Ihor Radchenko <[email protected]> writes: > David Masterson <[email protected]> writes: >> My goal was enlisting the LLM in helping keep the developer community >> healthy by better explaining its code/patches in ways that previous >> developeers never could (or could spend the time to) and, thus, teach >> those that come after. > > I did not get this paragraph. Could you elaborate?
When I suggested requiring that code/patches created for free software by LLMs be done in the fashion of Literate Programming, I thought the following: 1. In order to explain the code to humans in a literate manner, the LLM will not be able to take unnecessary "shortcuts" that may make the code harder to understand or the code will be rejected. 2. New software programmers who take an interest in a piece of free software, but do not have (or do not want to have) an LLM to help them interpret the code, will have a literate explanation of the code to begin their exploration. 3. LLMs should be able to not only generate maintainable code, but also maintainable documentation in a literate style, thus, promoting a synergy between developers using LLMs and those who aren't. 4. New developers can also add their thoughts to the literate programming documentation as they enhance the code. 5. LLMs later enhancing code will now also be able to see the goals of the code in the literate programming docs and continue to enhance it and the code accordingly. 6. Ultimately, I see this as a bridge for people (like me) who may want to (from time to time) patch a piece of free software without having the tooling for an LLM to analyze the code. Of course, these are ideas in my head without much experience behind them. Perhaps there are better ways of approaching this? -- David Masterson
