I have a four-hour long or so jazz for writing playlist in Spotify. No lyrics, which is less distracting for me.
Category: flotsam (Page 1 of 4)
Flotsam 16: new laptop
In my job I get a new laptop every 3 years or so; at least that is how it works with Apple laptops. You get a new one, together with Apple care, and it is depreciated during three years. Keeping computers for longer doesn’t make financial sense according to the bean counters. Coincidentally, it is roughly the time for the laptops to start falling apart, more likely by design.
On terms of features, I reached 1 TB SSD disk around 6 years ago (I don’t use half of that), 16 GB of RAM 3 years ago (I used to be quite comfortable with 8 GB of RAM 9 years ago or so. What I am trying to say is that spec-wise I’ve been OK for the last half decade, at least. The peak of my computing was a Macbook Air 13″ just before the appalling Macbook Pro 13″ butterfly keyboard fiasco. In 2020 I ordered a huge 16″ Macbook Pro, despite 13″ being my sweetspot for laptop size, because of covid 19. We didn’t know for how long we’d be working at home—which in NZ turned out to be not very long—so I ordered a larger screen and, gasp, a real ESC key (again). I don’t have much love for the 16″: too heavy, too noisy, meh battery life, got too hot, etc.
This time I went back to Macbook Pro 14″ because: real ESC key (ridiculous to mention this, but I was traumatised by the touch bar ESC), no touch bar (yay!), SD card slot (I like photography), HDMI connector (FINALLY!) so I can skip on one dongle, proper power connector. The screen notch looks funny, but it disappears from my mind when busy writing. Overall impression: solid, hefty, fast. It actually feels much faster than the 16″ with Intel processor.
I test a lot of software that I don’t end up using, R packages, etc. so I avoid moving my old setup to the new laptop, starting from scratch and avoid carrying over all the cruft accummulated over three years. Then it comes the unavoidable boring task of installing the software I need for my work (the university already install MS Office and other software that don’t use, like Endnote). I installed:
Homebrew: unix package manager*.RandRStudio: R stuff (see below for packages)*.- Apple command line tools: compiler, etc.
MacTex: everything and the kitchen sink LaTeX for mac*.Zotero(includingZotfileandBetter Bibtexplugins)Joplin: notetaking in markdown*NetNewsWire: reading RSS feeds, free, synchronises across mac and ipad*.Calibre: e-book management*.Digikam: photo management*.Rawtherapee: RAW photo processing*.- Visual Studio Code: free text editor, don’t think it is fully open source.
- Neovim: text editor*.
- pandoc: text transformer*.
- asciidoctor: text transformer*.
All starred (*) items are Open Source Software.
I use numerous R packages, but when I start with a new computer I don’t compile a list of packages to import in the new machine (lots of cruft) but I add a few packages that I know I use often and then add when I need to. Included in this list:
tidyverse: so I get ggplot, dplyr, reader, etc*.data.table: sometimes I use this for fread() and data management*.asremlr: multivariate + spatial genetic analyses.rjags: bayesian stuff*.rstan: bayesian stuff*.
I still have to install “a few” things (like QGIS) but I’m getting there. I’ll update the post later once I have added more software.
Flotsam 15: inference
Before I lose the link—as I’m deleting toots & tweets two weeks after I post the—I should save the address for “Introduction to Modern Causal Inference” by Alejandro Schuler and Mark van der Laan. It is a book draft that looks quite readable.
Read more: Flotsam 15: inferenceAlso love Xanthe Tynehorne, Esq.’s Compendium of Curious Words. Weird enough to make it interesting.
Count me fascinated by the Literature Clock by Johs Enevoldsen, which presents a text from a novel, poem, etc with the time of your computer clock.
I have kept on adding links until 5th February:
This Bayesian Data Analysis course, by Aki Vehtari, based on the classic BDA3 book (link to the free online version) looks really interesting. Even more so if you already have done some Bayesian stats work/study before.
Against Copyediting: Is It Time to Abolish the Department of Corrections? by Helen Rubinstein got me thinking about how we “correct” while editing texts, in my case mostly writings by postgrad students.
Flotsam 14: dev links
Back doing some coding, playing with R packages and collaborating with someone else. This is an unusual situation for me, as I write my own code, and never bothered learning to use git (or other code management system). With the help of the Twitter crowd I’ve been slowly learning to use some git so the first link goes to How to Rebase a Pull Request.
Another interesting link is Getting started with unit testing in R, which will come handy with some other programming work implementing research models.
Flotsam 13: early July links
Man flu kept me at home today, so I decided to do something ‘useful’ and go for a linkathon:
- Ed Yong discusses the effect of subject expectations in psychology experiments Nice Results, But What Did You Expect? At the beginning there was another article on The placebo phenomenon, and another one on The placebo defect.
- A googleVis tutorial to create Hans Rosling-type graphs from R.
- Google’s Python Class is material for an intensive 2-day course on Python.
- An opinion piece on Calculus and statistics by Daniel Kaplan, on teaching a different version of your typical introductory calculus course, so it is useful for statistics. He goes as far as teaching calculus using R. There is more information in Project MOSAIC.
- Nice graphs on what happened to Asiana Airlines flight 214. I didn’t know there was so much available data for a specific flight.
- Biased and Inefficient, Thomas Lumley’s personal statistics blog (he insists that posting 75% of Statschat is not enough to qualify as personal). You may know Thomas from the survey package (or a few others).
- If you are a postgrad student in New Zealand you can apply for a NeSI (New Zealand eScience Infrastructure) postgraduate allocation to access high performance computing facilities.
- My previous post the USA versus Western Europe comparison of GM corn was the first time that I received more traffic from Facebook than from R-bloggers. Five hundred readers in total.
Over and out.