bazaarlike, egalitarian, argumentative, unplanned, chaotic appearance, open innovation communities rely heavily on strong leadership to function effectively and to resist splintering, forking, and balkanization.”

They introduce two kinds of leadership: brokers and boundary spanners:

“Brokers can span boundaries, but not all boundary spanners broker. This unexamined difference leaves a variety of unanswered questions. For example, what are the different mechanisms by which the occupants of the two positions attain leadership positions? Burt’s early descriptions characterize brokers as calculating and politically savvy operators, while Allen and Tushman characterize boundary spanners as well respected guardians who redirect crucial information, both within and outside the firm. If these characterizations hold any validity, then colleagues of the broker and boundary spanner will surely hold different perceptions about individuals in each position. Colleagues will be less likely to trust a broker (Coleman 1988, Burt 2001b), and this lack of trust will be exacerbated within open innovation communities, which are inherently wary of balkanization and cooptation by commercial interests.

The most effective strategies and behaviors of aspiring leaders will therefore be different for brokerage- and for boundary-spanning roles.

More generally, open innovation communities provide an opportunity to develop theories of human and social capital in a novel context that lacks pecuniary incentives, hierarchical authority, and formal structure. Leadership in such communities depends more on the trust and mobilization of peers than on approval of superiors. To wit, members cannot be fired or forced to participate in any activity, nor can they be compelled to pay attention to any other member. Ascendancy in such relationships relies purely, to borrow a phrase from politics, on “the power to persuade”

Their research concludes that boundary spanners are more successful:

“Individuals who broker work collaborations are more likely to assume leadership, but the effect is strongly contingent on physical presence within the community, a consequence of the diminished trust inherent in brokered social contexts. Consistent with the argument that they must overcome lack of trust, brokers also encounter difficulties when they attempt to span technological boundaries within the community. Boundary spanners, in contrast, do not suffer from a lack of trust and are more likely than brokers to assume leadership positions. In summary, future leaders are most likely to be individuals that make a strong technical contribution from a structural position that can bind the community together.”

More details on boundary spanners:

“The boundaries within open innovation communities, like those in private firms, usually correspond to the interfaces between technological subsystems (Henderson and Clark 1990). Each boundary demarcates a distinct technological area or module. The boundaries are defined directly by leaders’ architectural decisions and then implemented by followers’ choices of where to volunteer their efforts. Torvalds, for example, delegates responsibility for different modules to a handful of trusted lieutenants (Hamm 2005), each of whom reviews submitted code, accepts the best, and works with Torvalds and the other lieutenants to resolve issues that cross technical boundaries. Bradner (1998, p. 5; 2003b) describes how individuals who aspire to leadership in the IETF must define their technological boundaries relative to other community efforts as part of their BOF proposals: “Is there a good understanding of any existing work that is relevant to the topics that the proposed working group is to pursue _ _ _ and, if so, is adequate liaison in place?” Although technical and social boundaries correlate, brokerage and boundary spanning remain distinct roles. Nothing prevents a boundary spanner from being the only engineer, in which case she would be a social broker. Alternately, she could be one of the many engineers who collaborate across a boundary, in which case the boundary spanner is not a social broker at all. Open innovation communities are well aware of these issues and value individuals that can span technical boundaries (Bradner 2004). Boundary spanners should be more creative and able to call on more-diverse resources, as explicitly recognized in the expectation that leaders promote “cross-pollinating between groups” (Davies 2003, p. 13). In addition to providing the integrating advantages of social brokerage, boundary spanners are in an even stronger position to control the potential for community forking, which generally occurs when alternative technical solutions attract dedicated coalitions that refuse to compromise or resolve incompatibilities.”


“Both brokering and boundary-spanning roles greatly increase the likelihood of leadership points to the importance of social positions that can unite open innovation communities. We argued that trust does not come easily to community members who fear cooptation by commercial interests or forking over technical disagreements. Because brokers by definition contrive less cohesive and less trusting contexts, the probability that they will assume leadership roles remains highly contingent on building trust with community members. We argue that aspiring leaders can build trust through physical attendance and, consistent with this argument, find a positive interaction with physical attendance. Also consistent with our emphasis on trust in open innovation communities, brokerage and boundary spanning demonstrated a negative interaction, indicating that brokers who span boundaries remain at a disadvantage. While brokerage alone demonstrates positive influence on becoming a leader, boundary spanning demonstrates a much stronger effect. Finally, we did not observe a contingent relationship between boundary spanning and attendance. Our results emphasize the importance of intermediary and integrating roles—for brokers within technological boundaries, and for boundary spanners across cohesive technological boundaries."

More Information

  1. Open Design Communities
  2. Boundary Spanners
  3. Leadership