It's one of these weeks

Currently working with a new corpus, and today we decided to run NER on it. However, the pipeline (spacy) would grind to a halt much more frequently than I am used to. Could be the pipeline (which is Swedish now), or something else. But I have a more interesting theory:

+++ NEW ARTICLE +++ It's Friday, and you know what this means: A new part of my series on WebGL! Today, I'm walking you through how to programmatically calculate color gradients using a huge amount of circle math: www.hendrik-erz.de/post/webgl-s...

Fun times are ahead.

It’s crazy how many usually pretty non-political YouTubers currently see the need to warn against impeding fascism in the U.S. After Technology Connections, now @adamneely.bsky.social has dropped a 90min epos with a 30min chapter on techno-fascism.

+++ NEW ARTICLE +++ After some theoretical planning and creating some triangles, today we move to actually animating them! Read on for part 4 of my series on WebGL: www.hendrik-erz.de/post/webgl-s...

New video by @techconnectify.bsky.social. I enjoy watching his videos because I like his nerdy exegeses of heat pumps and solar. BUT. Today I want you ALL (!) to watch a particular part of this video that starts at 1:16:35. The link below jumps to that part. Watch it. youtu.be/KtQ9nt2ZeGM?...

We’re now in the replacement bus for the replacement bus for the train. I take bets as to whether I’ll make it home this weekend.

Current State: we stood for about three hours, until we could get off the train into a replacement bus. And you won’t believe what happened next: the bus lost its engine and can’t continue either.

Currently sitting in a train where someone has just jumped in front of. Never happened to me yet. I’ve never been so sympathetic of a train delay. Condolences, incl. to the operator who likely had to witness everything first row …

What does it take to produce free and safe software for Windows? Quite a lot, it turns out. And the biggest enemy of small developers is Microsoft themselves. Read more in our third piece on "Selling Trust." zettlr.com/post/selling...

+++ NEW ARTICLE +++ After two theoretical exercises to understand and set up OpenGL and WebGL, this week it's finally time to draw something! If you always wanted to generate your own sun in a 3d rendering engine, read on! www.hendrik-erz.de/post/webgl-s...

Accidental finding: This great paper on norms that people hold, versus what norms they think other people hold: academic.oup.com/pnasnexus/ar...

What I said almost a year ago. Although I might’ve been too careful in outlining the dangers of this: www.hendrik-erz.de/post/this-is...

Welcome to the worst of all timelines.

+++ NEW ARTICLE +++ This week is part 2 of my WebGL series. We're still not drawing anything, but talk instead about state management. I promise it's getting more hands-on next week! www.hendrik-erz.de/post/webgl-s...

+++ NEW ARTICLE +++ Over Christmas, I embarked on yet another journey into some weird niche topic. It turned out to be extremely exhausting, so I wrote down what I did. Brace for an eight-part article series on WebGL! www.hendrik-erz.de/post/a-rabbi...

What the f***.

Happy holidays everyone! ;)

Academic Christmas: Last-minute paper submissions. It's 11:34pm on December 23, and I just hit "Submit" on a paper submission. Finally, no more work, just in time for Christmas! Happy Holidays! ;)

Formatting an R plot for a journal submission that requires points and millimeters is harder than it should be.

Automate the boring stuff 2.0: If someone provides me a list of keywords and tells me to pick from that list, I'm gonna ask my local Llama. (Pro-Tip: If you have a somewhat decent computer and a model with sufficient context, you can throw entire papers at that thing, and it'll work flawlessly!)

Me: Uses proper tools such as password managers to prevent typos and other human errors The journal submission system:

Great opportunity for B.A. and M.A. students around the world to submit CSS research!

„Ostdeutsch“ und „Westdeutsch“ als Zuschreibungen ist ein Tagungsbericht von Judith Nasdal, zu lesen in unserer aktuellen Ausgabe. Diesen und viele weitere Beiträge von Studierenden und Nachwuchswissenschaftler*innen gibt es bei uns zu lesen soziologieblog.hypotheses.org/soziologiema...

📢 Analytical sociology is coming home! Call 4 INAS26 is open 🌍 1-3Jul26 @nuffieldcollege.bsky.social and @sociologyoxford.bsky.social (☔) ☢️ Deadline: 1Feb26 🦹 Organisers: @aksoyundan.bsky.social , @kasimirdederichs.bsky.social, D Kretschmer, @awaldendorf.bsky.social Info: tinyurl.com/yc2tusjx

PhD opportunity at the University of Oxford. The Morelli scholarship funds a doctoral student to work with me on on democratic backsliding, strategies of democratic defense and regeneration, or the rise of illiberalism. Deadline Jan 9. More information at users.ox.ac.uk/~ssfc0073/

Recently I saw some guy on YouTube explain that it takes only few videos until the algorithm starts filling it either with Shrimp Jesus and/or right-leaning content. Let YT on during sleep tonight, and it took 4 videos of my regular channels before some 10h Christian meditation music video started.

And that's why every social science dept should have a house philosopher who can slap us on our wrists

Being on the job market makes me realize: The traveling salesman was me all along!

This administration appears to have been so successful that they now have time for small things. Except they don't, because there is much more to come in the solidification of the unitary executive theory.

We have just released the possibly final beta release before we can move Zettlr 4.0 to stable! If you'd like to have a sneak peek to what we have in stock, take a look here 👇🏼 github.com/Zettlr/Zettl...

Since search is dead, how soon do you think Google Scholar is headed for the Google Graveyard? I'm betting it's soon, and academia is NOT prepared

This looks yummy 👀

I'm always shocked how quickly eugenics is losing its “dirty” connotations these days, especially in scientific subdisciplines that broadly work with AI. And yes, this also includes my own related scientific fields.

Situation: Doing peer-review for a journal. Footnote in the proof starts with the words: “As we do in an earlier study (NAME and NAME, YEAR)…” Hach. The unavoidable pitfalls of peer review…

And thus it continues…! Looking forward to 2026!

+++ NEW ARTICLE +++ Recently, I had my first encounter with a “vibe coded” PR. Vibe coding is performative production for the sake of production, detached from community and social norms. I argue why vibe coding is harmful from a social perspective. www.hendrik-erz.de/post/vibe-co...

I'm increasingly convinced the European model of scientific funding is a waste of time and talent and needs to change. Just one of many stories... 1/4

Today, I’m excited to be a guest at the Technical University of Braunschweig, where I will talk about my dissertation!

So @mims.bsky.social is doing some of the best work in the mainstream media on AI and data centers, and is one of the few who is asking the difficult questions rather than looking for convenient ways to back up market consensus, it’s nice to see

Shutdown, capitulation, and austerity: how Democrats failed to defend affordable healthcare - Peoples Dispatch #united states#access to health care#donald trump#affordable healthcare act

Excited! Looking forward to it! 🤩

Signs of an 8-hour workshop

John Mohr has once called interpreting topic model results a form of "Rorschach Test", and I agree. I just had a topic with completely odd word stems and had no idea what to do with them. And then it hit me. They were all (slightly butchered) names of Neural Network scholars.

I'm back in the phase of sitting idly in front of my computer while it works *sips tea* ☕️

Dear potential contributors to ANY software. Please do not just open PRs entirely generated by an AI thingy. Clean them. I just had a pretty good feature PR opened to my main repository, adding support for RTL languages (Arabic, Hebrew, etc). BUT, it’s impossible to review on a volunteer basis:

This is amazing

How doing Computational Social Science with R sometimes feels

Can AI-generated images work as visual vignettes in survey experiments? I argue yes - and discuss how . in my new @sscratsage.bsky.social paper: doi.org/10.1177/0894...