Thomas Hühn

Thomas Hühn

@2uo.de

CompSci, EdSci. Functional safety. Open Source Compliance. OT cybersecurity.

7 followers31 following50 posts

We are bad at estimating a single value. But at least we know deep in our hearts that we're bad at it. But who can claim to choose their estimates such that on many repetitions a given confidence level is reached? That is why we need to calibrate our estimates. www.thomas-huehn.com/calibrated-e...

thomas-huehn.comCalibrated EstimatesWe are often required to supply estimates. Project managers would like us to say how long a task will take. That number should be on firm grounding. Sometim...

Totally not unrelated: the term "thought leader" is already stupid. But for the love of God, don't call yourself thought leader in your web site bio!

Is it because one of the original researchers who coined the term saw a way to commercialize the concept and himself became the original consultant for this niche, paving the way and ensuring some quality?

Compare with the snide remarks in the competence literature where researchers make quite clear that they think what HR departments do is bullshit (only phrased more politely). Yes, they may envy the consultants salaries, but still.

It seems to me that the concept "Community of Practice" has survived the transfer from academia to company implementation much better than other concepts. While we probably all know instances where CoP means "monthly Powerpoint presentation", by and large they seem to fare well.

Thomas Hühn reposted

Delighted to learn that @gretchenmcc.bsky.social and @superlinguo.bsky.social did an entire podcast episode structured around explaining two words—“morphosyntactic alignment”—in one of my comics (xkcd.com/2421)

Lingthusiasm@lingthusiasm.com

"I visit them" and "they visit me" mean different things But with "I laugh" and "laugh me" either would work just as well to convey the meaning In this episode, we get delighted by languages that pick this second option, a pattern known as ergativity!

Thomas Hühn reposted

I described these estimates last week in my intro lecture for the human evolution core course at my institute. Surprises a lot of ppl. And odd for a primate to be ~half mammal biomass, given how insignificant other primates are. More bat biomass than non-human primate. Humans are weed monkeys.

Matthew Freeman@mjafreeman.bsky.social

OK this is the mind-blowing image of the week for me: the total global biomass of mammals. Wild animals are a tiny proportion. From a paper by Ron Milo Lab at Weizmann Institute, via a great Oxford lecture by @marionkoopmans.bsky.social www.pnas.org/doi/10.1073/...

Encyclopedias that really exist and that I need to flick through someday: * New Grove Dictionary of Music and Musicians * New Dictionary of the History of Ideas * Encyclopedia of Food and Culture * Encyclopedia of Historical Treaties and Alliances

* Encyclopedia of Journalism * Encyclopedia of Law Enforcement * Food Cultures of the World * International Encyclopedia of Dance * Literature of Travel and Exploration: An Encyclopedia * Routledge Encyclopedia of Language Teaching and Learning

Merlin Mann's Wisdom project has some nice nuggets. "Organizing your email is like alphabetizing your recycling." Guilty as charged.

I even tried to convert it with Pandas, but the default sqlite driver in Python cannot cope with so many columns. Or something. I didn't spend much time, only wanted to see if I can get a quick Datasette frontend to the data,

JASP looks approachable, but is strange in ways I cannot put into words, and SPSS (I've got a license via the university) is a sprawling monster. I guess as a programmer learning R will be quicker.

I am currently evaluating a possible Bachelor's thesis, and oh my God. PISA and TIMMS datasets are complex as hell, and statistics is simply frightening. Hopefully I can get something plausible to compute with @rmcelreath.bsky.social's book soon. So I'm learning R on the side.

You are up to date.