Gentler introduction to recursion

[katrinabrock]

This is an unsolicited PR so feel free to close without merging you don't think this needs fixing. Goal is to actually explain what learners need to know about recursion rather than linking out to wikipedia.

Changes:

• Explain the concept in simple words when first introduced
• Remind learners of the in-context meaning throughout

This is unfortunately a net add of content. I think it is possible to make the explanation(s) more concise, but likely will still be a net add since it adds an explanation that wasn't there before. I think overall it makes the lesson more clear to a learner who is not already familiar with the concept of recursion.

[github-actions]

🆗 Pre-flight checks passed 😃

This pull request has been checked and contains no modified workflow files, spoofing, or invalid commits.

It should be safe to Approve and Run the workflows that need maintainer approval.

[bkmgit]

Maybe "Recursion" should be in the glossary? This might allow simplifying the paragraph.

[katrinabrock]

Agree it should be added. I'll add that to this PR. I still think a short inline explanation would be helpful. The instructor will need to verbally explain -R/-r somehow so I think coming up with some nice concise and approachable wording will be helpful.

[bkmgit]

Suggested change

	: A recursive precudure is one where one step of the procedure is to apply the procedure itself
	(usually to a different input).
	In the context of shell, it often refers to applying a command to a directory and to all directories
	contained within that directory.
	: A recursive procedure is one where the output of an iteration
	is the input of another iteration of that procedure.

Is context of the shell lesson helpful? Maybe should be added to
other terms if it is.

[katrinabrock]

I put this because for all except ls, the command itself is not recursive. Really the only "recursive" part is finding files that we apply the command to. e.g. in grep, we are not applying some kind of recursive algo to search ie feeding the output of the search back into the search algo multiple times (not sure what that would even mean, maybe something akin to a recursive regex 😬 ). We are recursively crawling the filesystem to identify files to search, and then searching them (once each).

For this reason, I think the main thing learners need to know is that -r/-R means "apply this to the entire directory tree within the specified directory". They don't need to fully understand recursion as a general concept. In fact, imo, trying to dig into that would be distracting/confusing. I think it's enough to cover that recursion is a more general concept that has something to do with applying a process in a tree-like manner and in this context when applied to a directory means applying to everything inside as well.

This "apply to everything inside" is probably novices default way of thinking of commands and the fact that it's not the default for shell commands and -r/-R needs to be added is something to actively learn. Meaning, when you copy a folder in a GUI, typically the contents are also copied. When you search a folder in a GUI, typically that means searching all the contents and subfolder contents etc.

[bkmgit]

Probably better to have shell lesson context portion in the main text rather than in the glossary so that if one reads the glossary on its own, it is self contained. A clearer but general explanation of recursion in the glossary would also be welcome.

[bkmgit]

Perhaps if

The `-R` option to the `ls` command will list all nested subdirectories within a directory.

is changed to

The `-R` or `--recursive` option to the `ls` command will make `ls` a [recursive] procedure
which lists all nested sub-directories within a directory.

to ensure recursive links to the glossary add

just before the Keypoints section, see #1510
Then the above can be shortened to

Suggested change

	The `-R` option to the `ls` command will list all nested subdirectories within a directory.
	`R` in `-R` stands for "Recursive" which refers to the idea that the output of a process
	becomes the input of another iteration of that same process.
	So in this case, `ls -R` lists the contents of a directory, and for every directory
	it finds, it lists the contents of that directory,
	and for every directory inside those subdirectories,
	it lists the contents, and so on.
	The command `ls -R` lists the contents of a directory, and for every directory
	it finds, it lists the contents of that directory,
	and for every directory inside those subdirectories,
	it lists the contents, and so on.

[bkmgit]

Suggested change

	the `-r` option for `cp`
	the `-r` or `--recursive` option for `cp`

long form options are great for readability

[omsai]

@katrinabrock I second the suggestion from @bkmgit about describing recursion in the glossary, but I do not think we should be discussing the concept in the main text because it adds unnecessary cognitive load. I would only want the learners to know about cp -r for copying directories with the minimum explanation for why that is necessary. Using find or grep is more typical than ls -R; I don't see as much use of ls -R to focus on it here.

[omsai]

I think ls -R should be removed entirely so that the learner only needs to know about cp -r. We can describe in the instructor notes that ls -R has been removed in favor of tools discussed later like find that is more practical. The learner explanation for recursion should be confined to the glossary.