Chip Hogg

That's a good point, and maybe it's what we'd do if we were starting over. That said, I think we're small enough in scope to get away with one "detail"...

Note that the first step will be to give people a migration path such that they're not specializing `detail::UnitAvoidance`. We need to remove this public-facing usage of the `detail` namespace....

Overnight measurements were not encouraging. It looks like we'd be incurring a roughly constant 25 ms penalty for adding this feature. I don't think the cost/benefit adds up here. It...

I've decided to simply close this instead of making a utility that nobody is asking for. We can reopen it in the future if somebody requests it, and use 3d7aea3...

What a great question! You're right that this behavior was by design. And your argument about `-Wfloat-conversion` is a very compelling reason to revisit that design. Au 0.5.0 added some...

Thanks for calling this out! As you say, there are really two issues here: better error messages for trying to "quantitify" a quantity, and giving our troubleshooting guide a refresh....

@connorjak, check out [this comment](https://github.com/aurora-opensource/au/pull/294#issuecomment-2364882277) to see the before-and-after on the error messages for the implementation that's currently under review. (Fair warning: you'll have to scroll through a _lot_ of...

Side question: for something like `volume / length`, are we doing the Anthony Williams thing of not simplifying numerator and denominator? Or is it just _already_ basically equivalent to `area`?

> One possible problem is to decide which representation type should be used in case we want to allow CTAD (e.g., `quantity{si::standard_gravity}`). From my point of view, the right answer...

Yeah, it's on me to write a doc to explain the concept of "shapeshifter types". These are [monovalue types](https://aurora-opensource.github.io/au/main/reference/detail/monovalue_types/) that can "shapeshift" to become any type from some family of...