A little review of my current Safari session since it is probably time to close all those tabs bookmarked on my iPhone.

I have long been aspiring to some kind of digital minimalism, and I think I am almost there, with my using only essential applications–to write, code, and read–and almost very reliable ones–Vim or Emacs, a Terminal, and macOS builtin and optimized apps. I would have a harder time to live without my bookshelf or my 25-year old collection of records. However, it seems that we can, provided you live like a real bohemian. (I came across this by reading one of Irreal blog post recently.)

I don’t use Chrome, and I don’t think it is a good idea to use any Google products by now. Facebook is just a little toy compared to the power that Google has accumulated over years. I am using a Mac and I subscribed to Apple Music, though, and you could retort that it’s all the same. I don’t think so, since it’s not the same offer of services, unless you think of Chromebook (or Android) users. Anyway, here is a little joy for Lisp saavy users: Dinosaur and Lisp. Note, however, that you will need to have a working Chrome(ium).

I didn’t knwo there was such a thing, but here it is: Tutorial on Good Lisp Programming Style, by Peter Norvig (PDF, 116 pp.). The Little Book of Python Anti-Patterns also provides good tips regarding Python this time.

Some CS ressources here and there: Stanford CS Education Library (via HN), The Ultimate Guide to Learning Clojure for Free; and a particular HT to Greg Wilson for his Tidynomicon (I don’t like the tidyverse but this could people in a hurry, plus there are some interesting tips on interacting with Python or developing web app through Shiny) and, of course, the JavaScript for Data Science website.

Some fresh articles for the interested scientific reader:

• Please Stop Permuting Features: An Explanation and Alternatives: An interesting take on why permutation-based statistics may lead to erroneous assessment of model accuracy, especially in the case of strongly correlated predictors;
• Advances in epigenetics link genetics to the environment and disease: A great review of epigenetics and its role/implication in biology and health-related research;
• SciPy 1.0—Fundamental Algorithms for Scientific Computing in Python: Surely a great read for people interested in scientific computing, at least for the historical side and the resulting popularity of Python in data science and computational statistics;
• Lagged Explanatory Variables and the Estimation of Causal Effect (PDF, 15 pp.): In which the uthors used DAGs and Monte Carlo simulations to demonstrate that lag identification often leads to incorrect inferences;
• Generalised linear models for prognosis and intervention: Theory, practice, and implications for machine learning: All you need to known about GLM vs. ML for prediction and causal inference, as a complement to Frank Harrell’s excellent past discussion of this hot topic.

Do you like quines? How about making compressed file quines, step by step? This blog post provides a nice description of how GZIP files are encoded and how they can be manipulated using Python.

Vim vs Emacs: Detailed Comparison, or the Holy war again. Interesting post, albeit a little too generalist for my taste (thx to Irreal!). On a related side, I happened to read a nice post on the history of Vim recently.

Program Design by Calculation is a (draft) textbook all about functional programming, functors and monads, and Haskell:

Functional programming has a tradition of absorbing fresh results from theoretical computer science, algebra and category theory. Languages such as Haskell have been competing to integrate the most re- cent developments and therefore are excellent prototyping vehicles in courses on program calculation, as happens with this book.

You may also like the related blog post: A Short Skinny on Relations & the Algebra of Programming.

Well, that’s all for this month!