We could also discontinue the traditional way of publishing (there is little evidence that peer review improves the quality of the research). That would also redirect money from for-profit publishing companies to other things.

This slide unfortunately generalizes well 🥲

And if this article is right, this is the perfect example of the underspecification of the theory. If people double-downed on the cult, it's because they reduced cognitive dissonance. Now it may turn out that people disbanded the cult. Well, it's because they reduced cognitive dissonance 😃

One example is that he never refers to 3I/ATLAS as an “alien spaceship” but as an “interstellar object” when strictly referring to it (I’m talking about the concept of referring expressions, not about when speculations about the object are made). People should learn a bit of linguistics perhaps.

I agree with Timo! Of course applicability varies by discipline, but in linguistics it is often very difficult if not impossible to come up with a SMES (Smallest Meaningful Effect Size, as I call it in Bayesian inference, although the concept is similar to SESOI).

I know it is difficult to do this here, but could you expand on what you take "science" to be? (I ask because my positionality makes it so I don't much care for "science", but I want to make sure we are talking about the same thing). :) (Love your work by the way! And the students like it too!)

But why doing equivalence testing when one can learn from a full posterior probability distribution? (Maybe I am biased because I don't think dichotomous decision have a place in knowledge-oriented research)

It’s not been an issue so far. Students who engage, engage, those who don’t, don’t (these are very very few tho. The course ha about 90/100 students). I see most of these back again during dissertation writing and they are on top of things.

That said, assessment is a university policy that cannot be changed so this is what I do: uoelel.github.io/qml/assessme...

(Funny enough, while courses have to be assessed, there isn't anything in the UK law that says that they have to be marked and what the marks should be)

I don't think we can ever "effectively assess" knowledge, so I gave up.

"Until a system of absolute mental units has been invented, it is futile to try to make grades represent absolute accomplishment." doi.org/10.1126/scie...

In Jurassic World Rebirth, in contrast you learn a lot about the characters story from inference based on minimal dialogue, facial expressions, stuff like that. A lot of what is out today is just overly redundant (probably meant for busy people that multitask while watching stuff?)

For me it’s not so much that there is much going on but that everything is over-explained… like in episode one when they give a lecture to the lad about the five companies and so on… stuff like that. It’s a trend now, and I don’t enjoy it 🤣 in contrast, Jurassic World Rebirth

Til the end of time (1946): not so much cerebral but not so obvious like contemporary movies.

Mother! (2017): biblical inspiration with pagan layer.

The Game (1997): cerebral, brainful, gray mattery, thriller and action.

Also commented on this Discussion thread: github.com/posit-dev/po... Although the issue clearly shows a lot of interest in the feature already