A

abstraction: A way of thinking about a problem that makes it easier to reason about. For example, say we want to write a program that fetches some data from a server. We might write a function that abstracts away the nitty-gritty details of actually making the request, handling errors in case the server is down, etc. Once we have implemented this function (and assuming there are no bugs), we can then use the function wherever we want to in the rest of the program to get data from the server without having to worry each time about the details of how it works under the hood.
address: A location in memory. Most high-level languages abstract the concept of memory addresses away from the programmer.

B

branch: A splitting off from a common source of code (either another branch, or the master branch) that allows you to work on a feature without affecting the progress of other parallel branches. See also: version control.

client: A computer (or, depending on the context, a program) that requests data from a server.
commit: In Git, a logical collection of changes that can be annotated with a message to describe what the changes are and the reasoning behind them. The commit is the main object that Git allows you to manipulate—you can reorder commits, change their contents, move them between branches, and more.

directory: A grouping of files on the filesystem. Synonymous with folder. By placing files and other directories into directories, we form a hierarchy of files on the filesystem that makes it easier to locate and reference files.

filesystem: A way of representing and managing how data on a computer is stored and accessed. Filesystems provide things like files, directories, and links (also called aliases in macOS, shortcuts in Windows) to help users organize and access data on a computer. Typically, we use this term to refer to files stored on the hard drive, but filesystems can, among other things, also be for storage devices that networked (i.e. not in the same computer).
function: A basic unit of computation that does something useful, possibly using one or more input parameters and returning one (or many, depending on the programming language) values. Often times, we group logically related computations together in one function. See also: abstraction.

hard drive: Also called hard disk, hard disk drive, HDD. A mechanical device consisting of magnetic platters (literally a metal disk) that store data and an arm with a magnetic head that reads and writes data to the platters. Reading and writing data to disk is slow compared to reading and writing data to RAM, but hard drives can store much more data for less cost and can keep this data even after the computer is powered off.
high-level programming language: A programming language that abstracts away from the programmer more details about how the computer executes a program. Typically, high-level languages are easier to use and understand, but the tradeoff is usually lower performance (slower programs) or the inability to do more complex things that the language does not allow, whether by design or not. The definition of what programming languages count as “high-level” varies depending on the period of history we’re talking about, or even who you talk to, but most people will not argue that languages like Python, R, and shell script are high-level. See also: low-level programming language.

local: Refers to something on the same computer, as compared to something on a different computer (remote). See also: network.
low-level programming language: A programming language that abstracts away from the programmer fewer details about how the computer executes a program. Examples of low-level languages include C and assembly. See also: high-level programming language.

master: The name commonly given to the main, canonical branch of development. A common action to start working on a feature might be to make a new development (or dev) branch from master. Once you are done with development on that branch, you merge it back into master.
memory: Computers have hardware that allow them to store data for some period of time. Depending on the type of memory, data can be kept in memory for as long as it is receiving power (volatile memory), or it can be kept even after it is powered down (non-volatile memory). In most contexts, the term memory refers to random access memory (RAM). When you run a program, it is first loaded into RAM before it is executed so that the CPU can read it.
merge: The act of combining two branches back into one. Once your work on a feature in a separate branch is complete, you usually merge your feature branch into master.
metadata: Literally, data about data. One example might be metadata for files: the file itself is the actual data, and the time it was last modified, its size, etc. would be the metadata.

network: Many mutually connected computers that can talk to each other. The Internet is a vast network of computers located around the world that uses the Internet protocol (IP) suite to communicate.

pull: In Git, the act of pulling retrieves work from other developers (or even yourself) from a remote to your local machine. This allows you to work on the code locally.
push: In Git, the act of pushing publishes your work to a remote. This makes it visible to other developers and also saves a copy of your work to the remote.

remote: Refers to something on a separate computer, as compared to something local. In Git, a remote repository (simply called a remote) is a repository located on another computer. You can collaborate with other developers by pushing your work to and pulling other people’s work from a common remote. In Git, the main remote is usually named origin by convention.
repository: Also repo for short. A collection of files and associated metadata that help developers keep track of code and the history of changes made to that code. Closely tied to the concept of version control.

server: A computer (or, depending on the context, a program) connected to a network that can respond to requests for data from clients. When we visit websites on the Internet, our web browser (Chrome, Firefox, Safari, etc.) is a client that requests data from a server (e.g. Google servers, Facebook servers), which then respond to our browser with data (HTML, CSS, JavaScript). Our web browser then renders this content into something that it can draw on the computer screen as a graphical representation of the data provided by the server—the actual webpage.
source code: Also source for short. The file containing the actual text of code (hence the name source) that we write. For example, if we are writing Python, we would put our code in .py files, which would be our source files.

version control: A system to manage changes to source code. Especially important when working with multiple developers. Version control allows us to keep track of who made changes to what files when, to append notes to changes so that we can explain why we made those changes, and the ability to roll back changes if necessary. Popular version control systems include Git, Mercurial, and Perforce.