If you read German, I highly recommend @tobischolz.bsky.social recent piece on Raisina Dialogue and what it means in the broader context:
11.03.2026 14:56
👍 5
🔁 2
💬 1
📌 0
If you read German, I highly recommend @tobischolz.bsky.social recent piece on Raisina Dialogue and what it means in the broader context:
Even granted that the Home Office makes ministers at least 50% more evil, there’s something unnerving about Mahmood’s performative cruelty
My article in @thediplomat.com discusses what the United States' recent decapitation of #Iran’s leadership following the earlier abduction of #Venezuela’s leader means for Asia. A summary of the key points in the comments: @chasiapacific.bsky.social thediplomat.com/2026/03/amer...
Well damn, this is cool.
Birds That Don't Exist: Niche Pre-Emption as a Constraint on Morphological Evolution in the Passeroidea | Chia et al., 2026 | Ecology Letters
onlinelibrary.wiley.com/doi/full/10....
"would otherwise receive immediate access to welfare and social housing" is BS.
Correct: "after 5 years working in a care home, would be eligible to apply for welfare and social housing on the same basis as the rest of us."
www.theguardian.com/commentisfre...
Deliberately misleading/disingenous.
A quick plug for juliapackages.com/p/mixedmodels (by Doug Bates and phillipalday.com -- similar capabilities to lme4 but **much** faster
I'd love to see 2.75 M geese. Who doesn't love geese, don't even have to be Snow geese.
I rounded up a few Claude Skills for #RStats users.
Huge thanks to the creators who developed them. They share Skills for everything from tidyverse code to brand.yml files to learning while using AI.
Hope the list is useful, and please let me know what I missed! 🧡
rworks.dev/posts/claude...
🆕 BLOG POST
There’s so much more to spatial downscaling today than classic LM or kriging. In my new blog post, I show how flexible and powerful GAMs can be for this task. #rstats #dataviz
➡️ dominicroye.github.io/blog/canary-...
"This is not strategic autonomy. It is a risk multiplier. The question is not whether India can afford to alienate either the US or Israel; it is whether India can afford a West Asia engulfed in war.
If this truly is “not an era of war,” then Modi must act as if he believes his own words."
What a spineless worm, was it really that hard to stay out of it?
Thats it's mostly unnecessary. Our RSE team has developed a soft no-CRAN, R-universe first strategy after bad experiences with CRAN. My own (from a previous job) have been fine, but have heard it's often not great.
Otoh, I recently submitted to the Julia registry in error and had a good experience!
At least with the Iraq war, they prepared a dossier
Only on Bluesky was I going to find out about this in this framing
Published in @jappliedecology.bsky.social!😀
We show how (Hierarchical) Hidden Markov Models ((H)HMMs) can be tailored to different epidemiological scenarios to infer disease status directly from animal movement data.
🔗 ttps://besjournals.onlinelibrary.wiley.com/doi/10.1111/1365-2664.70323
"The number of birds captured in mist nets plummeted by 40% between 2001 and 2014. Likewise, bird numbers fell by 50% during systematic surveys of birds by sight and sound. The most severe declines occurred among species that feed on insects."
Well, thanks for the optimism. First off I don't think the RSE role is mainly teaching. I don't see how RSEs survive for any but the wealthiest labs when agent code is good enough for median use cases. In the aggregate it probably depends on whether funders are willing to cover RSEs.
This is how the RSE role goes extinct I fear
v1 of `bootstrapper`, an opinionated #rstats package to help you start building modern R packages. Wraps a number of usethis funcs to simplify starting new packages.
Adds some common GHA, dependabot, air + jarl (precommit hook and GHA), and a concise AGENTS.md which I'm rather liking.
Reminds me of my back pocket idea to design a course in which each session we discuss the history and political economy of a single element. Has anyone taught something similar?
To help myself get through these times, I capture evidence of better days to think back to. This was Tuesday evening, although it was nice most of the day, following which the grey, damp, drizzle, and rain has returned with a vengence.
I can normally get behind the Lib Dems but not this nanny state nonsense. I have a Lib Dem MP - pushing on OSA-related stuff any more, and supporting the UK's further descent into a surveillance state will lose her my vote.
supercells helps group raster cells into meaningful spatial units for analysis and ML.
v2 adds many new features.
Docs: jakubnowosad.com/supercells/a...
Please try it and share feedback, suggestions, or use cases -- your input will help shape the release 🙏
#rstats #RSpatial #GISchat
Excited to share our new paper in @natcomms.nature.com We synthesize causal discovery & inference approaches across traditions (regression adjustment, quasi-expts, SEMs, Granger causality, convergent cross-mapping, and more) into a unified workflow for ecologists. www.nature.com/articles/s41...
A drawing titled "psst genetic engineers: I have a proposal." It shows ten duck species being combined into one best duck, which has a long-tailed duck's tail plumes, a king eider's wing spikes, a hooded merganser's hood, a white-winged scoter's eye markings, a redhead's eye and red head, a ring-necked duck's bill pattern, a northern shoveler's bill shape, a northern pintail's neck pattern, a harlequin duck's blue, and a ruddy duck's upturned tail.
An oldie about ducks and ignoring scientific ethics.
Just having a look at this map; moused over my area and the tooltip showed a pop size of 10M, same for most of the SE. Is it possibly pulling from a larger regional area? In which case the county/borough wise stats are going to be off, probably.
Yet again we in Higher Ed get everything bad from being private institutions, and everything bad from being seen as part of the public sector. We can be sued and sued again for actions which we didn't cause, which hit *us*. And yet ppl think we are backed by govt. www.bbc.co.uk/news/article...
New post on how terminal AI tools (e.g. Claude Code) make my work easier.
As the discourse is caught between hobbyists and programmers, mass unemployment and slop, I thought some normie use cases saving me several hours this past month would be helpful.
newsletter.mikekonczal.com/p/three-ways...
#tinycc based #RStats #C #Transpiler, only 4x slower than #{quickr} for the convolution benchmark ! 2-3x faster than naive #C + marshaling. #tinycc transpilation will have an obvious advantage since it can always fallback when transpiling and use `Rf_lang3` since it has access to R's C runtime !
