September 2015 Summaries

Appium implements various locator strategies to find elements in mobile devices, specifically for Android and iOS. The class name strategy uses a string representing a UI element on the current view, while the accessibility id strategy reads a unique identifier for a UI element that remains constant during localization or other changes. The xpath strategy exposes the functionality of XPath language to locate elements within a mobile view, and in mobile environments, ids are not CSS ids but rather native identifiers. Appium also provides methods to use the underlying device automation technology directly, such as Android's UI Automator framework and iOS's UI Automation framework, allowing for fine-grained control over finding elements.

A definition of "done" is a clear and concise list of requirements that a software increment must adhere to in order to be considered completed. This allows everyone involved in the project to understand what "done" means, ensuring consistency and quality. However, organizations may need to create their own version of DoD as they apply Agile methods for the first time, addressing issues such as inconsistent standards, lack of transparency, and poor communication. A well-defined DoD can lead to better decisions, planning, and execution, minimizing risk and improving product quality. It captures activities that can be committed by the team, leading to improved processes and clearer communication at each level, including story, sprint, and release.

The concept of Minimum Functional Product (MFP) is introduced as a way to break down large user stories into smaller, testable pieces that still provide value to the customer. By identifying MFPs, teams can prioritize how to build and test their product, keeping stories small and manageable. This approach helps drive conversations around what is critical for delivery, forcing teams to think in smaller pieces and improve estimation, planning, design testing, building, and delivering processes. The MFP concept is particularly useful in Continuous Delivery environments, where large user stories can lead to scope creep and waterfall life cycles. By establishing MFPs, teams can establish clear boundaries around what is in scope and what needs to be deferred, ultimately improving the value delivered to clients.

Sauce has announced a public beta of its Real Device Cloud, a cloud-based platform that allows developers to automate mobile testing on real devices, in addition to emulators and simulators. The Real Device Cloud offers instant availability of popular iOS and Android devices, with no waiting or reservation system, and supports massive concurrency for parallel testing. It integrates with top CI tools, including Jenkins, Bamboo, Travis, Team City, and CircleCI, and provides full reporting, secure access to backend data, and enterprise features such as account management and SSO integration. The platform is designed to support test native, mobile web, and hybrid apps, and will initially include Apple iPhone 6 and Samsung Galaxy S4 and S5 devices, with more device types planned for the future.

Can You Test it All? The question of whether to test everything is often met with a resounding "no" due to time or resource constraints. However, ensuring a quality product can be achieved by testing smarter, not harder. This involves breaking down large user stories into smaller tasks, identifying key acceptance criteria, and understanding the hero workflows that users will interact with the system. By focusing on high-risk areas and using exploratory testing for complex scenarios, teams can achieve 20% test coverage that identifies 80% of bugs, while also considering factors such as the Pareto Principle and leveraging Big Data to inform test strategies. Ultimately, testing is a team effort, requiring collaboration and conversation to determine the right level of coverage, rather than relying solely on individual testers to cover everything.