What should testers have in mind for the New Year ahead?
As 2016 begins, we turn our thoughts towards what might lie ahead. There are several major IT and development trends that will have serious repercussions for testers in the months to come. It’s vital to plan ahead and prepare for what’s coming down the pike, but also to maintain a fluid, adaptive strategy.
These are the broad trends that we see on the horizon, with advice on how testers can rise to meet the challenges they pose.
Everything is Connected
Our definition of mobile technology is rapidly shifting to accommodate new categories such as ‘wearables’ and ‘smart car’ technology. Increasingly complex software is making its way into every facet of our lives, from home safety to home appliances, and joining it all together. Smaller and cheaper sensors with wireless connectivity are widening the net, and driving the Internet of Things (IoT) forward. Gartner predicts that 6.4 billion connected things will be in use this year, up 30% from 2015.
This ever-shrinking gap between the physical and the virtual world presents several new problems for testers. The scope is widening, as a setup of scenarios is more dynamic (and perhaps more unpredictable). To ensure quality across the board, testers need to look at the range of devices they test on and how they interact with others in this new eco-system. It’s impossible to have every physical device in hand, so the mobile device farm model is sure to spread. Solid emulation services are vital to keep costs down and make test coverage manageable.
Speed is of the Essence
Agile development methodology has already become the norm. As organizations move further towards continuous integration and continuous delivery, the emphasis on speed without compromise on quality is demanding more from test teams. The barriers between departments are breaking down to facilitate this trend. A full 25% of Global 2000 organizations will employ DevOps this year according to Gartner. As development and operations merge, it’s important that testing is not left behind.
Part of the solution to this need lies in test automation. Test automation is a prerequisite for faster, frequent software releases. Wherever possible, experienced testers should be developing their programming skills so they can write automation scripts. We shouldn’t have separate silos of manual testers and software test engineers. Demand for testers with a foot in each camp is going to soar.
With a proactive attitude of repeatable tasks within a risk-based testing assessment, testers can work out the best candidates for automation from a business perspective. They’ll need to be able to make quick, strategic decisions about what components to manually focus (test) on, with the limited manual resources available to them. This can’t be done effectively without a real understanding of the business value of the software.
Data Processing and Security
We can’t continue to treat separate devices as individual islands that are tested in parallel. The expectation that the underlying software will work as intended across a wide variety of devices means that testers will have to find ways to validate the exchange and synchronization of data. Data processing has to be absorbed into functional test suites.
The ramifications of security breaches are growing more serious. As we build access into cars and our homes, we simultaneously raise the stakes on security. Data must be securely collected and stored. Compliance is mandatory and it must be validated. Failure to properly test application security could lead to serious breaches, and the consequences could be devastating.
Behind the Scenes
Many modern devices are simply windows onto a remote interaction with software, often employing APIs to access a distant cloud-based infrastructure. There has to be a solid methodology for testing everything that goes on behind the scenes in the cloud. We need to make sure data packets aren’t being lost, that we have some analytical insight into data transactions, and that performance benchmarks (usually ‘expectations’) are met.
Validation is necessary for data and activity logging. Providing testers with access to a diagnostic assessment that can peek behind the scenes, will mitigate risk and enable us to minimize time spent on investigating issues. As the enterprise shifts from legacy systems to cloud infrastructure, or more commonly, relies on a hybrid of the two, we need to develop innovative testing techniques to provide the confidence to underpin fast and flexible business decisions.
There’s plenty for testers to consider as 2016 approaches, but it’s worth remembering that these challenges are also an opportunity to excel and improve software quality for all.
Mush Honda is QA Director at KMS Technology, a provider of IT services across the software development lifecycle with offices in Atlanta and Ho Chi Minh City.