Avoid the worst consequences, address complexity and balance security with other business needs by working through the issues using a systematic approach and your organization will come out OK.\nRecently, an\u00a0American Business Roundtable report\u00a0concluded: "Most of the significant cyber-incidents . . . have had at their root cause defective and readily exploitable software code." This comes at a time when escalating threats, regulatory creep, increased software reuse and dramatic cost increases associated with a seemingly endless cycle of patches and vulnerabilities have made software quality, or surety, a critical issue.Surety is the extent to which it can be made certain that a system attains the desired properties. Surety levels are associated with risk. High-surety systems are used for applications whose failure could lead to deaths or catastrophic financial loss. Medium-surety systems are used for applications whose failure could cause severe but survivable damages or losses. Low-surety systems are used for applications whose failure could cause lost productivity or minor damages.Most commercial off-the-shelf (COTS) products are available only as low-surety offerings. Even most security software - such as domain controllers, automated patch management systems and most identity management systems - are low surety because the market values low cost, flexibility, backward-compatibility, features and schedule over security. The economics haven't been there for the secure design, secure code, extensive security testing and strong change management required to build higher-surety software.But vendors can do better. Well-known vulnerabilities such as buffer overflows can be largely eliminated using commercially available code-scanning services for as little as $1 per line of code. Customers should demand at least this level of diligence from vendors. As an industry, we need to put more effort into researching and documenting secure development practices, training practitioners, changing license terms to favor users over vendors and imposing consequences on vendors for software failures.But let's face it: Market economics aren't going to change altogether. And in the end, software surety can only get so far in the face of ever-increasing system complexity. Ten years from now, serious software quality issues will remain. That's why properly constructed medium- or high-surety systems use COTS components but don't trust them too much. For example, a medium-surety manufacturing line might mitigate the risk in a low-surety COTS control system through a hardware failsafe that physically prevents the flow of fluid into a pipe from changing rapidly, no matter what the software does.Developing system-wide surety approaches is the long-term issue. Practice basic security by following\u00a0ISO 17799\u00a0and other control standards for management. Conduct risk assessments and map risk to the required surety thresholds. Demand higher software quality from vendors that write applications for higher-surety environments or whose widespread operating system and security infrastructure software aggregate risk. Deploy a layered defense to protect against inevitable failures. Avoid the worst consequences, address complexity and balance security with other business needs by working through the issues using a systematic approach and your organization will come out OK.Blum is senior vice president and research director with Burton Group, an integrated research, consulting and advisory service. He can be reached at firstname.lastname@example.org.