Hi Jon, Thank you for the prompt reply and the precious feedback. I realised that I might have sounded arrogant in my previous message, and I apologise for that. It wasn't my intention. I genuinely want to help and learn in the process. I will search the tickets in the kicad-python's and find a way to be of help, and I will also try to contribute to and learn from this mailing list.
Best, Eugenio. Il giorno dom 13 lug 2025 alle ore 23:03 Jon Evans <[email protected]> ha scritto: > Hi Eugenio, > > I can speak as the maintainer of kicad-python, but for kicad as a whole, > I'm only one of the lead developers and we have not discussed or released a > formal policy on this. > > AI code assistants may be used in many ways, ranging from fairly benign > (prompting one to explain a piece of code to you to help you understand it, > or using one as a refactoring tool to modify an existing piece of code in > straightforward ways) to potentially questionable (generating large chunks > of code or even entire projects, which may have questionable > authorship/licensing implications) and definitely unacceptable (wasting > people's time by posting the incorrect output of an LLM without disclosing > that it is from an LLM, and not having the knowledge yourself to > double-check the output). > > I would recommend using the following guiding principles when considering > using LLMs as part of a contribution to KiCad or any open-source project: > > 1) Do not make any technical contribution that you do not fully understand > yourself. Bug reports, code submissions, or discussion in the form of > comments are all contributions to an open-source project. One of the best > ways to burn goodwill by using AI tools that I have seen is to attempt to > make contributions that one does not understand, resulting in either being > dismissed as nonsense or (worse) wasting the time of actual human > maintainers to try to understand something that ends up being the > fabrication of an LLM. > > 2) Always disclose the use of an LLM in any work you submit, whether > saying "I translated this comment to English using ChatGPT" or "I used > Claude to try to fix this bug, and this is the solution it suggested". LLM > outputs can be very useful as a shortcut to getting work done, but they > have serious limitations: they are not true intelligence, and they will > easily invent solutions that do not work or have subtle problems. The > output of an LLM needs to be scrutinized at least as much as that from an > unaided human, because it is more likely that the human who prompted the > LLM either doesn't fully understand the output, or did not check it > carefully. > > As to whether or not the KiCad project will accept code contributions that > were primarily or substantially generated by LLMs, I cannot answer that on > behalf of the team, it's a complicated question. > > Separate from the AI question, I will also note that it is best practice > for KiCad (and most open-source projects) to discuss with the development > team before taking on the development of new features. This can be in the > form of asking on the mailing list, commenting on an open issue saying > you'd like to work on it, or opening a new feature request issue and > stating that you're willing to work on the implementation if it is > accepted. Starting by forking and writing code might work, but if you do > so without discussion, you risk having your code submission rejected > because it is not compatible with other future plans that you didn't know > about. > > Best, > -Jon > > On Sun, Jul 13, 2025 at 3:19 PM Eugenio La Cava <[email protected]> > wrote: > >> Hi all, >> >> I recently forked both the kicad and the kicad-python repos to work on >> feature that are missing. >> I have a question regarding the use of AI code assistant in the >> development of new features and in the fixing of existing bugs: is the use >> of such tools permitted, tolerated or frowned upon? >> >> -- >> You received this message because you are subscribed to the Google Groups >> "KiCad Developers" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion visit >> https://groups.google.com/a/kicad.org/d/msgid/devlist/0b635111-4421-4c6a-aeea-7d7d515810fbn%40kicad.org >> <https://groups.google.com/a/kicad.org/d/msgid/devlist/0b635111-4421-4c6a-aeea-7d7d515810fbn%40kicad.org?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to a topic in the > Google Groups "KiCad Developers" group. > To unsubscribe from this topic, visit > https://groups.google.com/a/kicad.org/d/topic/devlist/2-mAIjt971c/unsubscribe > . > To unsubscribe from this group and all its topics, send an email to > [email protected]. > To view this discussion visit > https://groups.google.com/a/kicad.org/d/msgid/devlist/CA%2BqGbCAgXpM6xVigUoWQDwsjOOxzOXOb_s%3Dx1hLSKczsZ0ebKg%40mail.gmail.com > <https://groups.google.com/a/kicad.org/d/msgid/devlist/CA%2BqGbCAgXpM6xVigUoWQDwsjOOxzOXOb_s%3Dx1hLSKczsZ0ebKg%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- Dr. Eugenio La Cava Dottore di Ricerca in Scienze Agrarie ed Ambientali - Dr. Eugenio La Cava Philosopher Doctor in Agricultural and Environmental Sciences -- You received this message because you are subscribed to the Google Groups "KiCad Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/a/kicad.org/d/msgid/devlist/CAP-G1zsmXSBmWtoyaucQLc5-TEUTXd1_xNvePgW0KxBAL4DhcA%40mail.gmail.com.
