User and developer mediation in an Open Source Software community: Boundary spanning through cross participation in online discussions

{{Summary
 * title=User and developer mediation in an Open Source Software community: Boundary spanning through cross participation in online discussions
 * authors=Flore Barcellini, Françoise Détienne, Jean-Marie Burkhardt
 * tags=open source software community, cross-participants, boundary spanners, distributed design, role emerging design, open source, open source software, design-use mediation, role analysis
 * summary=This paper distinguishes 5 different roles in open source projects:
 * 1) project leader
 * 2) administrators
 * 3) developers
 * 4) active users (who participate in discussions, report bugs, and propose new modules)
 * 5) 'passive' users who may lurk but do not discuss or document the software

Can users who are neither administrators nor developers impact the design choices of an open source project? This paper investigates how users participate in and contribute to the design process, focusing on boundary spanners, who bridge the user community and developer community. Boundary spanners' contributions to the "management and knowledge sharing activities" are also investigated. "Despite the idealistic picture that users may intervene freely in the process, we will question whether users who are neither administrators nor developers in the core Python community can really have an impact on the design choices and decisions.

Methodology is field interviews and analysis of two mailing lists (developer-oriented and user-oriented). The focus is on cross-participants, who "take part in same-topic discussions, occurring in parallel in both mailing-lists.", with the hypothesis that cross-participants are either key participants or boundary spanners. Messages were coded along two lines: activities and references, as follows. Tables provides the number and percentage of references (in total) by role on each mailing list, and some analysis is given regarding the roles. (For example, that cross-participants "tend to provide more references about the application domain" and that they transfer "references about programming users to the user community"). The paper also observes that "the cross-partici- pants are central in the social network linking the user community and the developer community. On the user side, the cross-participants tend to quote and be quoted by the users. On the developers side, the user-champion, a specific cross-participant, tends to quote and be quoted by the administrators (including the project leader) and the developers of Python."

Activities

 * 1) Coordination
 * 2) Synthesis
 * 3) Decision
 * 4) Social relationship