DEVOPSdigest

Industry experts — from analysts and consultants to users and the top vendors — offer thoughtful, insightful, and often controversial predictions on how DevOps and related technologies will evolve and impact business in 2021. Part 4 covers testing and quality.

Start with: 2021 DevOps Predictions - Part 1

Start with: 2021 DevOps Predictions - Part 2

Start with: 2021 DevOps Predictions - Part 3

TESTING: DEVELOPERS LEAD THE WAY

In 2020, we saw the beginnings of a sea change in the world in testing, one in which empowered developers took a more active role in shaping the testing landscape. That trend will accelerate even further in 2021, as developers take the next step and begin to assert clear leadership over organizations' testing strategies. We'll see more developers incorporating testing directly into their workflows. We'll see more quality engineers embedded directly into development teams to enable tighter collaboration. Most importantly, we'll see organizations increasingly look to developers to determine the tools, systems, and processes upon which they rely to deliver quality at scale. In 2020, developers became empowered. In 2021, they'll lead the way.
Alissa Lydon
Director of Product Marketing and Technology Evangelist, Sauce Labs

ADDITIONAL STAKEHOLDERS JOIN THE QUALITY CONVERSATION

That quality assurance and development teams are collaborating and sharing ownership of product quality already represents a marked improvement from years past. As we move into 2021, however, even more personas will demand a seat at the quality table. That includes product teams with a vested interest in quality as a means to drive product engagement, operations teams that need to monitor and optimize the post-production experience, and go-to-market teams who want to understand the role user experience can and should play in their sales and marketing initiatives. In a world where quality is everyone's responsibility, everyone will desire a seat at the table.
Alissa Lydon
Director of Product Marketing and Technology Evangelist, Sauce Labs

QUALITY ENGINEERING

QA is dead — long live quality! The adoption of continuous integration/continuous delivery (CI/CD) practices has set the foundation for the emergence of modern Quality Engineering (QE). If you look at a typical schema for the software development life cycle (SDLC), it does not offer visibility into the production side. Organizations tend to fly blind once they’re in production because CI/CD tools live almost entirely on the left-hand side — the pre-production side. But the right-hand side — the production side — is where customers are using it. On that side, you need stability management tools and software health monitoring to implement testing in production. Although gradually gaining ground, mid-sized and large organizations have yet to instrument testing in production correctly to achieve rapid iteration. While apps are being released to customers more frequently due to growing pressure for new versions and new additional features, there is a desire to launch timely releases with minimal bugs. Implementing strategies such as A/B testing and phased rollouts (where you give your software to a subset of your customer base to see how it performs) will help mitigate risks and deliver a quality experience to customers. Stability insights from such an approach can also help inform decisions on future rollouts and help create feedback loops between pre-production and production. As such, in 2021 we will see QE budgets align with testing in production strategies and stability will serve as the data-driven layer across the entire application lifecycle.
James Smith
CEO, Bugsnag

THE END OF BUG-FREE SOFTWARE

Something we predict will happen that might surprise others is the end of bug free software. Meaning bugs or errors aren't going to be the fearsome headache they once were. There are new tools that help make bugs easier to catch when they happen, but also harder to replicate in the future. And, because there continues to be an increase in demand for developers, leading to more junior or no code developers, proactively solving issues before they reach users will be critical.
Brian Rue
CEO and Co-Founder, Rollbar

DIGITAL APP PERFORMANCE TESTING

The growth of new digital app types (such as progressive web apps, app clips and Android APKs), will require teams to better focus on digital app performance testing pre and post-production, to meet the expectations of a better user experience.
Eran Kinsbruner
Chief Evangelist, Perfecto at Perforce Software

CODELESS TEST CREATION

As part of software test automation, codeless test creation will become a widely among many more DevOps teams, to enable faster, more reliable, robust and maintenable testing accessible to both non-coders and developers
Eran Kinsbruner
Chief Evangelist, Perfecto at Perforce Software

AI-BASED TEST GENERATION

In 2021 as enterprises adopt release automation best practices, we will see heavy investments in AI-based test generation. Development teams, striving to bridge the technical debt caused by years of code inflation will make the transition from manual test creation to test automation.
Dori Exterman
CTO, Incredibuild

INTELLIGENT AUGMENTED ASSURANCE

Automation and AI will play a larger role in the testing life cycle going forward as intelligent augmented assurance comes into play. The testing industry will increase quality by improving coverage, the throughput of the testing, environment, data and defect management. Taking out a lot of the human guesswork from the testing life cycle will result in better software. It will also free up testers to do other jobs and explore new roles.
Jayanti Murti
CTO, Digitate

