Larefly

Hi all, I took a stab at a possible solution. The approach I took was to pass `httr::write_disk` to `httr::VERB` when a output location is given. **Edit:** Updated `issue` function...

> @DyfanJones I've already forgotten how httr works, but the chances are it'll be much easier with [httr2](https://github.com/r-lib/httr2) @hadley no worries, from a quick look at `httr`, it should call...

PR (#458) has been merged and will be in the next paws release.

Hi @wlandau and @davidkretch, I am happy to become a maintainer for paws. I am still getting to grips with the backend of the SDK. Any maintainers documentation would be...

Hi @davidkretch, that sounds great. Congrats on your new job and move hope it is going well. I would be very grateful for any PR reviews as I would like...

I believe so, from memory each category needs to be smaller than 5Mb (cran package restriction). Will run `make check` to see what our current status is for each package...

Here is a summary from the check: ``` paws.machine.learning: - Total: 5.1Mb - R: 2.0Mb - help: 3.0Mb paws.satellite: OK paws.compute: - Total: 7.2Mb - R: 3.0Mb - help: 4.0Mb...

After splitting the big packages it looks like we aren't out of the woods just yet ``` paws.machine.learning: OK paws.machine.learning.p2: OK paws.satellite: OK paws.compute: - Total: 7.2Mb - R: 3.0Mb...

Will explore the sub category option previously discussed. For completeness, large categories will be split into sub categories. The category then will then reference the sub-categories similar in how `paws`...

Split categories into ~ 700 operations