chris <[email protected]> writes: >> Preliminary ideas are the following: >> >> 1. First-time contributors should be discouraged to use LLM > > I completely understand the problem of having to sift through large amounts > of > poor code that arrive frequently. > > But how can you tell whether the code was generated by an LLM?
I have seen too much of LLM code and texts myself. In many cases I can recognize it. (For example, some of the recent patches likely used LLM assist for documentation, but I judged it as acceptable and did not raise any concerns). In any case, "discouraged" is not the same as "prohibited". What I want to say in (1) - "please write the code yourself first time, for your own learning". > The problem with LLMs is that poor code can easily resemble good code, which > makes triage more difficult. That's not true. LLM code written out of poor instructions, can only resemble good code until you actually sit down and read it, and not just scroll (or run without much testing). Sometimes, LLM code can contain subtle errors related from incomplete understanding of the codebase. However, I do not expect human authors to have a complete understanding of Org codebase. There is no significant difference for me in this regard. -- Ihor Radchenko // yantar92, Org mode maintainer, Learn more about Org mode at <https://orgmode.org/>. Support Org development at <https://liberapay.com/org-mode>, or support my work at <https://liberapay.com/yantar92>
