Ron Garcia

Yup, highly recommended!

In the book Labyrinth of Thought, Ferreirós discusses how Dedekind and Cantor kept having falling-outs because Cantor kept stealing and publishing Dedekind's results without giving due credit.

On Brian Cantwell Smith and the Promise of AI Today I had the bittersweet pleasure of participating in a symposium honoring the late philosopher Brian Cantwell Smith, a good friend whom I’d known for over 30 years.

I wrote about the late philosopher Brian Cantwell Smith, and his profound thinking about AI and the nature of intelligence.

aiguide.substack.com/p/on-brian-c...

That's a *little* better than the Nazca boobies, where the older sibling straight up kicks the younger one out of the nest and the parents ignore it, so it dies 😢.

Psychological Inquiry Volume 34, Issue 4 of Psychological Inquiry

TFW when a meme sends you careening down a rabbit hole (albeit not about the meme because I too am old):

www.tandfonline.com/toc/hpli20/3...

Clarification questions: does this hold even if the regression has no categorical variables? And if so, is that because that circumstance can be construed as having an implicit single-level factor?

From Searle, Casella, and McCulloch: "In endeavoring to decide whether a set of effects is fixed or random, the context of the data, the manner in which they were gathered and the environment from which they came are the determining factors. In considering these points the important question is that of inference: are the levels of the factor going to be considered a random sample from a population of values? “Yes”-then the effects are to be considered as random effects. “No”- then, presumably, inferences will be made just about the levels occurring in the data and the effects are considered as fixed effects. Thus when inferences will be made about a population of effects from which those in the data are considered to be a random sample, the effects are considered as random; and when inferences are going to be confined to the effects in the model, the effects are considered fixed."

Slide from a Richard McElreath lecture on varying effects about superstitions. (best considered after reading Gelman's blog post)

Trying to hold these two in my mind at the same time 🥲

In classifying data in terms of factors and their levels the feature of interest is the extent to which different levels of a factor affect the variable of interest. We refer to this as the eflect of a level of a factor on that variable. The effects of a factor are always one or other of the two kinds, as has already been indicated. First are f i x e d eflects, which are the effects attributable to a finite set of levels of a factor that occur in the data and which are there because we are interested in them. In Table 1.1 the effects for the factor sex are fixed effects, as are those for the factors drug and marital status. Further quality discussion of fixed effects is in Kempthorne (1975). In a different context the effect on crop yield of three levels of a factor called fertilizer could correspond to the three different fertilizer regimes used in an agricultural experiment. They would be three regimes of particular interest, the effects of which we would want to quantify from the data to be collected from the experiment. The second kind of effects are random eflects. These are attributable to a (usually) infinite set of levels of a factor, of which only a random sample are deemed to occur in the data. For example, four loaves of bread are taken from each of six batches of bread baked at three different temperatures. Whereas the effects due to temperature would be considered fixed effects (presumably we are interested in the particular temperatures used), the effects due to batches would be considered random effects because the batches chosen would be considered a random sample of batches from some hypothetical, infinite population of batches. Since there is definite interest in the particular baking temperatures used, the statistical concern is to estimate those temperature effects; they are fixed effects. No assumption is made that the temperatures are selected at random from a distribution of temperature values. Since, in contrast, this kind of assumption has t…

Gelman's cryptic definition #2 inspired me to look up Searle, Casella, and McCulloch, which to me at least provides some useful terminological context:

that this is almost literally one of the first things we teach in my intro to CS course (entitled "Systematic Program Design") makes me feel pretty good rn 😎

I see you're doing penance for your timeline cleanse 😱

Sounds like I am bound to like Greenland's interpretation of Feyerabend better than the batch strength version. Thanks!

lol that paints a picture!

Any chance you could explain this joke (and thereby ruin it, I know sorry :( )? I've neither read Marx nor Feyerabend, so only know of them via caricature.

OTOH I enjoyed that Greenland not only read Feyerabend, but took his class!!

For and Against Methodologies: Some Perspectives on Recent Causal and Statistical Inference Debates - European Journal of Epidemiology I present an overview of two methods controversies that are central to analysis and inference: That surrounding causal modeling as reflected in the “causal inference” movement, and that surrounding nu...

Sander Greenland has an interesting take on manipulability in this banger of an article: (e.g. the section entitled "Feasibility and precision: Not necessary, but desirable")

link.springer.com/article/10.1...

Description Please note that job descriptions are not exhaustive, and you may be asked to take on additional duties that align with the key responsibilities ment...

Computing @ Imperial are hiring four Ass. / Assoc. Profs! Priority areas:

- PL
- Systems
- Security
- Software Eng.
- Computer Architecture
- Theoretical Computer Science

Applications from individuals from underrepresented groups especially welcome!

www.imperial.ac.uk/jobs/search-...

No Shame, R's kinda neat!

BTW what language are you implementing this in?

Just in case this might help:
www.cs.tufts.edu/~nr/cs257/ar...

Dave, sometimes you have to speak to the children in small words they think they understand 😉

course schedule as a table. Available at the link in the post.

I'm teaching Statistical Rethinking again starting Jan 2026. This time with live lectures, divided into Beginner and Experienced sections. Will be a lot more work for me, but I hope much better for students.

I will record lectures & all will be found at this link: github.com/rmcelreath/s...

5 dimensions being high-dimensional, with intuitions from 1 and 2 dimensional spaces utterly failing, is a pretty good rule of thumb.

You might need the "input" to determine which disjunct holds!

Godel figured out the translation to S4;
then Kripke came up with the possible-worlds model for S4;
then Kripke smashed the two together:
www.princeton.edu/~hhalvors/re...

Causal foundations of bias, disparity and fairness The study of biases, such as gender or racial biases, is an important topic in the social and behavioural sciences. However, the literature does not always clearly define the concept. Definitions of b...

Curious if you've seen this manuscript from some years ago, and if so your thoughts:

Traag and Waltman, Causal foundations of bias, disparity and fairness

arxiv.org/abs/2207.13665

For an interesting approach (presumably happening close by you), you might want to check this out:
jpolitz.github.io/notes/2024/0...

“You know, I’m not unaccustomed to people being confidently wrong at me. I’m a woman with a PhD. I’ve been training for this my whole life.”
— @cfiesler.bsky.social 🧪

New on my blog:
50 years of proof assistants
lawrencecpaulson.github.io/2025/12/05/H...

Gonna be hard for me to take the guardian seriously now.

Just out: Functional Data Structures, edited by Tobias Nipkow

ICLR Points: How Many ICLR Publications Is One Paper in Each Area? Scientific publications significantly impact academic-related decisions in computer science, where top-tier conferences are particularly influential. However, efforts required to produce a publication...

One facet I think about a lot is the massive differences in publication rate across areas of CS when evaluating CVs (arxiv.org/abs/2503.16623 tries to quantify it). How this affects/should affect the early part of open-area searches is really something.