iiinova — where your digital self keeps evolving

A refresher in Boolean algebra, which broke our brains for a sec but we got there: !(A & B) is equivalent to !A | !B !(A | B) is equivalent to !A & !B

Two Venn diagrams depicting De Morgan's law.

121 2

Single column data frames will return just the content of that column as a vector. To preserve dimensionality, use drop = FALSE when subsetting, particularly within functions.

R code highlighting the difference between df[, "x"] and df[, "x", drop = FALSE]. The first returns an integer vector, and the second returns a data frame.

R only stores the first and last number of sequences made with a colon : , so all sequences are the same size, no matter how large!

R code that says:
obj_size(1:3)
#> 680 B
obj_size(1:1e3)
#> 680 B
obj_size(1:1e6)
#> 680 B
obj_size(1:1e9)
#> 680 B

Objects don't have names, names have values. A name is a reference to a value, and the value isn't copied unless it is modified.

R code that says x <- c(1,2,3) and y <- x with a comment saying #Creating y does not copy c(1,2,3) but creates a second binding to the existing object

I then tried #PositronDatabot, and asked it to explore the processed data and make an interesting plot. The figure was GREAT off the bat tbh, but I iterated with it to tweak the colors/theme and add crab/sturgeon icons 🤓 The best part was working on icon sizing (for a moment we had BIG CRAB)

An in-process snapshot of the dot plot where a crab icon meant to replace a single data point is hilariously large, obscuring half of the plot.

100 1

Much of my research focuses on #fisheries, so I couldn't help focusing on seafood product tariff rates for my own Week 17 #TidyTuesday submission 🦀 🐟 I also used it as a means to play around with some of the #AI tools in Positron! A bit about my process below ⤵️

A dot plot showing the average most favored nation (MFN) US tariff rate for seafood groups.

1152 15

Echelle was also one of the first in our group to adopt #targets, a Make-like pipeline tool for R 🎯 She put together this great guide to targets, including how to start converting your existing #rstats code to use this powerful tool! Tutorial: echelleburns.github.io/targets-tuto...

A line of code that reads "targets::tar_visnetwork()" and its output. The output is a visual network graph of targets and functions.

280 8

Fisheries data expert Echelle Burns created a 💯 tutorial for the `marlin` bio-economic modeling package (dev by @emlab.ucsb.edu alum @danovando.bsky.social!). We're using this package to explore the potential effects of Marine Protected Areas 🎣 Tutorial: echelleburns.github.io/marlin-tutor...

Rasters of spatial fisheries catch data to represent the outputs of the `marlin` package.

120 2

Another fab #rspatial resource from Jason: an intro to spatial data manipulation with terra (the successor to raster). Tutorial: jflowernet.github.io/intro-terra/

A four panel plot that shows an original raster in the top left depicting a raster of sea surface temperature in Australia with the boundary of the Great Barrier Reef Marine Park overlaid. The top right panel shows the raster when it is cropped to the boundary. The bottom left panel shows the raster when it is masked to the boundary. The bottom right panel shows the raster when it is both cropped and masked.

130 3

First, an intro to #tmap, developed by spatial data wizard Jason Flower! 🌏️ Since working collaboratively through his materials, I'm pretty convinced there's no better way to get #rspatial data neatly plotted in just a few lines of code. Tutorial: jflowernet.github.io/intro_tmap/

A short snippet of tmap code that says: tm_shape(gbr_sst) + tm_raster() + tm_shape(gbr_boundary) + tm_borders(). The resulting plot shows a raster of sea surface temperature in blue colors off the coast of northeast Australia with the boundary of the Great Barrier Reef marine park overlaid in black.

120 2

Honorable mention: VSCode Pets extension. Do I need to throw a ball to a tiny animated pet built into my IDE? No. Does it make me happy? Yes.

030 3

The quick summaries of data (including any missing data 🔎 ) in the data explorer is pretty handy. Plus, I've noticed that large objects load much faster here than in RStudio!

151 5

RStudio's Git pane had strength in its simplicity, but Positron's "source control" provides a visual commit graph, detailed diffs, and includes many more git commands in the GUI ⚡️ Screenshot from the current @posit.co Positron cheat sheet: rstudio.github.io/cheatsheets/...

141 4

For me, the biggest plus was that if (okay, when) my code crashes, my IDE is #stayinalive. Crashes taking down RStudio too was a big pain point for me!

241 4

5️⃣ Curate a small set of reliable learning sources (that you’ll actually look at) It's impossible to keep on top of *every* new feature in the R ecosystem! But here are some resources that R-Ladies SB members love: ⭐️ rweekly.org ⭐️ @rfortherestofus.com ⭐️ #rstats on Bluesky and LinkedIn

210 2

3️⃣ Embrace the alerts, errors, and warnings You know the story: you revisit some old code, and get hit with a bunch of "this function is deprecated" messages 😩 Fear not! Often, the answer is right in the message. Familiarize yourself with function lifecycle stages and what they mean.

