That’s different. The warning on head appears in red because it’s the result of an actual WARNING pragma that emits a warning if someone uses head and has x-partial enabled.
IIRC last doesn’t have one since accessing the last element of a list can still be a reasonable operation without an unambiguously better total alternative
1 Like
I understand. And it still is bad.
I would propose to at least mention that mod returns the remainder of the division.
Oh I see, you’re objecting to prescribing “a curated list of words” for all documentation of anything in the Haskell ecosystem. I agree. I wasn’t thinking of prescribing it for everything, I was thinking of first-party components, like GHC and base. But I realise even that is too strict when we have concepts like “semigroup” in base.
So unfortunately I’m no closer to knowing whether you object to replacing less familiar with more familiar words on the grounds that the less familiar ones are nonetheless in the dictionary.
By the way, I would love to find a better name than “handler” for the Bluefin concept, but I have no idea what to call it.
I think it is fine to use words like semigroup as long as we define what that means.
I recommend this great talk (although it is not super relevant to this discussion on documentation):
2 Likes
So are you objecting to the change rather than the choice of words per se?
Can you explain more of your reasoning? It was confusing enough for a user to create a Discourse account to report the issue, and that’s strong evidence to me that it’s worth considering improvements.
Thinking one word is supposed to be another seems like an archetypical example of confusion!
We’re not! We’re discussing on Discourse. Would you like us to stop discussing it here?
(And by the way, documentation changes don’t require involvement of the CLC.)
2 Likes
@egmaleta, your experience suggested to me that an improvement was possible, so I’ve raised a merge request:
3 Likes