You are here

Best of TechBeacon 2018: Quality comes first for app dev and testing

public://pictures/Jaikumar-Vijayan-Freelance-Writer.png
Jaikumar Vijayan, Freelance writer

With more organizations implementing automated testing practices these days, developer and QA focus has shifted from "Do we need test automation?" to "How do with get the most out it?" The use of AI and machine learning in test automaton received a lot of attention this year, as did other core topics such as training testers to become automation engineers and developers to become better testers.

TechBeacon's top app dev and testing stories for 2018 cover some of the most significant trends and topics in this space over the past 12 months. Check out what experts in the field have to say about how AI is changing test automation, the mistakes organizations make in preparing testers to become automation engineers, and tips from a pro on how to drive a culture of quality in your organization.

Gartner Magic Quadrant for Software Test Automation

How AI is changing test automation: 5 examples

AI and machine-learning approaches are increasingly being used in automated testing scenarios. Testers are using the power of ML to find patterns in software that are predictive of coding failures, to do more precise testing, to automatically write tests, and to pursue several other use cases. Here Joe Colantonio, veteran developer and implementer of multiple software automation and performance testing tools, lists five ways in which AI is changing test automation, how test automation training is broken, and the mistakes organizations make.

7 habits of highly effective SDETs

Software developers who can also do testing are a hot commodity these days among organizations of all sizes. But while there's no dearth of developers claiming testing abilities, the proportion of coders who actually also have the skills to make for good software developers in test (SDET) is pretty small, says Angie Jones, senior developer advocate at AppliTools. Jones draws on her extensive experience in recruiting people to the role to identify seven traits that the most effective SDETs possess.

[ Conference: ADM Summit 2019: Optimize Your Deployment Pipeline ]

Two years with no testers: What I learned

One drawback with having dedicated QA testers on the development team is that other developers pay less attention to quality. The safety net provided by dedicated testers can often cause developers to pay less attention to writing quality code than they should. Rouan Wilsenach, software developer for TES Global, describes this and other lessons he learned as a member of a team with no testers for two years. The main takeaway: An absence of dedicated testers is not always a bad thing in a development environment.

World Quality Report 2018-19: Key trends shaping QA today

Malcolm Isaacs, senior researcher at Micro Focus, looks at some of the key takeaways from the company's latest World Quality report. Among them is the fact that end-user satisfaction has become one of the top priorities for QA, testing teams are increasingly adopting AI, and new roles are emerging in QA and testing. Isaacs, who has held various positions in the application development field, breaks down the five key trends revealed in Micro Focus' report, based on responses from 1,700 software professionals in 32 countries.

Intelligent test automation gives Orion spacecraft a boost

NASA is using intelligent automation in testing all aspects of the Orion deep-space mission program. The goal is to ensure that all onboard software and equipment works as expected when Orion launches on its mission sometime next year. Computer engineer and Eggplant CTO Anthony Edwards explains how NASA is using and benefiting from intelligent test automation, the testing challenges that organizations in the aerospace and defense sector face, and the six steps to maximize mission success.

7 container design patterns you need to know

Design patterns are key to using containers properly. They help solve common problems that can crop up when deploying containers and can help make containers reusable. Design patterns can also provide a common language for communicating about the architecture of containerized applications. In this report, AutoWeb cloud/DevOps engineer Christian Melendez explains the importance of design patterns and highlights seven common ones that every DevOps team should consider.

How I created a culture of quality

Quality assurance is about a lot more than just your testing tools or your bug count. It's a mindset and a culture that the entire development team should be engaged and invested in, explains Angela Riggs, QA engineer at Vacasa. Here Riggs uses her experience at a previous job to talk about the challenges involved in creating a culture of quality in software development and explains why factors such as change management, onboarding, and roles and responsibilities are key to overcoming them.

Test automation training is broken. Here's how to fix it.

The growing demand for test automation engineers is driving increases in test automation training courses. Unfortunately, many of the courses put too much emphasis on automation tools and not enough on software testing itself. Test automation and service virtualization consultant Bas Dijkstra discusses why test automation training is important, explains why it is broken, and offers tips on how to clean up the current mess.

9 ways organizations screw up continuous delivery

Implementing a continuous delivery practice can help improve the reliability of your software development process. However, many organizations are not getting the most out of CD because of a failure to adopt key best practices. Matthew Skelton, head of consulting at Conflux, who has helped 30 organizations worldwide implement CD practices over the last seven years, identifies nine mistakes to avoid when implementing CD. Among them: using slow deployment pipelines, having no logging or metrics, and not investing enough in build and deployment.

Why separate test automation teams don't work

Organizations often use one team for doing development work and a separate team writing automated tests for it. The reasons are they don't want the team doing the core development work to be slowed down by writing automation, automation is viewed as a specialized skill, or automation is an afterthought. Stephen Frein, software engineer at Comcast and former manager of software projects at the US Department of Defense and Treasury Department, explains why separating test automation teams is a bad idea.

Why converting test teams to automation is a challenge

Turning your entire team of hands-on quality testers into automation engineers can be a big challenge. While a handful might make the cut, many others won't, because typically most testers don't know how to code, do not understand algorithmic performance, or can't solve logic problems, says Paul Merrill, CEO at Beaufort Fairmont Automated Testing Services. Merrill draws from his experience accelerating test automation in agile teams to explain why trying to transition your entire testing team to automation engineers is most likely not going to work.

Should you test trivial code?

There's little consensus among developers whether it's a good idea or not to test trivial code. While some have argued that there are perfectly valid reasons to test other people's code or implementation details, others have argued the opposite. Context is important to getting the answer right in your particular situation, says Carlos Shults, developer at SmartBio Technologia. You need to be able to detect when trivial code becomes non-trivial and have mechanisms for consistently maintaining the quality of your test suite, says Shults, an experienced developer in web and desktop environments.

[ Partner Resource: Learn more about key trends in performance testing. Attend the PerfGuild online conference, which runs from April 8-9. ]