TESTING IN PRODUCTION

Expect to see heavy investments in enhancing the customer user experience and service availability, translating into a more rigorous performance analysis and testing in production, and into more shift-left application performance assurance on top of traditional functional testing.
Eran Kinsbruner
Chief Evangelist, Perfecto at Perforce Software

With the end of the monolithic applications also comes the end of the "finished" product release. In 2021, organizations will increasingly focus on the concept of experimentation in production, using new technology to continually A/B test different digital experiences and understand in real-time how users interact with an application, and then continuously translating those learnings into near-real-time improvements. In other words, beginning in 2021, no product release will ever truly be "done."
Alissa Lydon
Director of Product Marketing and Technology Evangelist, Sauce Labs

RISK SIGNALS

To date, the success or failure of DevOps initiatives has been defined by the extent to which organizations can release high-quality software at rapid speed. But as organizations continue to grow more comfortable with automation and more confident in their ability to deliver quality at speed, the barometer for success will start to change. Quality at speed will become table stakes, and in its place, DevOps success will increasingly be defined by our ability to unify and discern risk signals. Today, most risk signals exist disparately throughout the DevOps toolchain, spanning everything from code quality to APIs to application security and UI functionality. Beginning in 2021, the ability to harness those signals into a unified view of customer-experience risks will come to define the success of DevOps initiatives
Alissa Lydon
Director of Product Marketing and Technology Evangelist, Sauce Labs

FUZZ TESTING

The automation of unit testing: Amid demands to push more testing left, the time is ripe for automation technology to evolve unit testing approaches. Property-based fuzzing: technologies like fuzzing, combined with property-based testing techniques, are significantly more effective and efficient than traditional unit testing methods. Developers spend less time on testing while writing better tests. As organizations see gains from property-based fuzzing, we will see deeper integrations between fuzzing techniques and source code management systems in 2021.
Alex Rebert
Co-Founder, ForAllSecure

DATA EMERGES AS MOST IMPORTANT TESTING ASSET

Data is poised to emerge in 2021 as the most important asset in the testing ecosystem. As they grow increasingly proficient at automating testing on a massive scale, organizations will begin to prioritize platform-driven, data-centric strategies that enable them to make sense of the many quality signals they receive throughout the testing and development process and use that data to continually iterate and improve. This will in turn drive increased focus on the ability to aggregate data in a meaningful way and leverage methods like machine learning to identify where risk is the greatest and glean insight into how to quickly remediate issues.
Alissa Lydon
Director of Product Marketing and Technology Evangelist, Sauce Labs

BACKUP/HA DATA FOR TESTING

Companies will look to use backup and HA data for DevOps. Companies will look to get more value from replicated data "at rest" to use it for more than simply disaster recovery. Companies will tap stored data for a variety of testing, including DevOps testing, and availability and failover testing.
Cassius Rhue
VP, Customer Experience, SIOS Technology

ACCESSIBILITY TESTING

We expect organizations to rely increasingly on automated and guided accessibility testing next year. As a time-saving supplement to their manual efforts, automation can identify a wide range of problems that keep their content inaccessible to people with disabilities. Recent research found that automated testing, supplemented with simple guided tests, can identify 84 percent of accessibility issues. Achieving this level of coverage with automation, while avoiding quick-fix widgets, enables Dev Teams to make a huge impact on users without slowing existing processes.
Dylan Barrrell
CTO, Deque Systems

2020 LAYOFFS IMPACT 2021 APPLICATIONS

One of the many unfortunate effects of the COVID-19 pandemic in 2020 was widespread layoffs impacting QA teams, especially in hard-hit industries such as travel, hospitality, dining, and retail. But while many QA teams were thinned out, organizations still moved forward with their digital initiatives, and in many cases, greatly accelerated them to account for the loss of in-store revenue. Many organizations continued rapidly delivering new application features and functionality to market, only without the level of quality and testing scrutiny they would have normally received under the watchful eye of fully staffed QA teams. With encouraging progress on vaccine development and a more effective national and worldwide response strategy portending the potential reopening of the economy at some point in 2021, organizations in recovering markets will likely move quickly and aggressively to rebuild their QA teams. But the damage may already be done, and for many, there will be a price to pay early in 2021 on the customer experience front for the blistering pace at which under-tested features were delivered in 2020.
Marcus Merrell
Senior Director of Field Services, Sauce Labs

Go to: 2021 DevOps Predictions - Part 5, covering CI/CD, infrastructure and more.