A meme of Robert Downey Jr grimacing with the text "Me: I finally understand how this function works" and "Library developers: That function is deprecated"

100 1

Last fall, we hosted a facilitated discussion for our R-Ladies Santa Barbara community to discuss: How do people *actually* keep up with the ever-evolving R ecosystem? Then, we wrote a blog (and gave it the hashtag #stayrelevant 😉): 🔗 rladies.org/blog/2026/sa... Here are our 6 considerations 🧵

We Are RLadies+ @weare.rladies.org · 27/04/2026

Ever come back to an #rstats project and…it just doesn’t work anymore? Or struggle to keep up with all the latest R packages & tools, much less integrate them into your code? Keeping R workflows up-to-date can feel daunting - but luckily, @rladiessb.bsky.social has some thoughts ⤵️

A weary pink monster looking at a large stack of books with the titles "New package", "Cool updates", "K now change it all!", "Now add Python." An enthusiastic green monster with its arms in the air and a speech bubble that says "and there's something new every day!"

2215 21

Outside of work, I love to spend as much time as I can in the playground that is Santa Barbara, California! ☀️ You can often find me on my bike, on a trail, or by (or in!) the ocean.

A woman exploring a tide pool in Santa Barbara, California during sunset.

We Are RLadies+ @weare.rladies.org · 27/02/2026

Today I debugged code for a simulation study on the association of home environment and child development 🤓📚 Previous versions of this plot included random jitters due to some unknown cause (looking at you #OpenMx 👀) but I fixed it now 📈 How have you used #RStats this week? #Rladies #ggplot2

MacBook screen showing RStudio in dark mode. The bottom right corner shows power as a function of measurement time point. Power to detect an effect the further apart measurement points are.

1122 12

We Are RLadies+ @weare.rladies.org · 26/02/2026

So, a thread about how I have learned #SEM in #R as a #PhD student in #StatisticalGenetics. I’ve heard SEM be described as a “system of regressions”, but in addition to observed variables, we also have latent/unobserved variables. We can represent these variables in a path diagram (lavaan.org) :

Path diagram of a structural equation model with three latent variables called ind60, dem60, and dem65. ind60 has three indicators, dem60 and dem65 each have four. dem65 is regressed on dem60 and ind60. dem60 is regressed on ind60.

1205 20

We Are RLadies+ @weare.rladies.org · 23/02/2026

Hi, I’m this week’s curator, Josi! I‘m a PhD student in statistical genetics 🧬 and use #RStats to do #SEM. I dabble in #DataViz and recently wrote my first #Rpackage. My hobbies are reading #Fantasy books 📚 and #maximalism. You may catch me at a conference matching my outfit to my poster 🤓🧜🏽‍♀️ #rladies

Foto of a feminine nonbinary person in a purple floral shirt in front of a purple poster. The poster is displaying a structural equation model analysing the influence of parental education on children’s school grades. The person is smiling cheekily to the side.

Foto of a child with four fluffy buns and blue sunglasses. The child has lowered the sunglasses below their nose and is gazing over them.

26010 60

We Are RLadies+ @weare.rladies.org · 24/10/2025

Here's the relevant config from the second example

A snippet of code in `_quarto.yml` that sets under `website` the keys: `open-graph`, `title`, `description`, `image` and `site-url`.

030 3

We Are RLadies+ @weare.rladies.org · 24/10/2025

Can you spot the difference? If you've wondered how to get those pretty previews when you share your Quarto websites on social, you need to enable `open-graph`. Docs: quarto.org/docs/website...

Two draft blue sky posts: on the left a link with a preview with just a url and title; on the right a link with a preview that includes a picture, title and description.

1122 12

We Are RLadies+ @weare.rladies.org · 23/10/2025

How about a Positron (or VS Code) snippet that picks a random number for your seed? Example snippet: gist.github.com/cwickham/820... Snippets: positron.posit.co/migrate-rstu...

0143 14

We Are RLadies+ @weare.rladies.org · 23/10/2025

The best I've managed is to use citations as usual, then edit a CSL file to make the style you would normally see in the "References" section, the style you get inline. Code example: github.com/cwickham/ref... Preview: cwickham.github.io/references/

261 6

We Are RLadies+ @weare.rladies.org · 21/10/2025

Yup, as @incidentalfindings.org pointed out, you can use Quarto's usual citation syntax and you'll get the "References" for free. If you just want the references (without having to cite them somewhere) you can use special `nocite` syntax, see: quarto.org/docs/authori...

Code cell for the file `refs.qmd`. In the YAML header the option `nocite` is set with value `@*`.