Shane F. Carr

@erights It was suggested that we bring this PR to TG3. Could it be added to an upcoming TG3 agenda?

TG2 conditional approval; we will defer to TG1 on exactly what bounds to use: https://github.com/tc39/ecma402/blob/main/meetings/notes-2025-09-11.md#normative-increase-limits-on-intl-mv-and-explicitly-limit-significant-digits-1022

Discuss with: - @robertbastian - @Manishearth - @echeran Optional: - @sffc

I like the marker type approach; but can we be more explicit about how we specify the marker types? For example, we already have `pub struct CanonicalCombiningClass`. Do we add...

I proposed one solution in https://github.com/unicode-org/icu4x/pull/5101. What other solutions might be considered?

> @sffc I'm confused, this issue _is_ proposing the other solution. It's listing out the macros that could potentially be written. But I wrote these macros in the PR, and...

In terms of what's needed: honestly just `ref-cast` functionality that works across all standard pointery types solves ~60% of the call sites (by personal recollection; the exact list can be...

I wasn't distinguishing between ULE and VarULE. I don't immediately see why that matters for the design of the solution: `struct U8Wrap(u8)` and `struct StrWrap(str)` both need ref-cast impls from...

Some general principles for a wrapper type `Foo(T)` (ULE only, not VarULE): | | `T` impls `ULE` (like `u8`) | `T` does NOT impl `ULE` (like `u16`) | |---|---|---| |...

One thing that might be making properties special is that there are some properties that are `u8` and others that are `u16`, so then we need to decide whether to...