In the event I can't get all this off my chest on my panel at OSBC on Open Source Innovation, here are my thoughts on a couple of levels.
While we picture the inventor as a solitary, if not asocial genius -- this stereotype is not often the case or on the topic of innovation. Invention is a disruptive breakthrough from pure research, that provides a platform for others innovate. Innovation is implementing something new, largely an incremental process of improvement at the margin.
Innovation springs from diverse groups with specialized expertise. The greatest breakthrough for open source, IMHO, is applying collaborative methodologies for development. Inherent in collaborative practices is a greater opportunity for innovation than competitive practices.
Some norms such as the right to fork, open participation and self-organizing contribution strengthen this opportunity and provide models for consideration beyond software development. When a project can be forked, it provides a balance against poor management (albeit at a cost) and fosters a leadership style that lets other express ideas and have them be heard. Leadership forms the core of a social network of innovation, being an arbiter of information and quality outcomes. Open participation is essential to innovation, to bring in new people with new ideas. By self-organizing I don't mean some high falutin' emergence, but the simple freedom for people to choose where to contribute based on their expertise and personal motivation.
One body of work I suggest you follow is John Hagel and John Seely Brown leading up to the release of their book next week, The Only Sustainable Edge. In their recent HBR article, they suggest that innovation is the only source of sustainable competitive advantage. Most of the opportunities to compete on the basis of efficiency (automating transactions) are gone. Now the opportunity is managing exceptions with groups of diverse specialization under constraints, particularly across organizational boundaries -- where productive friction occurs, a source of innovation. They emphasize the use of a reference model and rapid prototyping. While their approach to sustainable competitive advantage emphasizes continual process innovation, my understanding of this model is very similar to open source methodologies.
Open source practices are proliferating outside software development, most noticeably in open or free content communities such as Wikipedia. This is the topic of a larger post, and in fact my contribution to an O'Reilly book on open source, but I wish to emphasize what might be the core issue of open source innovation.
Part of how Wikipedia is able to enable collaboration at unprecedented scale is not just how there are less dependencies than with developing software. Having a reference model for what an Encyclopedia should be provides an organizing framework, design benchmark while reducing coordination risks and moderation costs.
Consider for a moment the most successful open source projects. Linux, Apache, MySQL, JBoss and even Firefox -- all have established reference models. Further, there is an almost competitive drive to outperform against proprietary incumbent products.
You may think this means that open source is not innovating when building substitutes. But behind the curtain there is a remarkable level of innovation in community process. In fact, a third model of production, commons-based peer. I'm writing part of this while sitting in an Open Source Initiative meeting, watching this process at work (Just overheard in a debate about licensing, Nelson's Maxim: There are no such things as problems, there are only unmet business opportunities). The initial innovation is perceived as driving cost while pooling risk in production processes. Geoffrey Moore suggests an increasing focus on mission critical software that is context, not core to business. But this isn't how many people think of innovation.
As Larry Augustin points out, the next wave in open source may be in applications. I would suggest areas where existing reference models are well established are under the greatest threat. Applications are more visible than Infrastructure (BitTorrent which accounts for half of the net's traffic, but how well is this innovation known?), and the story of product innovation within open source may very well come out.
Innovation does work hand in hand with journalism. Innovation is implementing something new, journalism covers what is new, readers learn what is new and perhaps use or talk about what's new. The point is that some of the best innovation occurs in the substrate of what is otherwise infrastructure or process, but marketing shapes our understanding of innovation and validates what is new.
But let's bring this back to the individual developer. What is truly new is their ability to freely contribute. For companies, they have wonderful opportunities not only to drive down cost and pool risk -- but to tap into the social incentives that drive people to produce. The problem is these cooperative opportunities require enterprises to give up some control.
At Socialtext the interesting open source innovation isn't our support for Kwiki, the hybrid open source business model or how users of our tools employee open source practices. It is how we encourage our employees to participate in communities outside the organization. About 20% of our development time goes back into our open source product. There are times where a certain feature may not be ripe for commercialization. But giving them the freedom to experiment with creating open source plugins may lead to something that directly benefits the business. Community participation goes beyond coding, to blogging, volunteering with non-profit organizations to political activism. All of which not only helps work-life balance, but refreshes creativity and builds relationships across boundaries.
I would suggest that open source could improve management practices, if we can get past treating our employees as competitors. Through sharing, innovation proliferates.