Carl A. B. Pearson

Maybe something like using the [assume-unchanged](https://mirrors.edge.kernel.org/pub/software/scm/git/docs/git-update-index.html#_using_8220_assume_unchanged_8221_bit) feature in git? Some discussion on [stackoverflow](https://stackoverflow.com/questions/655243/ignore-modified-but-not-committed-files-in-git).

I suspect there are some updates to the CONTRIBUTING + STYLE_GUIDE files that codify / clarify these notions.

A bit more, I think - `coerce_dt` just handles data-like-inputs (which is indeed a fair chunk), but nothing else (nor do I think it should - would muddy up its...

Sounds obviously preferable to me? I do get the desire for completely automated use, but this seems automated enough. About the only nice-to-have tweak would be to require a final...

note re this: currently the public facing function (`rw(...)`) documentation refers to the internal function (`construct_rw(...)`) arguments. Probably makes sense to split into public vs internal elements, but will have...

snip from `rw`: ``` #' @return A list defining the time frame, group, and type with class #' "enw_rw_term" that can be interpreted by [construct_rw()]. ``` ...so not quite inherit...

my brownian motion is a bit rusty, and there may be some explanation gaps here, but it's not clear these are equivalent. please elaborate?

Is there a diagram anywhere in this vignette? Cursory glance suggests no - possible to nab something from the severity monitoring work, perhaps?

I do not, but I recall roughing in diagram on whiteboard in SA - did you snap some shots of that for the proverbial 'gram?

future issue for that + this package: stabilization of the representation of quantile summaries, such that they can be systematically discovered / introspected from outputs rather than hard-coded.