Ben

What about keeping these traits around, but marking them as deprecated with a link to an issue documenting the context of its deprecation and planned re-introduction? If it's deprecated in...

I had a look at the semver documentation, something I really should have done first, sorry. I initially read your response wrong (read it as something like "removal by means...

The problem arises when you want to apply constraints such as length. This fix addresses a limitation of compile-time constant generics, but doesn't fully leverage the type system. For example,...

I wasn't familiar with that as an option. I can see it coming in handy, but I still hold the view that there is merit to type-level expression of length...

I've updated with the following changes: - removal of the deprecated `const_len` - un-deprecation of the `len` function - un-removal of `const Self::LEN` - putting `typenum` and its use behind...

Big issues with the documentation. For me, I don't like the idea of `docs.rs/frunk` documenting the re-export of `typenum` or use of `Self::Len` without being very explicit that it's only...

> To be clear are you saying that the compiler will error out [...]? You are correct. I wrote "warn" when I meant "error". I'll go with opt 3

CI fixed on my end, and added typenum feature to the published docs. My main concern now is that with the typenum length, HList implementations need to add where-clauses that...

...hold off on morging for a bit. I'm getting a feel for using it in practice.

....there's a problem: ```rust impl PWMPinSetter { fn set_pin(dc_list: HCons, pins: HCons) { pins.head.set_duty_pct(dc_list.head); PWMPinSetter::set_pin(dc_list.tail, pins.tail) } } ``` ...that's pretty damn close to what I want, but I get...