Take your plots from drab to fab! #ggplot2

I got introduced to @randyau.com's 'Data Cleaning IS Analysis, Not Grunt Work' post during the #dataBS Conf this week: www.counting-stuff.com/data-cleanin... . I just finished--it was a great read. Here are some quotes and thoughts I'm walking away with 👇 1/9 #RStats

Enjoyed Hadley's keynote at Posit 2025. But disagreed with the statement that AI can't do maths. My understanding (& testing) of Claude.ai is that it can. If it identifies it's a maths question, it runs code in the background, and then provides the answer back in natural language #rstats #posit2025

Got a #Shiny app idea but don’t know where to start? 🤔Shiny Assistant can help you get unstuck and start building: shiny.posit.co/blog/posts/s... #RLang #ShinyApps #DataScience

Posit::conf(2026) will be in Houston, Sept. 14-16 - announced now at #Positconf2025 #Positconf2026 #Positconf #RStats

I am beyond excited to announce that ggplot2 4.0.0 has just landed on CRAN. It's not every day we have a new major #ggplot2 release but it is a fitting 18 year birthday present for the package. Get an overview of the release in this blog post and be on the lookout for more in-depth posts #rstats

Playing around with some new annotate_* helper functions in development #rstats

I'm creating a bunch of flexible themes for a mega update of #ggblanket.. They're all the same, but different defaults for colours to make things easy. I've currently got a grey, beige, light, and dark one. Then I wondered about alternative grey/beige options with white gridlines. Thoughts? #rstats

I really love trailing commas. You often never know how many args you need when you start, and it makes it easier to incrementally add to it. Think all DEVs should support where they can #rstats

I've developed a bunch of useful functions that I have as extras in #ggblanket. They're just #ggplot2 helpers really. I'm not sure if they should ideally be in a separate package? There's not heaps. I'm not sure if a new package might be hard to get on CRAN, and I'm low on energy for that #rstats

New #rstats blog post: Deep Dive into ellmer: Part 1. I explore the source code behind ellmer's starter workflow - creating a new Chat object and calling the $chat() method. www.howardbaik.com/posts/deep-d... Thanks to @hadley.nz , @grrrck.xyz, @atheriel.bsky.social, and others for 🐘 ellmer 🐘 !

#AcademicLife in Ukraine: preparing for lectures at night during a massive missile attack on Kyiv in an overcrowded underground station #EconSky #UkrainianView #RStats

Ross Ihaka, co-creator of #Rstats, reflects on growing up Maori 🌿 in rural New Zealand 🇳🇿 and how he discovered his passion for data youtu.be/LY3FOLZ_Zpk?...

Hey #rstats, #pydata, #julialang & #dataviz folks! 😊 What are some iconic data visualizations, similar to @edhawkins.org‘s warming strips & climate spiral?

🌲 New R package: forestdata makes it easy to download forestry and land cover data from multiple sources (Copernicus, ESRI, EU-Trees4F, and more). Supports sf, SpatRaster, and tidy outputs. Explore it here: cidree.github.io/forestdata/ #rstats #rspatial #forestry #landcover

Introducing my new #rstats package {kitchensink} Not sure what the right model to fit is? Should you allow random intercepts, slopes, both? What do Bayesian methods say? Just call {kitchensink::throw} to fit every possible model and see how your results differ!

Hello #rstats, I'm searching for work opportunities and wanted to put out a friendly inquiry. My work has focused on shiny and package development, github.com/nteetor/. I would appreciate any advice or leads and I am happy to talk more. Thank you!

I feel like maybe I'm an #rstats anomaly, but I'm finding LLMs particularly Claude.ai an amazing help

Hi #RStats & #PyData people! 😊 How do you like to plot 𝘥𝘪𝘳𝘦𝘤𝘵𝘦𝘥 𝘢𝘤𝘺𝘤𝘭𝘪𝘤 𝘨𝘳𝘢𝘱𝘩𝘴⁣ (𝐃𝐀𝐆⁣s)? 🤔 I'm looking for someting like this 👇 Links to blog posts & shameless self-promotions are very welcome! 🤗

For complete themes, there is a #ggplot2 convention to name them with a theme_ prefix. For when you have a bunch of theme args that do something, but it is not a complete theme (nor is it intended to be), is there a naming convention?? #rstats

🦆 My new R package {avilistr} is live on CRAN! 🎉 The first unified global bird taxonomy (11,131 species) is now accessible to #rstats users. I built it this weekend after Cornell Lab's AviList release last week! 🪿 Blog: www.dalyanalytics.com/blog/avilist... 💻 GitHub: github.com/dalyanalytic...

