Hi everyone! I have been working on a new idea related to eBPF since last week, and I think it might be interesting. It might help answer the questions: - Can we understand the high-level design and evolution of complex systems like the Linux kernel better? - Can AI help us with what's never possible before? Stop forcing stupid AI to do buggy kernel coding, or just using RAG or fine-tuned models to give wrong answers. We are doing a completely different way: - By carefully designing a survey, you can use LLM to transform unstructured data like commits, mails into well organized, structured and easy-to-analyze data. Then you can do quantitative analysis on it with traditional methods to gain meaningful insights. AI can also help you analyze data and give insights quickly, it's already a feature of ChatGPT. Imagine if you can ask every entry-level kernel developer, or a Graduate Student who is studying kernel, to do a survey and answer questions about every commit/patch/email, what can you find with the results? - The Github repo: https://github.com/eunomia-bpf/code-survey - Some early experiments and reports related to eBPF subsystem: https://github.com/eunomia-bpf/code-survey/blob/main/docs/report_ebpf.md - The commit dataset: https://github.com/eunomia-bpf/code-survey/blob/main/data/bpf_commits.csv This approach is in early stages, but it's general. I'll be at the LPC and OSS Summit and would love to chat if you're interested! Also, if you’re uncomfortable with your data being used, let us know, and we’ll filter it out. (I just came across the idea last week, and didn't get the early experiment results until yesterday. It seems LPC CFP is closed...?) Best, Yusheng
