You have heard about agile software development techniques such as eXtreme Programming (XP), Scrum, and Agile Modeling (AM). The industry is buzzing with everything from "this is the greatest thing ever" to "it's just hacking with a fancy new name." Comments like "there is no place for testers because developers and users do the testing now" and "testers play an important role in the agile methods" are both common. Scott Ambler, an early proponent of the agile movement, explains the fundamentals, values, and principles of agile development. He describes a range of agile techniques and explores many myths and misconceptions surrounding agility. Agile software development is real, it works, and it may be an important part of your future in testing. Better testing and improved quality are critical aspects of agile software development, but the roles of traditional testers and QA professionals on agile projects remain unclear.

Beta testing is an industry standard practice to obtain user feedback prior to general availability of software. Have you ever considered that the Beta release can be used to validate the software's value to customers and application users? Extending the Beta concept will result in higher customer satisfaction (and higher revenue for commercial products). Also, you can employ Beta testing to evaluate not only the software product, but the distribution (and sales) process, training, customer support, and usage within your customers' environments. Far beyond just finding defects in the product, you can focus Beta testing on how well the software is meeting your customers' needs. What does that mean to the Development team and the organization as a whole? What are the risks and challenges that we face? What are the rewards?

Security is a complex and often overwhelming issue. You cannot rely solely on trying to prevent hackers from entering your systems. Instead, you must ensure that the system safeguards itself if a hacker does break in. Three of the most common internal software weaknesses hackers exploit are dangerously constructed SQL, buffer overflows, and runtime exceptions that are not properly handled. Although testing existing code for these defects can help, it is not fool proof. You also need to make a concerted effort to prevent security vulnerabilities from being introduced as the team is writing code. Through the application of practices, such as static analysis, dynamic analysis, unit testing, and runtime error detection, you can jumpstart your security efforts and keep the hackers at bay.

• The most common internal software weaknesses that hackers exploit

eXtreme programming emphasizes test-first coding-you write the tests before writing the implementation code. You can apply the same approach in design when developing a complex system, including an architecture to support testing. To be successful, systems developed with agile methods must support a high level of testability and test automation. For large distributed systems, more sophisticated testing is needed to help determine which components may be contributing to failures. For such complex systems, you should architect the system for testing rather than add testing functionality as an afterthought. Ken Pugh presents a framework that employs polymorphic-style internal and external interface patterns to ease the work of testing and debugging. He also covers adding test-only functionality, test-only outputs, and test-only logging to interfaces.