Python has dethroned JavaScript as the most-used language on GitHub. This is mainly thanks to the language’s frequent use for AI and data science. Contributions to specifically GenAI projects alone increased 59 percent this year.
GitHub’s annual Octoverse report shows that Python usage has skyrocketed, with developers worldwide using it to build models, crunch data and generally shape their machine-learning workloads with Python. The number of AI-related projects using Python has nearly doubled from 2023. Other languages, such as JavaScript and TypeScript, also remain popular, and Rust is on the rise.
Small applications
The report tells not only which languages are used the most but also who is building them. There is a lot of enthusiasm for the language in India, Germany, Japan, and Singapore. GitHub’s user base has grown considerably overall, with many newcomers taking their first steps in open source. They often build smaller, custom-built applications without huge computing power requirements. This includes not only AI applications but also web development and software for smart homes.
Tip: Cognizant CEO: “AI will actually create jobs”
According to GitHub, there is little sign yet of the sometimes predicted doomsday scenario that AI would put programmers out of work. “As AI rapidly expands, developers are increasingly building AI models into applications and engaging with AI projects on GitHub in large numbers. At the same time, we’re seeing an unprecedented number of developers join GitHub from across the globe,” the company said.
Python’s popularity also extends to Jupyter Notebooks, which chalked up 92 percent growth this year. A Jupyter Notebook is an interactive document where users can add code, text and other media. Code can be run, tested, and modified in the Notebook and is a visual way to provide insight into what code ‘actually does.’ Jupyter is a contraction of Julia, Python and R, three languages that are very common in computer science and STEM fields.
Lots of growth in emerging markets
Overall, GitHub predicts a growing number of developers in India. That country could even provide the largest share of developers active on the platform by 2028. GitHub also predicts substantial growth in Africa and Latin America, which is unsurprising since there is still a lot of growth potential in such emerging economies.
GitHub Copilot, the platform’s code assistant, can now handle many more AI models. It originally came with the GPT-3-based Codex but has since expanded to include GPT-4, GPT-3.5-turbo, and several “mini” models of the OpenAI offering. Support for Anthropic’s Claude 3.5 Sonnet and Google’s Gemini 1.5 Pro is expected to become available soon. The fact that the assistant can handle more models is good news for users who need specialized models with various quality and latency options.
Also read: Popularity of Python continues to rise against backdrop of AI race