Thinking outside the military box

John Pike likes to talk at length about Moore's Law and how it applies to the future of warfare. The now-famous maxim, created by Intel Corp. founder Gordon Moore in the 1960s, posits that computing power doubles every 18 months. Pike sees the law borne out in military technology with sweeping advances leading to the use of unmanned aerial vehicles, stealth fighters and bombers and network-centric communications. He believes similar advances may lead to a future battlefield where robots perform many perilous missions now performed by humans.Pike is director of GlobalSecurity.org, which he founded in 2000. Before establishing the organization, he worked for nearly two decades for the federation of American Scientists, where he developed the Federation's popular Web site.Pike spoke recently with Deputy Editor William Welsh about future combat systems, net-centric warfare and military procurement. If we think about precision munitions or unmanned aerial vehicles, the reason that UAVs have gone like gangbusters is Moore's Law. The autopilots have gotten better, the image compression has gotten better. All of those things are simply commentaries on Moore's Law.When a particular domain of activity gets swept up by Moore's Law, you can have incredibly rapid change. I believe that the military has increasingly become divided between those things that have gotten swept up by Moore's Law and those that haven't. The big challenge facing the military is to see if we can't get Moore's Law into play. Maybe you don't. There are some things for which it is just not relevant. If we think about tanks, for example, they are remarkably unchanged from three or four decades ago. If I am going to improve land combat systems, I've got my work cut out for me. But we've seen Moore's Law in play with airpower with a vengeance.If you think about the way that UAVs and precision munitions have begun to revolutionize air warfare over the last decade, it's just Moore's Law. Various aspects of airpower have gotten swept up by some manifestation of Moore's Law. The problem across the board is to figure out, "Where can I get Moore's Law working for me?" The Army's Future Combat Systems program is doing it with robots. The Army is trying to figure out what kinds of problems can be fixed with microprocessors. Many systems are labor-intensive. All of the services are having to figure out what they can do to cut labor costs.Maybe there is a limit to how much Moore's law can help us reduce the number of soldiers in a tank crew, but we had better look around the battlefield and find some other place that Moore's Law can help us; otherwise, we are simply going to price ourselves out of the market. They've got to do something to get more realistic cost estimates for programs that they can actually execute. I blame the services more than the contractors. Because it is up to the military services to make sure that they start out with realistic cost estimates. It is up to the services to understand ? based on the last half-century of experience ? how much some of these things are going to cost.The services have a problem with "requirements creep." They have a hard time telling the difference between must-have and nice-to-have. A program that is on track ? on budget, on schedule and is meeting its performance requirements ? is the exception rather than the rule. More important proportionally because that is one of the things that is less susceptible to automation. This is the individual who is going to be calling in the indirect fire. They will have a variety of Future Combat Systems automated tools, such as non-line-of-site cannons and armed robots. Either a lot of the money or half the money is wasted, but I just don't know which half. But if we are talking about the absence of al Qaeda attacks on the homeland, then we are doing something right, but I'm just not sure what.