David Hodge
@davidhodge931.bsky.social
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
September 22, 2025 at 4:40 AM
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
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
September 3, 2025 at 2:55 AM
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
Just googled.. Like this? Looks like it'd take a bit to get used to! Think I'll stick to trailing
September 2, 2025 at 1:24 AM
Just googled.. Like this? Looks like it'd take a bit to get used to! Think I'll stick to trailing
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
September 2, 2025 at 12:59 AM
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
August 5, 2025 at 5:14 AM
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
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...
June 3, 2025 at 6:08 AM
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...
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...
May 27, 2025 at 8:51 PM
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...
Could be to do with the releases?? dplyr looks more like it's stabilized than decreased
May 26, 2025 at 12:26 AM
Could be to do with the releases?? dplyr looks more like it's stabilized than decreased
You can also use this function with ggplot2 code too
May 8, 2025 at 9:50 PM
You can also use this function with ggplot2 code too
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
May 8, 2025 at 9:41 PM
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
Still doesn't seem to always work for me :/ Was trying to use this font for example.. fonts.google.com/specimen/Com...
April 30, 2025 at 12:28 AM
Still doesn't seem to always work for me :/ Was trying to use this font for example.. fonts.google.com/specimen/Com...
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...
April 28, 2025 at 9:49 PM
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...
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
March 19, 2025 at 9:40 PM
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
Think I've just learned how to make plots have a specified _panel_ size. For non-faceted plots, use patchwork::plot_layout. For faceted plots, use purrr::map with patchwork::wrap_plots. This right? #rstats #ggplot2 #patchwork
March 18, 2025 at 11:14 PM
Think I've just learned how to make plots have a specified _panel_ size. For non-faceted plots, use patchwork::plot_layout. For faceted plots, use purrr::map with patchwork::wrap_plots. This right? #rstats #ggplot2 #patchwork