John Noll, “Innovation in Open Source Software Development: A Tale of Two Features”

alt=”” title=”Firefox Tabbed Browsing” width=”350″ class=”alignleft size-full wp-image-900″ />

Noll traces two significant features implemented in popular open source software products, from proposal to final implementation: tabbed browsing in Mozilla Firefox and “edge magnetism” in the Gnome desktop environment. In both cases the features were suggested or at least implied by online postings from users of the software. In both cases, the initial posts spawned a sizable discussion among users and developers. In the case of Firefox, initially no developer took on the task of implementation, leading to a non-core developer (i.e., a user not connected to the Mozilla development team) writing an extension that added tab functionality to the software. Influenced by the plugin, a developer later added the functionality to the core code itself. In the case of edge magnetism in Gnome, a feature that causes windows to snap together when dragged in proximity to each other, the feature was dropped when an overly complex module of Gnome’s (it’s old window manager, Sawfish) was replaced with a simpler, leaner, less customizable version (Metacity). A user complained in a public post, which lead to a debate about feature richness versus maintainability: what counts as bloat? A prolonged discussion then took place among developers about how this feature could be implemented in Metacity; three years later (in 2005) they had implemented the feature in a satisfactory way.

Noll emphasizes that both of these examples, and many others, emphasize both the importance of the user community to open source innovation and the transparency of its networked communication. All such discussions are public; anyone can read them or join in if they wish to do so. These open communication channels are enhanced by “extension mechanisms that allow users with programming skills to demonstrate ideas by contributing working functionality.” (118) The user do far more than merely provide feedback for the developers: they implement the changes they want to see, at least as a demonstration. Noll also emphasizes that open source allows for alternate paths for ideas to become released features: Mozilla and Gnome are organized differently and their respective products face different “market” situations, and yet each possesses multiple possible avenues by which innovations can occur; these multiple pathways are very difficult to maintain in closed source, hierarchical structures.


Article info here.
From this book (Amazon).

Tagged with:

Leave a Reply

Comments are closed.