Bloggo back to the blog
Is testing for mobile really that different?-->
Mobile has been on the rise lately, and testing teams are looking for ways to support their businesses by building a mobile testing company. Some argue that for mobile testing, you need to start building a new capability from scratch. Frederik Carlier from Jamo Solutions thinks there is another way, and in this blog post he explains his vision on automating mobile tests.
When you spend time researching the latest news in automated mobile testing, you probably have noticed how most vendors try to scare you into thinking that mobile testing is radically different from testing on the desktop. They have a point – the mobile world comes with a long list of new things you need to take into account. All devices have different form factors and tend to behave slightly different, even if they run the same operating system. Network conditions on mobile devices are often less than ideal, so testing you application on your corporate WiFi network will probably not give relevant results. Resources, such as CPU, memory and battery are scarcer on mobile devices.
On a closer view, mobile is not that different at all
Yet, in spite of the new challenges we face on mobile platforms, many of the tasks for testers remain the same. If you’re into UI testing, you will need to write tests that automate the user scenarios on the UI, and make sure that the application behaves as expected. If you are into performance testing, you will need to automate user actions and measure the response time of the application. And if you are a test manager or business sponsor, you will want to manage the quality of your application over time.
Similar challenges on mobile means re-use of existing best practices
So I’d argue that if most of the challenges for mobile testing are the same as they were on the desktop, so must be the solutions.
Automating test cases of Windows applications is done by accessing the individual objects of the application – if you are using HP’s toolset you will be familiar with the object repository, on Microsoft’s toolset you will use the UI Map that ship with the coded UI test. It’s a time-proven method that delivers clear, readable test scripts that are easy to create and maintain. We’ve long realised that using screenscraping technologies don’t provide the same efficiency. The good news is that you can reuse the same approach on the mobile world.
Similarly, there’s a difference between software testing and hardware testing, and you need to know which business you are in. I remember the testimony of a customer at a large trade show, saying how they spent time and resources testing their app on about 100 devices. They found an issue on one device. It turned out to be a driver issue, and they reported it to the manufacturer who fixed it. That’s great – but is that really what software testing is about? Most teams tend to test on the 10-20 most used devices in their customer base, and that works well for them.
In the end, testing isn’t just about executing tests – it’s about managing the quality of your application. So when you set up your mobile testing capability, you will also need to make sure there is a reporting capability in place, so you can update your stakeholders on the status and quality of your mobile application, helping them answer questions such as ‘is the quality of our application good enough to release?’ and ‘are we improving on quality?’.
Managing the quality of a mobile application is not different to managing the quality of a non-mobile application. There are a lot of great quality management tools out there. You are probably already familiar with some of these tools, so why not reuse them for your mobile testing?
Getting started with your mobile testing capability
If you have a good test organization in place, you can easily extend your capabilities to mobile testing, by re-using your existing tools, processes and people. You just need to add mobile to the mix.
I find that for most of the testing teams I interact with, the key to setting up an effective mobile testing capability is to re-use existing investments, start small and then grow. For example, when you start, you may want to try to reduce the complexity of mobile by focussing on one or a couple of devices and applications. When your team has gotten the hold of it, you know the time has come to grow and add new devices and applications to the mix.
If your organization is about to embark on a mobile testing adventure, I’d encourage you to focus on not only on how mobile is different, but also remember the similarities. It will help you get started faster and help you achieve a more effective and efficient testing capability.
I’d be more than happy to discuss your mobile testing plans with us, so feel free to contact me or come visit me at EuroStar!
Working as a Technology Strategists at Jamo Solutions, a firm that delivers automated mobile testing capabilities, Frederik Carlier spends most of his time working with quality assurance professionals, assisting them with implementing an effective mobile testing strategy. Frederik regularly speaks at conferences and seminars on mobile testing, discussing topics such as functional testing, performance testing and test management. In the past, he worked as a consultant for Microsoft. You can follow Frederik on the Jamo blog or on Twitter @FrederikCarlier.