In reviewing some recent PRs I've noticed that our various AIs have gone to town annotating some git tracked tomls, while others have been left untouched. The different AIs seem to have used different conventions for toml file header annotations and/or section annotations. A standardization sweep is in order.
I like the idea of the headers, especially for the examples. The headers can explain the purpose/context of the example and call out any key features of the settings. They should not be overly large - maybe 2-10 lines.
All variable setting lines should have in-line annotations describing what the variable is/does. These are super helpful for new users, who inherit copy and modify the example tomls and do not want to search source code to remember what each variable does.
Section annotations seem overkill, but perhaps I can be convinced otherwise if there are important ones.
In reviewing some recent PRs I've noticed that our various AIs have gone to town annotating some git tracked tomls, while others have been left untouched. The different AIs seem to have used different conventions for toml file header annotations and/or section annotations. A standardization sweep is in order.
I like the idea of the headers, especially for the examples. The headers can explain the purpose/context of the example and call out any key features of the settings. They should not be overly large - maybe 2-10 lines.
All variable setting lines should have in-line annotations describing what the variable is/does. These are super helpful for new users, who inherit copy and modify the example tomls and do not want to search source code to remember what each variable does.
Section annotations seem overkill, but perhaps I can be convinced otherwise if there are important ones.