Friend of mabl Stack Overflow recently released their 13th annual Developer Survey, which explores the latest trends, technologies, and tools lighting up the developer community. Built on Stack Overflow’s passionate community of aspiring and professional developers, the survey is a treasure trove of insights into how software is being built today.
For quality professionals, the report helps shed light on what technology is working - or isn’t working - for their closest collaborators and quality contributors. Given the size and breadth of survey respondents, which numbered over 90,000 this year, the survey’s insights help quality teams and the broader industry understand the state of enterprise technology tools and their impact on transformations like CI/CD, DevOps, cloud, and AI.
One of the most significant changes in the 2023 Developer Survey was the introduction of a section dedicated entirely to artificial intelligence and machine learning. Like virtually every other industry, software development isn’t shying away from adopting smarter tools. If anything, developers are eagerly experimenting with AI and machine learning in an effort to further accelerate delivery cycles. Yet old challenges remain: knowledge silos persist, much to everyone’s frustration. Collaboration is a work in progress, but reliance continues to grow for both asynchronous and real-time communication. And of course, some classics never go out of style, especially for programming languages. Let’s take a look inside the world of developers, according to Stack Overflow’s 2023 Developer Survey.
The new section of the Developer Survey asked how respondents felt about AI tools in the development process, and what AI tools (if any) they were using.
Less than half of professional developers (44.17%) reported using AI tools in their work, with another 25.88% saying they planned to start using AI tools “soon.” Roughly 30% shared that they weren’t using any AI tools to support their work and had no plans to do so. Of those using AI tools, the vast majority (82.55%) reported harnessing AI to write code, followed by debugging (48.89%), documenting code (34.37%), and learning about their codebase (30.1%). Less than a quarter of respondents (23.87%) said they were using AI tools for software testing.
Most interestingly for quality professionals, trust in AI tools remains significantly lower than adoption. Just 2.85% of respondents say they “highly trust” the accuracy of AI tools, though 39.3% “somewhat trust” them. Considering that one-third of respondents cite productivity gains as a key benefit of AI use, the need for greater oversight at greater scale is clear. According to the survey, developers appear to agree: 55.17% said they were interested in using AI for testing, the highest level of interest across all use cases. AI is clearly seen as a productivity booster for many developers, and quality teams have an opportunity to play an important role in ensuring code quality through this transformation.
Read about AI and machine learning skills for quality engineering.
Despite the wave of AI adoption and its potential impact on pipeline productivity, some challenges remain the same. When asked about the factors limiting their productivity, most cited knowledge silos and cross-functional collaboration as routine challenges. 38.19% of developers reported spending 30-60 minutes per day searching for answers or solutions to problems, with another 17.83% spending 60-120 minutes per day. This means that a plurality of engineers are burning almost a full day per week searching for information, and roughly 1-in-5 developers are spending 1.5 work days, which adds up to a significant amount of time in two week sprint cycles.
Quality teams are likely already familiar with these types of delays. QA and developers are increasingly collaborating in tight-knit testing and feedback cycles as part of shift-left and continuous testing efforts. Developer delays in finding solutions are also quality assurance delays. The issue doesn’t appear to be a lack of collaboration tools: 58% of developers reported having access to Jira at work, 37.73% use Confluence, and approximately 20% use Notion and Trello, with a long list of other asynchronous work tools supporting respondents. Similarly, 51.71% of respondents use Microsoft Teams, 47.59% use Slack, and 45.25% have Zoom to support real-time collaboration. Yet these disconnected tools aren’t enough to help development teams quickly find the information they need. Solving knowledge silos is a shared challenge and a shared opportunity for impact.
Discover how collaborative testing can help solve development silos.
The good news is that even though knowledge silos and cross-functional collaboration still plague development organizations, QA and developers still share a strong love for JavaScript. For the eleventh year in a row, JavaScript was named the most commonly used programming language. 65.82% of professional developers reported using it at work. HTML/CSS narrowly edged SQL out for second place, with 52.83% of developers relying on the former and 51.52% using the latter.
JavaScript has long been a popular choice for quality professionals. The language has been around since the 1990s, is a common language for aspiring programmers, and is a staple of low-code and legacy test automation frameworks. Its continuing popularity among developers is a positive indicator for experienced QA engineers who already rely on JavaScript in their day-to-day work, as well as useful knowledge for manual testers starting to learn automation and programming concepts with low-code test automation frameworks.
Explore how low-code test automation can help testers learn new skills on the job.
Continuous integration (CI) and continuous delivery dominated tool rankings, with 71.93% of professional developers reporting having CI/CD available at their organization. Echoing mabl’s Testing in DevOps Report, the Stack Overflow Developer Survey found that automated testing and DevOps tools had near-equal levels of adoption at 60.79% and 60.45%, respectively.
The neck-and-neck use of test automation and DevOps points to the growing role that software testers are playing in DevOps adoption. Unless an organization can break down the Waterfall-era silo between development and testing, they’re likely to lose momentum before fully adopting DevOps. Adopting the right test automation framework is essential for making the most of DevOps tools.
But the most important quality trend in the tools section of the 2023 Developer Survey is hidden between the data points. Though the majority of developers have test automation tools, testing ranked highest in developer interest for applying AI. There’s clearly room for improvement in testing tools, especially when it comes to applying AI and machine learning. As they look to mature their quality engineering and software testing strategies, quality teams should consider how to capitalize on developer interest in intelligent test automation.
See why mabl is a 4-time winner of AI Breakthrough’s Best AI-Based Solution for Engineering.
Software is getting closer to being built in a day, but the tools that unleash that velocity are still evolving. As the role of software testing becomes elevated in the enterprise, quality leaders should find ways to improve the work lives of testers, QA engineers, and developers. Exploring the insights from Stack Overflow’s 2023 Developer Survey is a valuable look into what development teams need from their partners, their tools, and the future of technology.