Xia Li-yao

I'm happy to keep this library working.

This would be very nice to have for Coq projects on gitlab to render better.

I am in favor of adding this function, agreeing with @chris-martin's motivations above, and particularly: > it is difficult to explain why among the three major types (strict text, lazy...

I also find those three classes reasonable enough (`ToText`, `ToLText`, `ConvertUtf8`). I think there can be other ways than laws to easily grasp what is going on with a type...

> will absolutely shred type inference without littering your code with explicit types or `@`-nnotations. Yes, it means people have to write slightly longer names - so what? Don't these...

Thanks for clarifying. So it's more about MPTC than lawlessness, but I'm still not sure how to understand this issue you mention > I've had cases where even completely known...

I would guess historical reasons, like that `fail` expects a `String`, or that libraries before `text` probably used `String` already for error messages. There may be value in making `Text`...

The motivation for these stylistic changes is quite subjective, it doesn't seem strong enough to merge this PR.

@Anton-Latukha It is unfortunate that we started with two disagreements, but I have nothing against you personally either. It is my job to vet changes, and I voiced my opinion...

This is a very benign breaking change. People push back against changes that cause unnecessary churn, but nobody depends on `replace "" _ _` being undefined. I will try to...