2019
Eight Obscure Bash Options You Might Want to Know About. (via O’Reilly Programming Newletter)
📖 Marina Tsvetaeva, Vivre dans le feu (Robert Laffont, 2005)
Recently, on BSAG’s website: Rethinking my dotfiles setup. Every time it seems I learn a new thing about the open source world when reading her blog posts.
I saw this pandoc LaTeX template popping out on Twitter yesterday. I remember trying it out at some point when I was looking for some good hand crafted templates for R/statistical reports. At that time, many people were also using the Metropolis theme for building slides. I wrote custom templates for both situations, finally.
IB Foundations of Data Science (PDF, 106 pp.). Yet another textbook for learning DS “quickly”. Beware, it relies on Python and is quite mathy.
Python for Epidemiologists: Tutorial in Python targeted at Epidemiologists. Over
the last few years, I came across several attempts at bringing the Python
statistical ecosystem to the Epi domain. Now trying this one. #python
TIL about Bulma, which is a free, open source CSS framework based on Flexbox. Looks like a solid and lightweight alternative to Bootstrap.
Writing a sqlite clone from scratch in C. I already learned a lot of cool stuff
by reading part of this tutorial. I’m eager to learn more. #database #clang
93% of Paint Splatters are Valid Perl Programs. So funny that you probably have to read the paper as well.
Statistics for Hackers. Be sure to also check the GH repo of the author, e.g.
code_py. #python
Convert images to LaTeX using Mathpix snipping tool. Maybe one day I will learn that I was wrong about the way I wrote some equations. A little test demonstrated that it works right out of the box with simple expressions:
Mine was, in this particular case:
$$ \hat f(x) = \frac{1}{nb}\sum_{j=1}^n K\left(\frac{x-x_j}{b}\right) $$
I somehow forgot about GitUp. A simple git log and more Magit when required, and
I’m generally done with tracking what I’ve done. Of course, this is because I
mostly work alone. Anyway, I like it when you have a such a clean and minimal UI
which helps to visualize at a glance (or even act on) a full Git repo.
Un excellent tutoriel
Emacs en français ! #emacs #fr
Oldie but goodie: normal random number generator in R. (via @BrodieGaslam) #rstats
Another command-line utility to deal with flat files: xsv (available on Homebrew).
Beta-blockers are so much fun. I just stayed out of it for almost four hours after I got home from work. And it’s been going on for weeks. Time for another beer I guess.
This is the 1000th post! 💪👈
Building sites …
| EN
+------------------|------+
Pages | 1000
Paginator pages | 160
Non-page files | 0
Static files | 617
Processed images | 0
Aliases | 34
Sitemaps | 1
Cleaned | 0
Total in 2825 ms
I don’t really like editing code in a browser. This is why I rarely use Jupyter
notebooks. Likewise, I found that the Emacs EIN package was clunky. This was
years ago, and I don’t think the project has evolved so much in the recent
years. TIL there’s a new interface to communicate with Jupyter kernels:
emacs-jupyter. #emacs
Built on top of Bootstrap, the Argon Design System comes with hundred of full featured components and is compatible with mainstream browsers. And the documentation is gorgeous.
You are being watched. Private and state-sponsored organizations are monitoring and recording your online activities. privacytools.io provides knowledge and tools to protect your privacy against global mass surveillance.
🎥 Matrix. Long time no see… wait, it’s still as topical as ever. Remember those who spoke about AI 20 years ago? Or wrote Black Mirror more recently? Anyway, my son and I had a good time watching this “viral” movie.
Remote pbcopy on OS X systems. Nice tip, as always.
I’m trying to use Eshell more consistently since a few days. Mastering Emacs is
(as always) quite useful in this respect. After having tried some custom
settings, including those found on Modern Emacs, I finally choose the full-featured
eshell-git-prompt. #emacs
Grav – a modern flat-file CMS. If you’re looking for an alternative to Jekyll or Hugo, there’s probably some good stuff behind this open-source project. (via Brett Terpstra)
Numerical Tours of Data Sciences, feat. Python, Julia and R.
I’m quite happy actually with how Spacemacs handles LSP for various modes I make
regularly use of (Python, JS, C), thanks to the wonderful lsp-mode. Today, I
discovered that there’s another “universal” package, eglot, for dealing with all
available servers. (via @hillelogram) #emacs
Nice finding today! Just when I thought I would need to write a full macOS
native app for viewing Fasta files or MAFFT-aligned sequences, I found it
already exists, and it’s so much faster and prettier than Jalview. Thank you so
much Mathieu Fourment! #bioinfo
Typing is not the problem. Nice take! This came just after reading Tom’s last post, where I also learned about conventional commits. The latter reminds me that at some point I was using some ideas from Modern Emacs to highlight commit leaders.
A successful Git branching model. See also What is wrong with this. Personally, I found that the Atom team has a pretty nice setup for working with stable and beta version.
relies on
I remember the time when I was using PLINK to perform genome-wide analysis,
before I switched to David Clayton’s excellent snpMatrix R package. Now, it
looks like some folks are interested in using Julia for this stuff. #julia
📖 Grégoire Delacourt, Mon père (JC Lattès, 2019)
Still no idea which TV series to watch, nor what to look for in this evening. So I’ll keep posting (not so) random links that have been hanging around on my iPhone for weeks:
- Time Series Analysis, by Kevin Kotzé
- Feather, CSV, or Rdata, by Vince Buffalo
- pandas 2.0 Design Documents, by the Pandas team
- Intervention Analysis, from the PennState Eberly College of Science
- If not SICP, then what? Maybe HTDP?, by Steven Rosenberg
- Matrix Algebra proGrams In Common Lisp, by Rigetti Computing
- Now You C Me, by Davis Vaughan
- Codex Seraphinianus, by John Borwick
- Reading large CSV files in R, on SO
- Advent of Code in Lisp, by Dimitri Fontaine
- Marijn Haverbeke’s homepage
- Modern Regression Analysis (PDF), by James C. Slaughter
- Variance reduction in randomised trials by inverse probability weighting using the propensity score, by Williamson et al.
- David Gohel’s Github account
- Should we ignore covariate imbalance and stop presenting a stratified ‘table one’ for randomized trials?, on Frank Harrell’s Discourse
- A Python package for exploring and analysing genetic variation data
- Replacing Disqus with Github Comments, by Don Williamson
Never heard of Qwant before, but it looks like a good alternative to Google or Bing.
Les données conduisent au refroidissement social. #fr
Well, I’m done for good with Occupied, my list of TV series is out of stock, and I have no idea what to look for. I guess I’m good at reading books and watching Minecraft gamers on Twitch.
Got a little upgrade under the hood in the morning:
Nothing really new, though, except perhaps the “more editorial highlights on a single page in the Browse tab” in iTunes.