Dominikus
LightNews LightNews
banner
do.minik.us
Dominikus
@do.minik.us
2.1K followers 240 following 250 posts
generative artist and prof, data visualization developer, identity-challenged

dataviz: https://do.minik.us
art: https://dominikus.art
Posts Replies Media Videos
do.minik.us
AI in higher ed
November 16, 2025 at 1:29 PM
do.minik.us
Love this map of the landscape of generative AI by @tallerestampa.bsky.social.
Thoroughly researched and all-encompassing, yet gorgeously designed.

cartography-of-generative-ai.net
Map of generative AI
October 14, 2025 at 12:07 PM
do.minik.us
Since I keep rediscovering it:

David Rumsey's Historical Map Collection is an absolutely fantastic resource, also for historical #datavis

www.davidrumsey.com
October 12, 2025 at 9:21 AM
do.minik.us
I completely missed that Tyler Vigen's "Spurious Correlations" project now also show AI-generated explanations and images as well as full, completely fake research papers. What could possibly go wrong?

tylervigen.com/spurious/cor...
screenshot of Tyler Vigen's Spurious Correlations website that shows a chart correlating the number of Johnny Depp movies with season wins for the LA Chargers, followed by an AI explanation and an AI generated image "depicting" the correlation.
September 24, 2025 at 1:13 PM
do.minik.us
love the unabashedly 1998-aesthetic of this website: skepticalscience.com
screenshot of skepticalscience.com
September 24, 2025 at 12:27 PM
do.minik.us
love this "Types of Deadlines" poster from the Timelines, deadlines and lifelines workshop www.temporalimagination.org/resources/de...
poster titled "Types of Deadlines" with 12 different types of deadlines illustrated as little particles bouncing against different types of "deadline" walls
July 1, 2025 at 8:20 AM
do.minik.us
I didn't realize that Jonathan Harris released an updated version of "We Feel Fine" for Apple Silicon Macs in 2024!

Hooray for keeping internet/ #dataviz history alive!

jjh.org/we-feel-fine
screenshot of jjh.org website screenshot of the We Feel Fine (2024) app running
June 23, 2025 at 2:52 PM
do.minik.us
Researcher-gone-investor writes 165-page article on how nothing can stop the growth of AI systems while tucking away the elephant in the room in a little sidebar.
situational-awareness.ai/from-gpt-4-t...
Quote from the website: "All of this is to say that data constraints seem to inject large error bars either way into forecasting the coming years of AI progress. There's a very real chance things stall out..."
June 9, 2025 at 8:38 AM
do.minik.us
Working on processes that bring p5.js input to Stable Diffusion. #creativecoding #genai
May 19, 2025 at 7:23 PM
do.minik.us
Maybe I've also spent too much time staring at ComfyUI workflows
Screenshot of a ComfyUI visual programming workflow
May 19, 2025 at 11:58 AM
do.minik.us
I'm obsessed with this:
Mystical, a visual representation of PostScript as magical symbols. What if software departments were catacombs of hooded figures scratching arcane circles into the dirt
github.com/denismm/myst...

(via @piterpasma.nl)
screenshot from the Mystical project. a visual representation of a PostScript file as several magical circles.
May 19, 2025 at 11:57 AM
do.minik.us
I must be doing something right since I'm on both the "boo, this guy uses AI!" as well as "boo, this guy doesn't use AI!" block lists.
screenshot of the clearsky tool, showing two block lists. first named 'AI "Artists"' and the second one 'does not get bitcoin/nft/ai'
May 18, 2025 at 2:48 PM
do.minik.us
love the convergence of brand names and AI gibberish text
Photo of a product carton for a manual coffee grinder by a brand called OrkeyDolk
May 18, 2025 at 2:22 PM
do.minik.us
Plato's Gorgias dialogue (380BC!) is interesting when it comes to LLMs. In it, Gorgias argues that someone whose only skill is being good at talking is more convincing to a lay audience than actual experts.

Which opens up two avenues for "improving" LLMs:
expert knowledge on everything
or rhetoric.
Screenshot showing parts of Platon's Gorgias dialogue: "Soc: Therefore in relation to all the other skills, the position of the rhetorician or rhetoric is the same. It does not need to know anything about the actual details, since it has discovered some means of persuasion by which, before those who do not know, it appears to know more than those who know. Gorg: Well, doesn’t it make things much easier, Socrates, when without having learned the other skills but just this one skill, you are no worse off than the other artificers? "
May 16, 2025 at 10:10 AM
do.minik.us
Not sure if the library's any good, but that website is 🤌

animejs.com
Screenshot of animejs.com which announces it as an 'All-in-one animation engine' and shows an abstract shape on a black background.
April 7, 2025 at 8:02 AM
do.minik.us
The future of #datavis
AI-generated parody of Minard‘s famous visualization of Napoleon‘s march to Moscow, embellished with all kinds of Studio Ghibli-inspired characters
March 28, 2025 at 11:56 AM
do.minik.us
Me using Cursor
Screencap from the Simpsons-episode 'King-Size Homer' showing a close-up of a plastic bird toy pressing the 'Y'-key on a keyboard.
March 21, 2025 at 4:44 PM
do.minik.us
love (especially) the navigation and playful exploration of the medium web here: thehtml.review/04/
March 21, 2025 at 9:30 AM
do.minik.us
Am I doing it right? 😉
February 21, 2025 at 11:22 AM
do.minik.us
February 12, 2025 at 6:38 PM
do.minik.us
PIXEL_EARRING

#pixelsorting #glitch
February 6, 2025 at 3:08 PM
do.minik.us
an awesome wave 🌊

#pixelsorting #glitch
February 4, 2025 at 8:22 PM
do.minik.us
The pixel sorting will continue until morale improves:

Mary Gartside's "Blue", "Orange" and "Crimson" (1808) versus the algorithm.

#postgenuary #glitch
February 2, 2025 at 7:47 PM
do.minik.us
#genuary31 Pixel sorting.

The painting that (sometimes) shows up is "Marchesa Elena Grimaldi Cattaneo" by Sir Anthony van Dyck

#genuary2025 #genuary #glitch
January 31, 2025 at 8:54 PM
do.minik.us
#genuary30 Abstract Map

#genuary2025 #genuary
January 30, 2025 at 7:20 PM