Opening for a tenure track position in statistical computing:
wirtschaftsuniversitaet-wien-portal.rexx-systems.com/Assistant-Pr...
As dept chair, R Core member Kurt Hornik invites applications from people who could make substantial contributions to #RStats, including base R.
Deadline Mar 18!
But I think it's also fair to say that if you know what an MCP server is, then there's less value in Posit assistant for you.
If you want to connect R via MCP with open source you can try github.com/posit-dev/mc... or github.com/t-kalinowski.... But I see three advantages to posit assistant vs claude code + mcp:
1. single shared R session for R and agent
2. plots!
3. easier to see analysis code as its run
We released DuckDB v1.5!
This release comes with a “friendly CLI” client, a new (opt-in) PEG parser, support for VARIANT types and many lakehouse features. It also ships a new network stack, a reworked geospatial extension, Azure writes and an ODBC scanner.
Read more at duckdb.org/2026/03/09/a...
The person sending the email doesn’t know you’re doing this
If you regularly create plots in R, the tidyplots package is definitely worth exploring.
Today, I released a new module in the Statistics Globe Hub that explains how to use tidyplots in R.
More info about the Statistics Globe Hub: statisticsglobe.com/hub
#rstats #datascience #tidyplots #ggplot2
Totally unrelated to this thread, but I have sometimes asked Claude to reword particularly irksome emails into a neutral and supportive tone. And then I respond to that email not the original.
My spawn just sent me this. It appears you can get @posit.co support at halftime during Davidson bball games 🤣
#rstats
Do you think you could do a weekly version? I think the day of week variation makes it harder to see the trend.
Can you please tell the results of Sys.info()[["user"]]? Asking for a friend.
It really is essential!
Grant Witness is hiring! We're seeking a full-time data scientist to join our team building data resources to support journalism, litigation, and activism protecting science, public health, and the rule of law. grant-witness.us/apply.html #rstats
But personally what has really bought it all back for me is having a new arch nemesis: @dingdingpeng.the100.ci 🤣🤣🤣
Obviously we’re a business and we have to cover the costs somehow. Bundling things together is always a bit arbitrary but if we can’t build a sustainable business we can’t fund all the engineers writing open source/free software.
If we did BYO key, I suspect we would charge some nominal “corkage” in order to cover our costs.
If you’re paying $20/month for all you can eat Claude, no you can’t fill up your purse and eat at Posit instead.
Claude code is as good at writing R code as me, in the small. (Still isn’t as good in the large)
Can’t bring your own key for _yet_.
So you’re assuming it’s been rendered once
So always?
How does it access the rendered plot?
I am glad that Bluesky has now captured all the best parts of #rstats twitter: dank memes, sick burns, hot takes, and flame wars.
I've got to say no no no #RStats
RStudio now has next edit suggestions! I wrote a bit about how they work and have open-sourced the eval we used to engineer the system's prompt: www.simonpcouch.com/blog/2026-03...
Here’s a full draft of the upcoming second edition of my “Data Visualization: A Practical Introduction”: socviz.co
But not tibbles!
We are working super hard on it! So hopefully soon
My hope is we can do more cheaper plans once we have more data about how much it costs to run. We want it to be sustainable in the long term which means being financially cautious to start with.
This makes me think of a fun captcha game I played once called I’m not a robot -
neal.fun/not-a-robot/
Screenshot of a plot legend showing two colored lines: a red line and a blue line. The red line is labeled "blue" and the blue line is labeled "red"
Turns out if you put the color argument inside the aesthetic instead of outside by mistake, you can get the most confusing ggplot legend ever #rstats
