Can anyone explain why Kamala Harris hasn't download my R package from CRAN yet? I understand she's on the campaign trail, but all she needs is an internet connection. Technical difficulties? #rstats

#rstats friends, I have a question: given that I have some built R package on disk, what is the blessed/easiest way can I (programmatically) check if it's in source or binary format? Beyond just checking the file extension.

Back in November I did a webinar for the @NHSrCommunity on "Customising your R experience". The video is now up: #rstats.

New categorical colour scheme just dropped 🎤🤩 #rstats.

RT @NHSrCommunity: 📢 November Webinar 📢. We are very excited to announce that the incredible James Laird-Smith will be presenting a ‘Custom….

This Wednesday (Nov 29th) I'm excited to be giving a webinar to the NHS-R Community on "Customising your R experience"! It's going to be at 1pm GMT. Registration link is here, but it will also be recorded. #rstats

At 5pm (BST) on Thursday I'm going to be a guest on.@posit_pbc's Data Science Hangout speaking about the work I do for the @bankofengland and supporting its 1000 R users. #rstats There's no need to register, you can join the stream using this link:

Is it wrong that, when the NYT crossword clue is "something to stub", my first thought is 'function' and not 'toe'? 😅 #rstats.

There's a lot of be said for removing cognitive overhead for developers. By using withr I'm saved the decision of where to put temporary directories or remembering to reset compiler flags. A huge thank you to the authors for this very sleek and amazingly useful package!.

Here's an example where I create a function to install a package in a temporary directory, with temporary libpaths and with temporary compiler flags. And it's just so straightforward! I don't need to worry about global state because I know withr will take care of it for me.

I've been doing a bunch of #rstats package 📦 install testing this week. And the withr package (as usual!) just makes the experience so much easier and cleaner. 🤩👇

After a hard day of wrangling data from APIs, I'm very seriously considering naming my first child `tidyr::unnest_wider()`. I mean, just look how easy it makes converting highly nested JSON into a tidy data frame?! 🤩 Friday #rstats fun!

Have you ever wanted to generate random but readable strings? Like for files or folders? Today's Friday #rstats fun is thanks to @rgfitzjohn for the ids package📦 and its wonderful `adjective_animal()` 😻function.

Today's Friday #rstats fun: I discovered the (experimental) `tree()` function in the lobstr package for the pretty-printing of lists. So much cleaner than `str()`! 😍🗒️ Thanks @NicholasStrayer!

R: What if everything was a vector? #rstats.

Friends who use #rstats: if somebody was interested in making Quarto available from within one's organisation (there's a whole process), what would be an appropriate version to start out with? @fly_upside_down @dragonstyle @scheidegger @chrisderv.

So if you are looking to set some code to trigger in your .Rprofile, but only if you are in the R world's favourite IDE, you can do so most reliably with `commandArgs()[[1L]] == "RStudio"`.

`Sys.getenv("RSTUDIO")` doesn't work either, because it gets set even when you open R from the terminal pane in RStudio.

Having certain packages installed will do it (eg. rstudioapi), but you are then dependent on that package being installed to make the logic work.

For example, `"tools:rstudio" %in% search()` doesn't work, because (I guess) that environment hasn't been attached yet at the point one's .Rprofile is being processed.