I added more information to the installing software section, updated section title to specify Arch Linux, and added another section for Debian and Derivatives.
My tiny nitpick is that "touch" will create the file you specify if it doesn't exist. I've seen this usage a lot, so your example may benefit from mentioning it.
I find the references to file extension kinda confusing. Extensions mean a lot less in Linux cli, but I can tell youre just using them for examples. Maybe give more concrete examples instead.
I don't know how this would be useful to someone reading the cheat sheet, but here's something interesting I just indirectly found out while skimming it through:
Ctrl+D does the same thing as ENTER, except the latter additionally sends the end-of-line character to the reader while the former sends nothing;
as is the case for shells or interactive programs like the Python REPL, Ctrl+D causes them to terminate only because it sends a string that is 0 characters long, and 0-size reads are universally interpreted as files reaching the end.
To test this: enter cat, type "hello" without pressing enter, then Ctrl+D: you should see "hellohello".
An extremely rare case of this being useful would be using netcat to send a string somewhere, without sending the end-of-line byte at the end.
I updated "Log out" to "Exit (sends a signal indicating the end of a text stream)". Which I think is a lot more accurate, and still easy to understand.
Oh. My. Gosh. I love this. Thank you. And thank you for being --verbose about the provenance and history of the document. And big big thank yous for the Internet Archive links. Bravo.
Yes, the top most directory, /, is the root directory.
Each directory is a branch in one giant tree structure. For example, if you have a directory containing two other directories, that is a branch that is splitting into two branches.
All directories are descendants of the same root.
Done? I've never uploaded to GitHub before, and I was just doing what I thought I should do. I'll do my best to keep it updated with the version on my website.
Is there a way to make GitHub automatically detect changes to the file at cerium.cc and update the repo? Or do they not allow that? I know a scheduled script would be able to work but I don't really want to have to run it myself.
Alt + . inserts the last argument from the last command run into the current line. I find it helpful all the time.
less can be invoked directly, without having to be piped from cat: less <file> is mostly equivalent to cat <file> | less
I have considered making an alias/function that automatically determines if the file is longer than the terminal, using something like wc -l and stty -a | grep -oP "rows \d" and then either uses cat or less depending on that... but I already use sharkdp's bat, which has that baked in as well as many other conveniences
Don't forget tail -f <file> which is kind of like watch tail <file>
If you're going to have du, I would also have a section for df, I use the latter more often (but probably because I have like 5 mounts for my OS). Using them in combination is basically what all the gui disk usage analyzers do; something like df -h "oh, /var's almost full" (as previously mentioned, I have different folders on different partitions), then du -ah /var and so on to find problem areas
The "installing from source" section works maybe 50% of the time. It assumes a configure script, which isn't always the case. I've had a lot of source that comes bundled the way a .deb does: basically a compressed filesystem that assumes the $CWD is / (basically, if you uncompressed it in /, all the files would go where they needed to be). Sometimes they use language-specific build systems, so you might need go and rust and... Maybe it's best to just keep it your way and look up the rest, but do keep in mind the thing I said about compressed filesystems
find is great if you want to reindex everything from square 0; or if you only need to do small directory/tree. If you have the extra space to spare, install locate: it indexes the files beforehand (as a cron job) and yields results more quickly for searches that span entire filesystems; the only downside is that you have to manually reindex (sudo updatedb) to locate files installed the same day
In the Extracting, Sorting, and Filtering Data section, you might consider adding in sort -u and uniq which fill their own (overlapping) niches. sed and awk may be a bit more than beginner, but they are endlessly helpful. tr can be a useful shorthand for when cut and sed don't quite cut it, but you don't want to build a full in-line awk script.
Finally:
<command> 2>&1 <file> | Output and errors from <command> are redirected and appended to <file>
Should read "Output and errors from <command> are redirected to <file>" because the single > overwrites the existing file, as opposed to >> which, as you noted, appends to the end of the file
At the end, in redirection, <<: that's not how here-documents work. The example gives the impression it will read the given file up until "STOP", but in reality the shell expects you to keep writing your here-doc until you write "STOP" and then feeds it to the program as if it were all on stdin. I don't think wc even does anything with the stdin if you give it a filename...
Note that variable expansion will happen in here-docs, so it's a bit different than a simple cat.
Also look into here-strings. And process substitution, I find that quite handy.
A little over a decade & a half and I find that very useful. Should have GPG in for reading signatures on software and such too. If you'd like I can contribute to GPG terminal as I've been using it for a good portion of that time.
It is not as complicated as it looks! As a long time windows user I'm in the process of getting used to the command line and I love the simplicity and direct way of doing things. For some tasks and small programs it is amazing. Grep for example is something amazing that you couldn't do in a windows type UI.
For other stuff like visualizing a directory tree you can always use a mouse interface. You're rarely forced to use the terminal.
Makes as much sense as saying "I'll definitely avoid Windows now. I had heard that Windows was supposed to be easy to use", since most of this also applies to WSL2.