Congratulations to Matthijs Berends and team for the release of the AMR for R package version 3! I've been using this package to produce the AGAR Kids reports and can highly recommend. amr-for-r.org #Rstats #AMR #IDEpi #MedSky

We don't record all #ggplot2 extenders talk, but do have recordings from a few influential developers!🎥📊 - Hassan Kibirige - plotnine - Winston Chang - ggproto and R Graphics cookbook - Hadley Wickham - thoughts on the ecosystem Links on meeting page! ggplot2-extenders.github.io/ggplot-exten...

Trying to create a function to help with standard widths for bars across #ggplot2 plots into #ggblanket. I got it to a point where it works with position of identity or stack. But it doesn't with position_dodge() or position_dodge2(). Can anyone help? PR please! #rstats github.com/davidhodge93...

v12.4.0 of #ggblanket just pushed to CRAN. A couple of API tweaks (sorry!). New feature of bind_each_all, which I'm VERY happy with :) Enables easy plotting of Each category and an All value. You can use this function with #ggplot2 too. Give it a go! #rstats davidhodge931.github.io/ggblanket/re...

What is dead may never die! #rstats

I liked the old dplyr logo _way_ more than the one used now #rstats

Got a new function in development #ggblanket called bind_all_by. It's designed to support plotting all data against group data. Like {gghighlight}, but different. You can use this function & method with normal #ggplot2 code too. Feedback welcome #rstats davidhodge931.github.io/ggblanket/re...

For naming function arguments in #rstats, is it better to use x or var etc to refer to a single unquoted variable name? Trying to align with tidyverse syntax select_one <- function(data, x) { data |> select( {{ x }} ) } select_one <- function(data, var) { data |> select( {{ var }} ) }

While R is far from “dead” I do think we in the #rstats community need to consider this chart from @jscarto. R has seen remarkable growth over the past decade, but has declined since 2022. Why is this the case? Three notable things happened in 2022:

Wondering if the difficult process of getting a package on cran contributes to package bloat. E.g it feels easier to add functions to an existing package than to put a new one on cran #rstats

Mastering typefaces and fonts in #rstats has always been harder than it should. I have tried to collect much of my relevant knowledge in this deep-dive blog post so you can spend your time picking the right typeface instead of cursing at the computer

Another #ggblanket extra is the annotate_axis_line function. This is a hack that enables you to make your axis line sit behind any geom. Basically create the plot using a blank geom with gg_blanket, add the function which removes the axis and adds an hline in, and then add your geom in #rstats

Within every sufficiently advanced #rstats project workflow lurks a slow and bug riddled implementation of targets.

So fonts and #ggplot2.. I think I haven't understood completely how the font_add_google function works. I thought I just go to fonts.google.com & find a font & it should work?? #rstats font <- "Roboto" #works font <- "Tagesschrift" #doesn't font <- "Tuffy" #doesn't sysfonts::font_add_google(font)

A useful 'extra' in #ggblanket v12.3.0 is the `aes_contrast` function. It enables the colour of text or labels etc to be determined automatically for contrast based on the fill. This function is based on code developed by @teunbrand.bsky.social #rstats github.com/teunbrand/gg...

Do any #ggplot2 experts know how to make a y axis like this? Or if it's even possible? I've been noticing online a few graphs using a scale in this way #rstats

Fantastic news that R 4.5.0 has added the wonderful {palmerpenguins} practice dataset! But did they really have to change the variable names?! #rstats

"But, #tariffs are supposed to be taxes on other countries!!!" cried every sniveling MAGA who will see their cost of living shoot through the roof. If you voted for Trump based on "the economy," you were either an idiot or lying to yourself. #Trumponomics #Trumstagflation #Trumcession

Who voted for this idiot??

Today I learned how to use the {tidyxl} and {unpivotr} maintained by Duncan Garmonsway. These packages are absolutely brilliant! They solve a massive problem with trying to deal with hideous excel spreadsheets. Think headers everywhere with merged cells. This is like a supercharged {tidyr} #rstats

What's driving the Trump administration? Evil or incompetent? Krugman's answer of both sounds about right paulkrugman.substack.com/p/incompeten...

Well, they lied. Obviously.

Cool to see a Twitter post from someone else promoting {ggblanket} #rstats

I feel confident in predicting history will judge this decade or so as the absolutely stupidest time in American history. Honestly our kids and grandkids will read about this time and be shocked that adults existed.

To add saved #ggplot2 images unscaled into quarto, a way that works for me is to use knitr::include_graphics. The key here is to ensure the dpi is the same as the image. ggsave uses 300 by default. claude.ai helped me sovle this #rstats

One more question, good #rstats folk: how do you add a image into quarto with no scaling? I.e. the width and length of the image stay exactly the same. My purpose is to add #ggplot2 plots to a document and not scale them at all #rstats

This is a better way to specify constant panel sizes, if you can use the Dev version of ggplot2. It works the same way for faceting or non-faceting. Thanks @teunbrand.bsky.social for the fantastic new feature! #rstats #ggplot2