Hanser-Verlag interviewed me on the topics of creativity, agile projects, the power of the tester and future challenges. The interview originally appeared in the “Hanser-Update” blog.
5 questions - 5 answers, the new specialist format from our authors: this time with Richard Seidl about software testing!
Mr. Seidl, which topics are currently hotly debated in the field of software testing?
I am currently aware of two topics in my environment that move the testers:
- How do I make the switch as a tester from a traditional to an agile project?
- How do I make test automation really take off?
… can you please briefly outline both topics for our readers?
There are testers who experience the change to an agile environment effortlessly and are enthusiastic about the new possibilities. But just as agile projects are not a panacea, there are testers who have little use for this approach and feel lost in agile projects. Training courses such as the Certified Agile Tester are a recommendable basis for understanding, but cannot address concrete situations and problems in everyday life. I recommend that testers in this situation take a closer look at the values, concepts and motivation behind the agile approach. There is so much more to it than just another process. It can also help to ask the question: How can I best contribute my skills to the project in order to increase the value and quality - perhaps in an unconventional, new way?
If you don’t get anywhere, you still have the option of honestly asking yourself whether this is the right path for you. There are by no means only agile projects, and frustration within the team doesn’t help anyone.
With automation, it is much easier to learn the concept and the technicality and integrate it into everyday life. A major pitfall lies in the complexity of the test object and the automation tool or framework. The devil is in the detail here. I consider a meaningful proof of concept and a decision and implementation in the team to be essential. But above all, I sometimes miss having the courage to leave the chosen path again when there are major obstacles and perhaps pursue other approaches to automation. If an approach is not effective, there is little point in putting more energy into it. Similarly, several tools and frameworks can be used in parallel if one is not sufficient. In my opinion, people often react far too late here. You get bogged down in the existing concept and lose valuable time instead of looking at a more suitable alternative. Instead of test coverage, frustration increases. To stand out, test automation needs a lot of attention - from planning to project maintenance.
Is it true that agility gives testers more influence or more power?
In terms of their structure, agile projects offer a stronger focus on quality - under the responsibility of all team members. A lack of quality also becomes apparent much more quickly in an agile environment. The basic prerequisite that testers are listened to in the team is therefore more given than perhaps in traditional project structures. The decisive factor is what is done with it. There are traditionally set up projects that live quality to the full, but there are also agile projects that fail if quality is not supported by everyone in the team.
Do creativity and testing go together?
Definitely! Creativity is becoming increasingly important for testers and that is a very good thing. In my opinion, this shows us that some quality-enhancing measures - pushed by the agile approach in particular - are taking effect: developers are implementing more unit tests and working with TDD, for example. In many cases, this hardens the software and makes it more robust. Code reviews are carried out more conscientiously. Support from development and test frameworks improves. Test automation becomes an integral part of development. All of this makes things easier for the tester and they can focus their attention on more important things: designing good tests and scrutinizing the product from the user’s perspective. Both require more creativity and new skills.
How do you think software testing will develop over the next few years?
In my opinion, 3 movements will have a stronger influence on the topic of software testing in the coming years:
- new “playing fields” for software testing will emerge, similar to the testing of mobile apps. A wealth of new device classes are on the horizon. Whether smart homes, 3D printing, self-driving cars, drones or the Internet of Things - there is a lot of software that needs to be tested. Test methods need to be adapted or test frameworks created for this. 2 Quality is becoming an even more essential task for the entire project team. The classic tester, who is the only “quality controller” and who keeps on praying for quality, will become less important.
- software testing will have to focus even more on the needs of the user. The user experience, usability and intuitive operation will be given more weight than the formal verification of implemented requirements.
What skills will testers need in the future?
In order to be prepared for the developments described, I would emphasize the following skills and try to promote them in my teams:
- Creativity for the design of useful tests
- Consulting skills to support other team members (developers, etc.)
- “User-understander” who can represent the user and their behavior
- Technical understanding of the use of test (automation) frameworks
Frequently Asked Questions
Why do some testers find it difficult to transition from a traditional project to an agile one?
Because agile work is more than just a different process. Some testers are naturally enthusiastic about the new possibilities, while others feel lost because they don’t quite understand the approach. It’s helpful to explore the values, concepts, and motivation behind the agile approach, along with the question of how your own skills can enhance value and quality in the project—even in unconventional ways, if necessary.
Is a certification sufficient preparation for testing in agile teams?
No. In 2015, certifications like the Certified Agile Tester were considered a recommended foundation for understanding the methodology, but they couldn’t address specific situations and problems encountered in everyday work. Anyone who finds themselves stuck after the training should honestly assess whether the agile approach is the right fit for them. Agile projects are by no means the only type of project, and frustration within the team benefits no one.
What are the most common reasons for the failure of test automation in practice?
The complexity of the test object and the automation tool: The devil is in the details. Often, teams react too late. They get bogged down in the existing concept, pour more and more energy into an approach that isn’t effective, and waste time instead of exploring a more suitable alternative. Instead of increasing test coverage, frustration then rises.
Should you commit to a single tool for test automation?
Not necessarily. Multiple tools and frameworks can be used in parallel if one alone isn’t sufficient. A meaningful proof of concept should precede any decision, and both selection and implementation are the team’s responsibility. It’s important to have the courage to abandon a chosen path when faced with major obstacles and to pursue other approaches to automation.
Does good software quality depend on the chosen process model?
No. Agile projects are structurally more focused on quality, with all team members sharing responsibility, and quality issues become apparent more quickly. What matters most, however, is what is done with that information. There are traditionally managed projects that exemplify quality, and equally, there are agile projects that fail because quality isn’t supported by everyone on the team.
Will testers become obsolete if developers use more unit tests and TDD?
No, their work will simply shift. More unit tests, TDD, more thorough code reviews, better frameworks, and test automation as an integral part of development make the software more robust and lighten the tester’s load. This allows the tester to focus on what’s more important: designing good tests and scrutinizing the product from the user’s perspective. Both of these tasks require more creativity and new skills.
Is the tester losing importance as the sole quality control officer?
In 2015, the article predicted that quality would become even more of a responsibility for the entire project team, and that the traditional tester—who constantly harps on quality and acts as the sole quality control officer—would lose importance. At the same time, new areas of application were identified: smart homes, 3D printing, self-driving cars, drones, and the Internet of Things—all requiring adapted testing methods and new testing frameworks.
What skills, besides testing knowledge, does a tester need?
In addition to creativity in designing tests that add value, the 2015 article identified three other skills: consulting skills to support other team members such as developers; the role of the “user advocate,” who can represent the user and their behavior; and technical understanding of how to use testing and automation frameworks. User experience and usability should be given greater weight than the formal appraisal of requirements.
