The People Side of Test-First Development
August 12, 2016

Kevin Dunne
QASymphony

For teams moving to test-first methodologies, helping developers and testers acquire new technical skills, workflows and processes is job one.

With the mainstreaming of Agile and DevOps, software testing has been turned upside down. No longer a manual, last-stage effort of development, much of testing is now being automated and moved in front of coding. Developers must write code that will pass each test, ensuring that quality is built into products from the beginning. Meanwhile, testing earlier in the process speeds up the release of code for production. Consequently, specified tests can now replace verbose requirements as the primary design document – and that's quite a mind shift for developers and testers alike.

Test-first methodologies are quickly gaining in favor by software development teams. 51 percent of organizations have implemented test-first methodologies, including 37 percent in the past year, according to a recent global survey of 200 software testers conducted by our company. The majority (48 percent) said that improving software quality was a top benefit of adopting test-first development.

Test-first development practices and processes are still maturing. The practice has been defined across three broad terms: Behavior-Driven Development (BDD), Test-Driven Development (TDD) and Acceptance Test-Driven Development (ATDD).

TDD is an umbrella term for the practice, although it has traditionally focused more on technical, unit testing.

BDD and ATDD incorporate user requirements and business benefits along with technical testing.

BDD differentiates through its adoption of specific automation frameworks and language for writing and running tests.

When implemented effectively, test-first development creates significant efficiencies in the development cycle and also prevents costly and time-consuming redesign late in the development cycle. Another benefit is that automated tests (or their components) can be reused when new code is written. That saves time and helps standardize the testing process. Test-first development also allows for easier separation of features so that teams can improve pieces of the application separately.

Helping teams transition smoothly to a new way of working is the first priority, over purchasing tools. Here's what you need to consider from a people and skills development perspective:

1. Retraining

Both developers and testers will need to gain new skills and learn how to work differently. No longer can these groups work in isolation, but they need to work together and cross train in order for test-first development to work.

Testers will need more technical skills than ever before, specifically learning basic development techniques to work with new test automation frameworks.

Manual testing should be used judiciously as it can cause delays with the quick code revisions that occur in TDD and BDD environments.

Testers will also benefit from high-level knowledge of modern programming languages such as Ruby or Python. The good news is, those languages are easier to learn than Java and.NET, and there are plenty of online and in-person courses to help someone get up to speed. Some test automation platforms are language specific, but many of the open source ones are language agnostic so testers and developers can both contribute in the languages of their choice. All testers don't need to code per se, but understanding how the languages work is tremendously helpful for assisting in test design and debugging.

Developers need to understand testing processes and gain an appreciation for testing overall. While the latter requires education and incentives from senior levels of the organization, the technical piece shouldn't be hard for most developers to pick up. It requires creating the test immediately as part of the design phase, and then writing and running code against that test for it to pass. Again, understanding of automated testing platforms is critical.

Developers should also integrate the use of monitoring and logging tools into their work to mitigate the inevitable missed bugs that will happen as developers begin writing their first tests. The biggest hurdle for developers is that they are used to working independently, and are motivated (especially today) by speed. Helping developers understand the need to balance quality, user experience and speed is a critical responsibility of organizational leaders when moving to test-first development.

2. Collaboration

Progressive software development teams are already adopting more collaborative practices, given the move to Agile practices and DevOps processes. Because of the tighter integration between developing and testing in test-first development, team members will need to work together closely and exchange information frequently to keep all the wheels turning.

As a result, co-locating teams is ideal. This facilitates real-time collaboration and eliminates the impact of time zone differences. Splitting testers and developers between onshore and offshore locations will make any TDD or BDD effort many times more difficult.

Regardless, there are many tools to help teams stay in touch and exchange information no matter where they sit. Using code repository platforms for storing and archiving code is essential for allowing teams to have one version of the truth. Test management platforms will enable developers and testers to view, change and approve tests in popular project lifecycle platforms.

Collaboration will not be easy at first. Developers, testers and business analysts are not accustomed to working together and some will invariably oppose this change. Those who can't adapt, may not have a future with the organization. Yet with the right leaders in place who can set attainable goals and incentives for change, any team can shift with a little bit of time and patience.

Additional considerations for moving to test-first development include timing and organizational readiness. Given other business, leadership and operational challenges, can a major change be effectively supported now? Can the company change wholesale or will you need to operate in a hybrid manner for a period of time, with some groups still developing in traditional waterfall style and others moving to TDD or BDD? Planning ahead of time can help determine the resources you will need to make the transition without disrupting business.

Once a team moves to test-first methodologies and achieves initial success, it's hard to imagine any organization wanting to go back. Test-first development is the wave of the future; it can achieve the optimal balance of user-focused quality, speed and efficiency which defines competitive advantage today in software-driven businesses.

Kevin Dunne is VP of Business Development & Strategy at QASymphony.

Share this

Industry News

October 17, 2024

Progress announced the latest release of Progress® Flowmon®, the network observability platform with AI-powered detection for cyberthreats, anomalies and fast access to actionable insights for greater network and application performance across hybrid cloud ecosystems.

October 17, 2024

Mirantis announced the release of Mirantis OpenStack for Kubernetes (MOSK) 24.3, which delivers enterprise-ready and fully supported OpenStack Caracal, featuring enhancements tailored for artificial intelligence (AI) and high-performance computing (HPC).

October 17, 2024

StreamNative announced a managed Apache Flink BYOC product offering will be available to StreamNative customers in private preview.

October 17, 2024

Gluware announced a series of new offerings and capabilities that will help network engineers, operators and automation developers deliver network security, AI-readiness, and performance assurance better, faster and more affordably, using flawless intent-based intelligent network automation.

October 17, 2024

Sonar released SonarQube 10.7 with AI-driven features and expanded support for new and existing languages and frameworks.

October 16, 2024

Red Hat announced a collaboration with Lenovo to deliver Red Hat Enterprise Linux AI (RHEL AI) on Lenovo ThinkSystem SR675 V3 servers.

October 16, 2024

mabl announced the general availability of GenAI Assertions.

October 16, 2024

Amplitude announced Web Experimentation – a new product that makes it easy for product managers, marketers, and growth leaders to A/B test and personalize web experiences.

October 16, 2024

Resourcely released a free tier of its tool for configuring and deploying cloud resources.

October 15, 2024

The Cloud Native Computing Foundation® (CNCF®), which builds sustainable ecosystems for cloud native software, announced the graduation of KubeEdge.

October 15, 2024

Perforce Software announced its AI-driven strategy, covering four AI-driven pillars across the testing lifecycle: test creation, execution, analysis and maintenance, across all main environments: web, mobile and packaged applications.

October 15, 2024

OutSystems announced Mentor, a full software development lifecycle (SDLC) digital worker, enabling app generation, delivery, and monitoring, all powered by low-code and GenAI.

October 15, 2024

Azul introduced its Java Performance Engineering Lab, which collaborates with global Java developers and customers’ technical teams to deliver enhanced Java performance through continuous benchmarking, code modernization recommendations and in-depth analysis of performance impacts from new OpenJDK releases.

October 10, 2024

AWS has added support for Valkey 7.2 on Amazon ElastiCache and Amazon MemoryDB, a fully managed in-memory services.

October 10, 2024

MineOS announced a major upgrade: Data Subject Request Management (DSR) 2.0.