picture of two firefighters trying to extinguish a very large dumpster fire with a small hose. text on dumpster: INFERENCE IN THE BIOLOGICAL SCIENCES. text on fire fighters: ME AND MY FRIENDS
Giving a short presentation later today to my institute's external advisory board. This slide summarizes the last 3 years of my department's efforts
🌟Nominations are now open for the 2026 SORTEE Commendation Awards!
Know a project promoting #OpenScience in Ecology & Evolution (workshops, tools, outreach & more)? Projects can be local or global, team-based or individual
Submit a nomination here: sortee.org/awards/
I took two years of ancient Greek in college, so when this stuff appears in my feed, I need to post about it as therapy
Bro was buried with the most boring part of the Iliad, like 6 pages of naming captains and the towns. There's an interactive map of the passage, for shared obsession: ships.lib.virginia.edu/neatline/sho...
A black-and-white photo of an Oblique Strategy card. The card reads: 'Use an unacceptable color'.
Use an unacceptable color
#ObliqueStrategies
yay, since i am getting back into the weeds of 3rd edition, nice to have this motivation. i promise to do even better next time.
Finally finished! Never thought I would enjoy reading a textbook front-to-back, yet I did, immensely. The best thing about the book is getting you to think of data generating processes. The flexibility of Bayes in creating scientifically valid equations is incredibly motivating. #rstats #statsky
yes very common in canals here
Leipzig wildlife. These are called „beaver rats“ and they are canonically buddhist
Yeah transparency and explicit alternatives and why they were rejected is what i am asking for
public funded institutions should have to show their work at least. to be fair max planck has done its share of vibe policy too. there is a broad cultural problem with how academic institutions make decisions
„The only remaining way to mitigate this situation is to reduce the number of applications.“
They’ve tried nothing and are all out of ideas? Whether this is good or bad, it is unjustified. There is no impact analysis, no engagement with research on funding procedures. Government by vibes
Reposting this old thread on the "Pervasive findings of directional selection" from ancient DNA. While the authors extended their results in various ways, I think many of these points still stand.
beside the point, but that coeff of 0.00 with a SE of 0.00 having a p-value of * gave me an instant migraine
This classic experiment was underpowered. It should be replicated with more cats www.youtube.com/watch?v=O9Xt...
Using the cytoscape library (cytoscape.org), which does almost all the graph work. Still much javascript suffering.
workflow graph showing the upstream parents of the Effect_Area node
javascript intensifies. highlighting upstream dependencies now, nominating missing parents and tests, it's all coming together
this is continental philosophy
screen cap from "The Wire". Caption: MY NAME IS MY NAME
Pinchy!
Excellent new special Issue on the challenges raised by the ancient DNA of kinship
Special Issue: Kinship Trouble: Traversing Interdisciplinary Boundaries between Archaeology, Archaeogenetics and Socio-cultural Anthropology
www.cambridge.org/core/journal...
I now have a primitive interactive browser interface to the workflow graph. I can get this to support an interactive design mode I think.
I just want to say, and let me be very clear so that no one misunderstands: javascript f'ing sucks.
Hm maybe. I was focused on the uniqueness of eml, but maybe that's not the innovation.
I think this has been known since at least 1935? If I understand right, nothing unique about choice of operator in new quoted paper. (www.pnas.org/doi/10.1073/...)
Post Doc at Uni Tübingen! 100% position for 3 (+3) years; they're looking for somebody to analyze large-scale longitudinal datasets in education research.
Expertise in machine learning is an advantage, commitment to research transparency desirable 😌 proficiency in German beneficial but not required
Nice clear presentation of commonplace meta-analysis failure mode: Pooling coefficients that mean different things, because original models had different adjustment sets. A coefficient gets its meaning from the whole model, not just from the predictor variable it multiplies.
Very excited to announce that the #BayesianWorkflow book by @statmodeling.bsky.social, @avehtari.bsky.social, @rmcelreath.bsky.social et al publishes in June! routledge.com/9780367490140 #RStats #DataScience #Bayesian
yellow curry in pan
cat sitting on towel with eyes closed
Cooking with Mischka: Yellow curry with chicken. Garlic onion chili tomato graram masala curry powder turmeric coconut milk ginger, blend with hand mixer, add seared chicken. Mischka enjoyed chicken bits.
Workflow graph showing suggested tests appropriate to each typed node.
Suggestion system for tests working now. I think I can write an overlay for language-specific examples and workflow calculi for doing things like listing licensing assumptions. But that's v2 perhaps.
This is from a conversation I had recently with a speaker of Dutch and German who wanted to know why English is so weird. As usual, the answer is "because French".
