Cool! Not too bad of a friction. Too bad I'm not using Quarto for my blog (using Minimal Mistakes theme via GitHub Pages, I'm a simple man)

This Bluesky integration into the blog's comments is neat!
Is it simple to implement?

Great post!
Now I need to decide if I want to save those extra ms and just use mini.nvim :)
Currently, I'm using a total of 57 plugins 😳

Currently using GitHub Pages to host my personal site+blog (using a simple Minimal Mistakes theme). So my current domain is: menisadi.github.io
I wonder: should I buy a domain of my own, so I won't depend on GitHub Pages, create a sort of "personal brand," or is this just a waste of time and money?

The simplest UI improvement ChatGPT needs imo is simply a ToC for a chat, allowing me to navigate between responses in the chat timeline.

Thanks!

Oh! The truth is that I myself have not really tried Positron yet. I'll give it a chance, thank you

Sounds good. I think we'll really save graphs for later. Let's start with data analysis

Interesting idea. But in order to really understand code, from my experience, you first need to have some experience writing code yourself

The intention is that later she will also learn how to perform statistical analyses with the help of scripts

​Looking for advice from the #rstats community!
​I'm was asked by my spouse to teach her R for her research analysis. She has zero programming experience and currently uses Excel.
​Main question: Do I start her with tidyverse or base R?

​Any tips or resources are welcome!

How would you answer the following: Let's say my spouse and I have 6 boys. Can I say something statistically about our inherent (unknown) probability of having a boy or a girl?
ChatGPT got this partially wrong at least on my first attempt

webR - R in the browser Interactive R environment running in your browser. Execute R code, create plots, and analyze data without installing anything.

PSA for anyone beginning to use #wasm #webR to share R code snippets: We have a new short URL for the webR app: webr.sh!

For those who follow Terry Tao's writing when it comes to LLMs, and who know that he is a fairly enthusiastic user of them, constantly looking for ways to use them in his work, this post is even more enlightening.

My thoughts on the crucial importance of methodology on self-reported AI performance on mathematics competitions, and my policy on commenting on such reports going forward: mathstodon.xyz/@tao/1148814...

The word "echos", for example, appears almost every time I ask for a folk-rock song (or even a name for a playlist of vb songs) I have no idea why. It became sort of a red flag in my eyes for AI music.
@switchedonpop.bsky.social

I played around a bit with song generators and "songwriting" with ChatGPT and noticed that there are lyrics that repeat a lot.

Not sure how I missed that!
Going to listen right away. Thanks

@switchedonpop.bsky.social we need a "Learning to love Ed Sheeran" episode

GitHub MCP suffers from the lethal trifecta for prompt injection: access to private data, exposure to malicious instructions + the ability to exfiltrate information

Be really careful with this stuff: attackers can trick your "agent" into stealing your private data simonwillison.net/2025/May/26/...

@nrennie.bsky.social @jbengler.de @tangandhara.bsky.social @eshekhova.bsky.social @weare.rladies.org

Learning R after several years of working with Python.
Should I begin with Base R, or jump straight into the tidyverse?
From skimming through blogs and tutorials, it seems like tidyverse has become the standard in the R community and that Base R is rarely used directly. Is that impression accurate?

Thanks!

I want to update my RSS feed with interesting blogs. I would love to hear some recommendations. Areas of Interest – Data Science. Machine learning (applied and research). Privacy (applied and research). Statistics. Data visualizations.

I'm open for recommendations in general.

💡 Tip: every statement looks more impressive when you add "Assuming the Riemann Hypothesis" in front of it.

"Assuming the Riemann Hypothesis, the algorithm runs in time O(n² log³ n)."

"Assuming the Riemann Hypothesis, the coffee machine on level 2 needs a descaling."

Which plugin are you using for this rendering?

אל דאגה, אפשר לקרוא בלי לפתוח חשבון :)

I'm not the one to answer, but I'll just point you to Terry Tao's posts about his experiments in such workflow:
mathstodon.xyz/@tao/1106010...

mathstodon.xyz/@tao/1099482...

mathstodon.xyz/@tao/1131325...