Photo: Luke Wolagiewicz/WPN; Kareem Raheem/Reuters

FAST, CHEAP & OUT OF CONTROL: Improvised explosive devices made from cellphones, radios, old mortars, and other low-tech mechanisms have exacted an ­enormous toll in Iraq.

And just as in the software community, information technology and the Internet play a pivotal role in bringing insurgents together. The resurrection of al-Qaeda is a good example, says Brian Jackson, a terrorism expert and associate director of the Homeland Security Program at Rand Corp. “Given the structural changes that were required of al-Qaeda to adapt to its loss of Afghanistan as a safe haven,” Jackson says, “the interconnections among disparate parts of the decentralized organization that the Internet made possible have been important for its survival.”

The reliance on IT also enables open-source groups to identify and respond to problems much more rapidly than a more structured, top-down entity can—be it the Pentagon or a large software company such as Microsoft. According to some estimates, it now takes Iraqi insurgents less than a month to adapt their methods of attack, much faster than coalition troops can respond. “For every move we make, the enemy makes three,” U.S. Brigadier General Joe E. Ramirez Jr. told attendees at a May conference on IEDs. “The enemy changes techniques, tactics, and procedures every two to three weeks. Our biggest task is staying current and relevant.”

Unfortunately, the traditional weapons­acquisition process, which dictates how the United States and other Western militaries define and develop new weapons systems, is simply not designed to operate on such a fleeting timescale. It can take years and sometimes decades—not to mention many millions or billions of dollars—for a new military machine to move from concept to design to testing and out into the field. Worse, the vast majority of the battlefield technologies now wending their way through the acquisition bureaucracy were intended to fight large force-on-force battles among sovereign nations, not the guerrilla warfare that typifies the conflicts in Iraq, Afghanistan, and elsewhere.

Meanwhile, time is on the insurgents' side. Since the start of the war, the consumer-grade products on which they rely have undergone several generations of improvement. Microprocessor speeds, for instance, have leaped by a factor of at least four in that time, while the cost per MIPS—or million instructions per second, a standard benchmark for processors—has dropped by roughly 70 percent.

This past spring and summer I interviewed dozens of current and former military officers, analysts, weapons developers, and others to try to understand why the coalition forces' technological might has proved so ineffectual. Nearly everyone I spoke with agreed there is a serious mismatch between the West's industrial-age approach to warfare and the insurgents' more fluid and adaptive style. All agreed, too, that the West will likely face more such confrontations in the years and decades ahead. The big concern, many people told me, is that once the war in Iraq has ended, the innovation that has occurred there and the lessons learned will be lost as the Pentagon returns to “business as usual”—that is, building enormously complex and costly weapons systems and training troops to fight large-scale wars.

To understand open-source warfare, it's instructive to revisit Eric S. Raymond's 1997 manifesto, The Cathedral and the Bazaar, in which he describes how a large community of open-source software hackers created the operating system Linux.

“Linux is subversive,” Raymond wrote. “Who would have thought even five years ago [1991] that a world-class operating system could coalesce as if by magic out of part-time hacking by several thousand developers scattered all over the planet, connected only by the tenuous strands of the Internet?” He likened the rise of Linux to the public marketplace of the bazaar. The programmers agreed to observe a few simple principles but were otherwise free to innovate and create. Raymond contrasted that style with the “cathedral” approach to software, in which a single organization, using highly planned, sequentially structured steps, maintained tight managerial control over every aspect of the process.

Eventually, the open-source culture would triumph over the proprietary world, Raymond argued, not because it was morally right “but simply because the closed-source world cannot win an evolutionary arms race with open-source communities that can put orders of magnitude more skilled time into a problem.”