Better Software East 2016 - Test & QA
Wednesday, November 16
SEBTE: Simple Effective Experience-Based Test Estimation
Testers are challenged to provide accurate estimates. Later, magically, test estimates become commitments cast in concrete even in the most turbulent contexts. In fact, most test estimates are based on shallow project insights. Estimators are often ignorant of key design, usage, and environmental factors. Many traditional project management estimation techniques fail when applied to testing projects. After all, what you test next depends on what you learned before, and knowledge evolves as the project continues. Rob Sabourin shares his SEBTE test estimation model...
Leading a Testing Center of Excellence
As companies look to drive down the cost of quality, they often leap to the conclusion that centralizing QA into a Testing Center of Excellence (TCoE) will magically result in increased quality and lower costs—without understanding the path required to get there. Change requires defining mission and goals, and putting into place a plan to move forward. Equally important is the need to establish cost of quality metrics to shift the focus from the cost of testing to the cost of building quality applications. Susan Schanta shares her experience transitioning...
Artificial Intelligence (AI) in Testing: The Future Is Now
When building new content for a game such as Candy Crush Saga that has more than 1000 levels, it is extremely important to make sure that the level of difficulty is balanced and that the user does not experience crashes or problems through some unforeseen type of play. Alexander Andelkovic shows you how King is training AI-programs (bots) to test its games by mimicking human interactions. Alex explains how King is taking testing to the next level by employing Monte Carlo tree search, automatic heuristic construction, and NeuroEvolution of Augmenting Topologies (...
Thursday, November 17
Exploratory Security Testing in the Browser
Much of the time, security testing is relegated to the end of development and then is assumed to be someone else’s problem. However, late security fixes can have a huge impact on your delivery dates. Although automated scanners offer some help by inspecting your application’s HTTP traffic, most common attacks are best performed in the browser, which should be the focus of your exploratory security testing. Using familiar automation practices, open source tools (Webdriver/Selenium), and a browser (Chrome/Firefox/Edge), Abraham Marin-Perez shows you how to drive a...
Simplify Performance Testing with Production Data
When creating models of system performance, the primary challenges are where and how to start. Whatever the performance characteristics to be estimated or modeled, we need to adopt methods that are repeatable and able to address both the business and technical needs of the project's delivery. All too often, performance tests inadvertently mix scenarios with little regard for how transaction rates can confound technical recommendations and business decisions. Join Will Hurley as he presents a real-world case study that demonstrates the impact on business decisions...