Peter Ljunglöf

Regarding Haskell vs Shell: I would prefer non-dependence on Haskell, if it's not too complicated. It would be nice if RGL authors do not have to install Haskell.

I have a solution too! 🥇 I don't suggest the Haskell way: to require parameters to be capitalised and variables lower-cased. But I suggest that we make it explicitly recommended:...

> I like the idea of warnings to encourage conventions. > Are you saying this should be implemented in the compiler for all GF grammars, or somehow targeted only at...

> You're saying this is "severe" as if it's a bug in GF, but in truth it is a feature of the language. That it can lead to hard to...

> Now I see that the overshadowing warning is already there ;-) Maybe we should just promote this to an important warning and tune down some less important ones (e.g....

I'm perfectly ok with a more complicated heuristics for generating warnings. Here is another alternative: - (my original suggestion) warn if a parameter is defined lowercased, and if a variable...

> Sorry for a mistake in my previous comment: this is not a bug, because the pattern is reached by for instance. Hence the overshadowing analysis is not sufficient to...

If the PR is about a new functor, the process must be even more strict, because functors influence so many languages.

> On this note, I am wary of using PRs a bit too superfluously. Specifically, updating a README like can surely just be committed directly into main by someone with...

Also, the languages do not use standard names or abbreviations: - Kikamba is called Kamba in Glottolog and Wikipedia - but the abbreviation "Kam" is at least the ISO language...