Given enough eyeballs, all bugs are shallow

“Given enough eyeballs, all bugs are shallow” is Eric Raymond’s maxim for the power of many programmers examining and improving Linux. It names a practical advantage of decentralized work: contributors with varied skills and perspectives can notice defects and offer fixes without waiting for a hierarchy to assign them a task. In the book’s broader argument, the saying also expresses a general principle of collective intelligence: a group can become more capable when its members contribute distinct information rather than simply follow one another.

Linux and the Open-Source Maxim

Linux has no owner, formal organization, organizational chart, or bosses directing contributors. When a problem arises, it is fixed only if someone independently offers a good solution; people work on what interests them and ignore the rest. The arrangement may look haphazard, but it has made Linux an important challenger to Microsoft.

That loose structure is also a source of variety. A conventional company can mobilize employees toward a particular task and build expertise in work it already knows how to do, but its possible solutions are limited by the number of workers, their available time, and organizational politics. Linux, by contrast, draws on a large supply of programmers willing to contribute. With enough contributors, the range of possible solutions expands, and someone is more likely to recognize a bug and devise a fix. Raymond’s phrase condenses this reasoning into a memorable image: many eyes make flaws easier to see.

Search, Waste, and Selection

The book compares Linux’s operation to a market and to a bee colony sending out many foragers in the hope that one will find the best route to the flowers. The process is less efficient than identifying the best route in advance or selecting one especially capable forager. Hundreds or thousands of programmers may spend time exploring solutions when only a few will find the one that works.

Yet this apparent waste is also a strength. Generating many alternatives makes it possible to discover solutions that a smaller, directed group might never produce; those alternatives can then be winnowed down. The maxim therefore does not promise that every contributor will find a defect or that every proposed fix will be useful. Its claim is about the advantage of a sufficiently broad search: variety increases the chance that a problem will be noticed and addressed. The book’s accompanying principle is to let many possibilities emerge and then select the strongest.

Independence and Collective Judgment

The expression fits the book’s wider contrast between independent judgment and sequential imitation. In classroom experiments on information cascades, people often let earlier choices outweigh their own private information. When participants were instead rewarded for the accuracy of the group’s collective answer, they paid more attention to their own information; although an individual could then make a wrong guess, the group’s judgment became more accurate. In that setting, collective intelligence depended not on everyone copying a visible consensus but on preserving distinct contributions.

Linux applies a related logic through decentralization rather than through a formal decision procedure. Contributors choose what to work on, and the resulting diversity supports a broad search for bugs and fixes. The saying thus serves as more than praise for a large community: it links the practical work of finding software defects to the book’s argument that groups can outperform individuals when they preserve independence and make use of varied knowledge.

Limits of the Maxim’s Logic

The book does not treat decentralized search as effortless or universally efficient. Linux’s many contributors can duplicate work, just as many market participants may explore alternatives that will not survive. Nor does the maxim by itself explain how a successful fix is selected or incorporated; it emphasizes the discovery advantage of having enough people looking.

Its significance lies in the contrast it draws with a system that relies on a few designated experts. A hierarchy can concentrate effort, but it can also narrow the pool of ideas and leave errors unseen. “Given enough eyeballs, all bugs are shallow” gives the book’s alternative a compact form: wide participation can make difficult problems more tractable, provided the system can turn diverse attention into useful corrections.