Micro posting in December
2025-12-05:
Quai de la Garonne, Toulouse, Dec. 2025
2025-12-05: /me is listening to “If I wanted someone” by Dawes
2025-12-05:
There seems to be some resurgence into the old web now, time will tell if it gains any real ground. It’s an uphill battle: Besides most online eyeballs now being glued to social media apps, we’re seeing AI take over the way people interact with the internet altogether. Back in the old days if you wanted to know more about something you’d google the term and start going through the websites Google said are most relevant. — Blogging in 2025: Screaming into the void
2025-12-05: Modeling temporal data with an unknown number of changepoints. #python
2025-12-05: System Reliability and Six Sigma in R and Python.
2025-12-07:
I’ve seen a number of editors grow popular and then wane. At various times I’ve seen Emacs, Eclipse, IntelliJ IDEA, Atom, SlickEdit, gedit, Notepad++, Sublime, NetBeans, Visual Studio, Xcode, IDLE, PyCharm, Android Studio, and more. Some of those aren’t around any more. I’m not worried about Vim getting eaten by a big tech company, or enshittified to feed advertising or AI ambitions. — Choosing Vim over VSCode
So, while I agree that it’s wrong to generalise that ‘Perl 6 killed Perl’, I would say that Perl 6 was a symptom of the irreconcilable internal forces that killed Perl. Although, I also intend to go on to point out that Perl isn’t dead, nothing has actually killed Perl. Killed Perl is a very stupid way to frame the discussion, but here we are. Perl’s decline was cultural
2025-12-07: Screenshots from developers: 2002 vs. 2015.
2025-12-08: osxphotos looks like a blessing to manage iPhoto library. The same applies to imessage-exporter. #apple
2025-12-09: QuickSort in 61 characters:
quicksort = # & @@@ {# //. x : {__} :> (## & @@ Reverse /@ GatherBy[x, (# < x[[1]] &)])} &
2025-12-09: [Rainbow array algebra](https://math.tali.link/rainbow-array-algebra/): I really like syntax-highlighted math notation and all those pretty diagrams. See also [MathTools](https://github.com/taliesinb/MathTools?tab=readme-ov-file). `#mathematica`
2025-12-09: [wolframite](https://github.com/scicloj/wolframite?tab=readme-ov-file): An interface between Clojure and Wolfram Language (the language of Mathematica). It works like a charm with latest release of Wolfram engine on macOS. `#clojure` `#mathematica`
2025-12-11: [Common Lisp, ASDF, and Quicklisp: packaging explained](https://cdegroot.com/programming/commonlisp/2025/11/26/cl-ql-asdf.html): It is important to realize that a “system” and a package are entirely different things: one is an entity in an add-on tool, the other is intrinsic to Common Lisp’s namespacing. `#lisp`
2025-12-14:
Cathédrale Notre-Dame de l'Assomption, Rodez, Dec. 2025
2025-12-14: /me is listening to "Zombie" by YUNGBLUD
2025-12-14: [I Fed 24 Years of My Blog Posts to a Markov Model](https://susam.net/fed-24-years-of-posts-to-markov-model.html).
2025-12-16: /me is listening to "Gallowdance" by Lebanon Hanover
2025-12-16:
The speakers are pretty good. And for a laptop, they’re actually amazing. They have depth, the upmp, and are loud enough to watch films with comfort - can’t say this about any other laptop I’ve had before. — GNOME is better macOS than macOS
Why readable math still matters in a world aided by LLM-assisted code generation. — In Defense of MATLAB: Why Engineers Still Need Whiteboard-Style CodeI too am a firm believer in simple syntax (R, Mathematica, or even Scilab or Octave): “The math is identical. But in the Python version, the engineer is thinking about computer science concepts."
2025-12-16: Interesting discussion on matrix multiplication with Mathematica: Speedup matrix number multiplication. #mathematica
2025-12-18: New kid on the blocks: ty: An extremely fast Python type checker and LSP. I’m pretty happy with basedpyright at present, but I will check this new tool. #python
2025-12-18: Introduction to Software Development Tooling: Lot of good stuff and very clear handouts.
2025-12-23:
I agree that it can be difficult to make sense of a model containing billions of parameters. Certainly a human can’t understand such a model by inspecting the values of each parameter individually. But one can gain insight by examing the properties of the model—where it succeeds and fails, how well it learns as a function of data, etc. — On Chomsky and the Two Cultures of Statistical Learning
Note that this updates cells live in response to previous cells that they depend on. Not pictured is that it doesn’t update cells if their dependencies haven’t changed. You can think of this as a spreadsheet-like Jupyter, where code is only rerun when necessary. — the terminal of the futureStill I don’t really like web-based notebooks.
2025-12-23: Flocking Quadtrees.
2025-12-23:
2025-12-27:
Jardin Tino Rossi, Paris, Dec. 2025
2025-12-27: /me is listening to “Albatross” by Alela Diane
2025-12-27: Recent Python bookmarks:
- How uv got so fast: uv is fast because of what it doesn’t do, not because of what language it’s written in.
- profiling.sampling: statistical profiler via periodic stack sampling introduced in Python 3.15.
- Exploring Mathematics with Python: mathematics with Python as a replacement to Turbo Pascal.
2025-12-29: Mole: Deep clean and optimize your Mac.#apple
