Software Security: Building Security In

The software security touchpoints exist to drive out technical risk. Critical to proper application of the touchpoints is the notion of keeping track of security risks as they are uncovered and making sure they are properly dealt with. The RMF is about identifying, tracking, and mitigating software risk over time.

Central to the notion of risk management is the idea of describing impact. Recall from Chapter 1 that risk is defined as probability x impact. Without a clear and compelling tie to either business or mission consequences, technical risks, software defects, and the like are not often compelling enough on their own to spur action. Though the risks I focus on in this book are all tied directly to software and all have clear security ramifications, unless they are described in terms that business people and decision makers understand, they will not likely be addressed. There is nothing more frustrating to a technical person than identifying a serious problem that never gets fixed. We can avoid that frustration by properly describing impact.

Put more succinctly, a major hurdle to the proper handling of technical risk has been the inability to tie risk clearly to business impact. This leads to the techno-gibberish problem. Software is a fairly geeky domain. It's about arcane technology that business people don't understand. The question needs to be: How do you get business people to care whether their software works or not? The answer has to be that software risk must be discussed, understood, and related in terms of business impact. As a technical person, you need to say something like, "If the flimflobble in sector four has a floosblozzle failure, that means we will miss the first quarter number by $2 million" (as opposed to just saying the first part). Business people can relate to the last part of the statement.

The RMF described here is a condensed version of the Cigital RMF, which has been applied in the field for almost ten years. An RMF is designed to manage software-induced business risks. For purposes of clarity, the RMF is described here in the context of a particular project; however, many of the activities can be applied at different levels. Through the application of these activities, analysts bring to bear their own technical expertise, relevant tools, and technologies.

The purpose of an RMF like this is to allow a consistent and repeatable expertise-driven approach to risk management. By converging on and describing software risk management activities in a consistent manner, the basis for measurement and common metrics emerges. Such metrics are sorely needed and will prove to be extremely valuable, allowing organizations to better manage business and technical risks given particular quality goals; make more informed, objective business decisions regarding software (e.g., whether an application is ready to release); and improve internal software development processes so that they in turn better manage software risks.