Toward a Theory of Remixing in Online Innovation Communities

Within online innovation communities, remixing (i.e., the community's use of an existing innovation as source material or inspiration to aid in the development of further innovations) is an interesting form of knowledge collaboration. This study investigates an open theoretical question: Why are particular innovations remixed by online innovation communities? Some innovations languish, while others are widely remixed. Community members (even those unknown to the innovation's creator) may remix, taking the source innovation in directions the original innovator may have never imagined. Within online innovation communities, remixing is not bound by some of the constraints to knowledge collaboration found in more traditional environments. To address our research question, we begin with variables constituent to innovation diffusion theory, essentially remixing this long-established theory to predict cumulative remixing in online innovation communities, using arguments grounded in the user innovation and learning literatures. We also consider two forms of communication that are relevant to knowledge sharing in online communities (online community interaction and front page presence). Regression analysis (using data pertaining to 498 3D printable innovations) shows that community interaction is highly influential in determining which innovations are remixed by the community. Conversely, the innovation having a presence on the community's front page does not have a significant effect on remixing. Observability has an inverse-U-shaped relationship with remixing; this suggests the value placed on experiential learning. Fuzzy set qualitative comparative analysis (fsQCA) is used as a supplementary analysis technique (with robustness testing), and the results are largely consistent with regression findings but offer interesting insight into innovation configurations that consistently result in remixing from the community. Within specific configurations, fsQCA results indicate a contingent effect of observability and complexity; that is, under certain conditions, complex innovations are more likely to be remixed by the community.