Terminal

Most people operate a computer with a mouse. You click on a picture of a folder, and the folder opens. A terminal works differently. It is usually a plain window into which you type instructions as text. For example, you write a short word and press Enter, and the computer lists the contents of a folder. The response appears directly below, again as text. This way of operating a computer existed even before the mouse, and it has not disappeared to this day.

Why professionals prefer typing over clicking

At first glance, typing seems cumbersome. The advantage becomes clear as soon as a task grows larger. Suppose you want to insert the date into the filenames of 3,000 photos. With a mouse, that would be days of work. In the terminal, it’s a single line.

There’s a second point as well: typed commands can be saved and shared. A sequence of commands can be stored in a file and run again at any time. Such a file is called a script. Clicks, on the other hand, can neither be saved nor described exactly to someone else. That’s why the terminal has remained standard in software development.

A third reason is distance. Servers, meaning the computers behind websites and apps, are located in far-away data centers. They often don’t even have a screen. You connect to them over the network and control them exclusively through text commands.

What happens between Enter and the response

Strictly speaking, the terminal is just the window. The actual work is done by a program behind it called the shell. The shell reads your line, breaks it apart, and searches for the matching program on the hard drive. Then it starts that program and passes along your additional information.

A command line almost always has the same structure. At the front is the name of the program. After that come the details it should work with, such as a filename. In between there can be options, often preceded by a hyphen. They configure exactly how the program should work.

What’s practical is the passing on of results. The output of one program can become directly the input of the next. This way, small tools can be linked into a chain, similar to an assembly line. That’s exactly what makes the command line so powerful. However, it is also merciless: a delete command doesn’t ask for confirmation, and there is no recycle bin.

Terminals in developer tools and AI products

Every common operating system comes with a terminal built in. On Windows it’s called Command Prompt or PowerShell, on macOS and Linux simply Terminal. Programming environments like Visual Studio Code also have one built in permanently. Anyone who installs, tests, or publishes software almost inevitably works in it.

In recent years, the terminal has additionally become a place for AI tools. Programs like Claude Code or Codex run as text programs within the terminal. You describe in plain language what should be changed, and the model then edits files and runs commands itself. The technical term for this is agent: an AI system that not only responds but carries out steps on its own.

For tech industry news, this is relevant because a lot of money is currently flowing into this area. Companies like Anthropic, OpenAI, and GitHub are competing over exactly these tools. A common misconception, by the way, is that the terminal is a relic of the past. In fact, it is the tool through which a large part of today’s AI applications is operated.