Section 17.7.  Exploratory Testing at Iteration End

17.7. Exploratory Testing at Iteration End At the end of the iteration, all the people involved in the development of RentEz met to carry out exploratory testing [Bac][Mar][KBN+02], an approach encouraged by the context-driven school of testing [Con]. Exploratory testing is nonautomated testing that is carried out to explore the quality of the system and to find out how well it achieves its goals. The tests are not preplanned but arise out of the exploration. They naturally go beyond the automated tests. At RentAPartySoftware, exploratory testing was carried out socially, with an emphasis on creativity and fun: finding problems by "thinking outside the box." It was also a good way to celebrate and explore the results of the hard work of the finished iteration. Questions & Answers Q1: What happens if the development team goes faster or slower than expected? A1: During the iteration, storiesor parts of storiesare added or removed by the Customers so that the iteration finishes on time. This encourages a focus on high value per cost and on getting clear feedback on story estimation. Q2: What happens if a new story comes up during the iteration? A2: It is written on an index card for now. It may be chosen in the next iteration. Even if it seems important and should be done "right now," it's not long before the current iteration will complete. As we saw in Section 17.6, the stories will be elaborated during discussions between the Customers and others on the team. It's a judgment call as to whether new ideas are a big enough change to be treated as a new story.