Benjamin Tyner

The help page says "These functions compute various weighted versions of standard estimators." therefore for wtd.var, the goal is to estimate a population variance. However, if the sampling weights (probabilities)...

Good point Frank. Is sampling probability also known as inclusion probability? https://en.wikipedia.org/wiki/Inclusion_probability

So that takes us into the realm of stratified sampling. I suppose we could refine wtd.var to accept inverse probability weights...would it also be desirable to also handle second-order inclusion...

Agreed that multi-stage is beyond the scope here. By second-order inclusion probabilities, I mean the pairwise probabilities of inclusion for a single-stage design. For other software: did you already try...

Absolutely in favor of deprecating and removing obsolete code. Especially if it gets shiny out of the doghouse.

Any update on this? If not, when might we reasonably expect resolution?

Alternatively, could call `s3fs::s3_file_exists()` which will return a logical value (or give an error if the permissions prohibit access). Also it is vectorized!

I concur with flodel. Also, dispatch_test seems to cause the file to be read twice: ``` function (test, file, parse.data = getParseData(parse(file, keep.source = TRUE)), lines = readLines(file), quiet =...

I wonder if this is related to this: https://github.com/ianmcook/implyr/issues/57

@DyfanJones unfortunately `profvis` depends on `purrr` which depends on R version >= 3.5.0 per [this](https://github.com/tidyverse/purrr/blob/fe11742256d7bebe4912a40d4ee21998d3e4109b/DESCRIPTION#L16) and I am using R version 3.4.3 because that is what ships with Amazon Linux...