Linux Crash Course
The command line, package management and Git — the working baseline for computational practicals.
This is a reference, not a lecture. It is written to be scrolled, searched and copied from while you are actually sitting at a terminal, and to be worked through before a practical rather than during one.
Start here
| Kickoff slides | A fast tour of all four parts, in about fifteen minutes |
| Part 1 — The command line | Terminal, paths, commands, pipes, wildcards |
| Part 2 — Installing and managing tools | pixi, and the conda/mamba world you will meet anyway |
| Part 3 — Git and GitHub | Keeping your work, and getting it back |
| Part 4 — Files you will meet | Markdown, JSON, tables, and why text files misbehave |
The slides are a map, not the material — click through them first if you want the shape of the whole thing before reading Part 1 properly.
Part 1 is the part that gates a practical. If you can open a terminal, say where you are in the filesystem, read a command with its parameters, and send output from one command into another, you have enough to start.
The rest matters as you meet it — Part 2 when you install your own tools, Part 3 when you have work worth keeping, Part 4 whenever a file misbehaves.
Who this is for
Anyone about to do computational work who has not used a command line before. It assumes no prior experience — not with Linux, not with a terminal, not with programming.
It deliberately belongs to no single course. Several modules link to it, so the examples here are ordinary files and folders rather than data from any one subject. You will meet the real data in the practical; this is about the tools you will meet it with.
Why a terminal at all
A fair question, and worth answering before you spend an evening on it.
- The tools only exist there. Most scientific software has no graphical interface. This is not an oversight — a program that runs from the command line can be run a thousand times by another program, and one that needs a human to click cannot.
- It is a record. A command you typed can be pasted into a methods section, re-run next year, and checked by someone else. A sequence of clicks cannot be.
- It scales without changing. The command that processes one file processes four hundred with the same typing.
The cost is that nothing is discoverable. A graphical program shows you what it can do; a terminal shows you a blinking cursor and waits. That is the gap this material closes.
How to use it
Type the commands. Do not copy-paste them, at least not at first — the typing is doing more work than it looks like, because the errors you make while typing are the ones you will make later under time pressure, and meeting them now with nothing at stake is the cheapest possible time to meet them.
Every command shown here is executed automatically whenever this site is built, in a clean scratch directory, on Linux. If an example is on this page, it ran. That does not mean it will do the same thing on your machine — versions and defaults differ — but it does mean it is not a typo.