Stack Overflow has released its groundbreaking 2023 Annual Developer Survey [1]. The report surveyed more than 90,000 developers to provide a comprehensive view of the current software developer experience. Next, this article will highlight several key findings, namely important programming language and tool preferences, the use of artificial intelligence in development workflows , and what these trends may mean for the DevOps field .
Developer technology preferences in 2023
Survey results show that more and more developers are learning to code online . The proportion of people using online resources to learn code will increase from 70% in 2022 to 80% in 2023. While many developers (47%) still hold a bachelor's degree in computer science or equivalent, these trends highlight a move toward other knowledge solutions, especially for younger programmers. Popular online resources include technical documentation, blogs , forums and how-to videos.
JavaScript remains at the top of the list - it has been the most commonly used programming language for 11 consecutive years. It is worth emphasizing that Python has replaced SQL as the third largest programming language . "SQL has been in the top three (JavaScript, HTML/CSS, SQL) since 2015, so it's a big deal that it fell below Python," Liuzzo said. "We've seen Python's popularity increase based on the number of questions on our public website, so we've been anticipating some changes."
TypeScript and Bash/Shell usage has continued to grow over the past few years . These two languages involve features of other popular programming languages, so they are very popular among programmers.
PostgreSQL has also surpassed MySQL to become the most commonly used database type . When it comes to web frameworks, Node.js and React.js are the main ones . Others, such as jQuery and ASP.NET frameworks, tend to be outdated, probably because they are older web frameworks.
Focus on artificial intelligence
The explosion of new AI innovations, such as large language models (LLMs) and chat-driven generative AI tools , have had a significant impact on technology discoveries this year. In fact, 83% of respondents have used ChatGPT in the past year. Followed by Bing AI (20.6%), WolframAlpha (13.36%) and Google Bard AI (9.86%). GitHub Copilot ranks as the most commonly used artificial intelligence development tool.
There has been a clear surge in the use of AI tools among those learning to code, often focusing on benefits such as faster learning and increased productivity and efficiency. Combine this with how they currently use AI tools for debugging and getting help (68%) and understanding the code base (50%), and the common thread found in the report is that AI tools promote learning .
Still, people are skeptical about the accuracy of AI, with only 13% citing improved coding accuracy as a benefit of using such tools. While these pain points may be resolved as LLM evolves, for now, human judgment is still required to catch errors and avoid misuse.
What impact these trends may have on DevOps
So what are the implications for practitioners in the DevOps field, namely DevOps tool providers and platform engineers? New tool preferences and habits will continue to evolve in the DevOps space in new ways. Here are some examples to consider:
-
More Markdown-based collaboration . Markdown is the hottest asynchronous tool, and its popularity continues to grow this year. New programmers are incorporating Markdown, GitHub Discussions, and Notion into their asynchronous tools. Various versions of Markdown have been used for comments, merge requests, issues, and more.
-
Continue to rely on Docker . Docker remains the most used tool, followed by npm and Pip. Despite the rise of managed container services, Docker remains a popular way to package and distribute software.
-
The pattern of the three giants remains the same . Amazon Web Services (AWS) remains the most used cloud platform, accounting for 48.62%. Followed by Microsoft Azure (26.03%) and Google Cloud (23.86%). It's worth noting, though, that newer developers want to make more use of Hetzner and Vercel.
-
Online learning is critical to knowledge sharing . Given the steady growth in the number of new programmers and their reliance on online resources rather than traditional schooling, the DevOps and open source communities should work to increase their online presence by making knowledge such as tool documentation and tutorials easily accessible.
Finally, generative AI will enhance programming capabilities in many areas and is already embedded in many environments. Just as AI can be used for code generation and debugging, LLM may also assist with DevOps efforts, helping to automate CI/CD pipelines, optimize infrastructure, detect anomalies, and provide more chat interfaces to easily launch these functions.
“AI tools are enhancing learning capabilities and making it easier to learn emerging technologies,” Liuzzo said. “Generative AI will lower the barrier to coding and allow the developer community to grow exponentially. More and more developers Humans will use the tools and validate and validate the output as they learn, bringing together the power of the developer community and the technical power of AI.”
Original link :
https://devops.com/key-insights-from-the-stack-overflow-2023-developer-survey/
Reference link :
[1] https://survey.stackoverflow.co/2023/#overview