Selective revealing in open innovation processes: The case of embedded Linux

This paper provides a quantitative study (N = 268) of patterns of free revealing of firm-developed innovations within embedded Linux, a type of open source software (OSS). I find that firms, without being obliged to do so, contribute many of their own developments back to public embedded Linux code, eliciting and indeed receiving informal development support from other firms. That is, they perform a part of their product development open to the public-an unthinkable idea for traditionally minded managers. Such openness obviously entails the challenge of protecting one's intellectual property. I find that firms address this issue by revealing selectively. They reveal, on average, about half of the code they have developed, while protecting the other half by various means. Revealing is strongly heterogeneous among firms. Multivariate analysis can partly explain this heterogeneity by firm characteristics and the firm's purpose behind revealing. An analysis of reasons for revealing and of the type of revealed code shows that different types of firms have different rationales for openness. Implications for management are that the conflict between downsides and benefits of openness appears manageable. Provided selective revealing is practiced deliberately, the opportunities of open development dominate. (c) 2006 Elsevier B.V. All rights reserved.