📢 Calling all #rstats #pydata #julialang #datascience folks! This year, a Data Science Learning Community (DSLC.io) goal is to crowdsource the curation of #TidyTuesday datasets to allow the DLSC board members and project maintainer @jonthegeek.com, more time to better the community in other ways 1/

{tidytuesdayR} v1.2.0 is now available on CRAN! Check out our new "Curating a Dataset" vignette (dslc-io.github.io/tidytuesdayR...) + associated functions, and help us prepare data for #TidyTuesday in #RStats! Only tested in RStudio so far, but let us know if you try something else!

When data managers and statisticians share data horror stories.

Wait, hold on, I think I blinked

The ls command was actually an acronym for "Lost Socks," an early experimental program designed to help Dennis Ritchie find missing laundry items by listing the contents of his drawers. It was later repurposed for file systems 🙃 #unix #linux

Right now if I were an international student I’d be seeking any way to transfer to a much less oppressive country because of the insecurity of trying to finish my degree here.

Shoutout to PIs who give a crap… I don’t buy the whole “your company is your family” thing but treating your team like they are valuable and involving them in the good times and the bad can go a long way #academia #acadamicsky

EVERYONE needs to watch @3blue1brown.com’s video on how LLMs convert massive amounts of text data into vector space and then individual vectors into human interpretable “meaning”, it completely explains how ChatGPT and friends can be likened to google search but smarter: youtube.com/shorts/FJtFZ...

#ChatGPT can take em dashes from my cold — dead — hands

We talking like, I should dump my 401k, soon, or like, I can cancel any Christmas plans, soon? Coz I’d like to cancel my plans if possible, thanks

👻 StringsAsFactors = TRUE 👻 #rstats

👻

In the words of Frank Reynolds, “I’m talking about community hunny,” and the #TidyTuesday community is indeed multilingual

ISTG if this person posts about #tidyplots one more time imma drop everything im doing and…probably play with it myself it looks really cool

Hot take: equivalence testing should be like, lesson number 2 in intro to statistics, right after differences of means

Sli.dev and #Quarto when?

requisite note that this all started from NIH work on Gila monsters

“An important part of courage is stupidity.” Thanks @hankgreen.bsky.social for reminding me just how fearless I am every single day

Flagging @ted-satterthwaite.bsky.social you might have some of this imaging data 🤔

My problem with the future of #AI and #LLMs is that I am constantly using em for “tip of my tongue” tasks. The kinds of problems that I know are totally solvable because I’ve seen them done in the wild — and I fully understand them — but I’m not motivated to go and learn *how to do them myself*

:(

Me waking up with some random song that I’ve not listened to in years stuck in my head.

I’ve now lived in Boston for 3 years and am embarrassed that I keep asking this question this time of year:

{plumber} got a facelift?! I didn’t even think it needed any work done, but this should be interesting… #rstats #APIs

Me: if my #rstats installation of rJava breaks one more time I’m going to throw my laptop out the window! Them: …you know you could just open that 200 line excel spreadsheet… in excel…right?

Haven't blogged in a little while, here's a new one: "R Version 4.5.0 is Out!" www.njtierney.com/post/2025/04... - `install.packages()` is faster - new datasets! `penguins`, `gait` - BLAS/LAPACK updates - My thoughts on reading NEWS files and updating R packages #rstats #rse #rpkgs #cran

Sometimes I laugh when job applications have “must be proficient in MS Office Suite”… And then sometimes I see screenshot images of excel spreadsheets, pasted into word documents across multiple pages…

Eric Jensen & I have a new paper out, based on a survey of research software funders, in a Sloan-funded research software policy project E. A. Jensen, D. S. Katz, "Awareness of FAIR and FAIR4RS among international research software funders," Scientific Data, v.12, 627, 2025. doi.org/10.1038/s415...

The functional work of an #RSE is growing

"But the most highly cited papers generally aren’t the most famous scientific discoveries. Rather, these works tend to describe scientific methods or software, the workhorses on which scientists depend." From doi.org/10.1038/d415... #ResearchSoftware

Teacher: what would you like to do when you grow up? Guy who directs OkGo videos: I wanna play with mirrors and robots and zero gravity and rollerblades and treadmills and umbrellas Teacher: 🙄okay sweetie, sure you can do whatever you want🙄 Guy who directors OkGo videos:

I haven’t read any briefs or anything, but I do believe that sending laypeople (ie not scientists) to space is a worthwhile experiment. I’m also not surprised that the only layperson who can afford it is a celebrity. Do I like it? No. But do I get it? Yeah, I guess

I just had my employment review at #Harvard last week…

are you a fan of the amazing {targets}📦 William Landau? are you a fan of my {rix} 📦? Then you might want to check out {rixpress}, a new 📦 still in dev that makes it easy to define reproducible analytical pipelines with steps in either #Rstats, #Python or #Quarto: github.com/b-rodrigues/...

It’s no #Severence, but if you’re itching for something to fill the void while we wait for the next season, #Silo has been pretty good at filing that vibe so far

The thing I hate most about #ChatGPT right now is that whenever I come to it with a new task that I’ve given some decent thought to, and come with some of the materials prepared, it sounds _genuinely_ thankful and for my pre-work and applauds me… it’s creepy how validated it makes me feel

No shade intended, but #python users getting in on #TidyTuesday can just call it that… #pydytuesday just feels… I don’t know… off

Okay I give up.. snakemake + hydra in #python is LEGITIMATELY bad.

I’m only just now realising that #databs is shorthand for Data BlueSky, and not, in fact, data BS

I’m becoming a bigger and bigger fan of Notebook Driven Development for both #python with {NBDev} and #rstats with {fusen}… seriously considering writing a full blown blog series on it thinkr-open.github.io/fusen/ nbdev.fast.ai

df <- read_csv(“importantdata.csv”) df_ <- df %>% important_function()

“The most important thing universities and academia should be doing during this #AI revolution is to figure out and work on the ethics of using these tools.” #LLMs #ArtificialIntelligence #AcademicSky #DataScience - Adam Selipsky

Our latest US-RSE newsletter is here! us-rse.org/2025-03-28-n... - A huge shoutout to @tinashemtapera.bsky.social for making it happen. Now back after a short break, we should be back on a monthly cadence! #RSE #RSEng

My pledge not to get Rick rolled this April fool’s lasted about an hour… so how’s your day going?

Python is the best language for statistical analysis and data science hands down. I'm currently porting all my reproducible pipelines (R in Nix shells) over to Jupyter notebooks (pip is all you need) running inside docker containers long live Pandas! this now a #pydata account

I would love to see more research teams employ data managers. While roles like data scientists and research analysts are crucial for analyzing data, those roles often don't have the time to develop standardized and documented processes that lead to better data. #databs cghlewis.com/blog/hire_da...