The function of a terminal is confined to display and input of data; a device with significant local programmable data processing capability may be called a "smart terminal" or thin client. A personal computer can run software that emulates the function of a terminal, sometimes allowing concurrent use of local programs and access to a distant host system.
Early user terminals connected to computers were electromechanical teleprinters (TTYs), such as the model 33 Teletype, originally used for telegraphy or the Friden Flexowriter. Later printing terminals such as the DECwriter were developed. However printing terminals were limited by the speed at which paper could be printed, and for interactive use the paper record was unnecessary.
By the early 1970s, many in the computer industry realized that an affordable video data entry terminal could supplant the ubiquitous punch cards and permit new uses for computers that would be more interactive. The problem was that the amount of memory needed to store the information on a page of text was comparable to the memory in low end minicomputers then in use. Displaying the information at video speeds was also a challenge and the necessary control logic took up a rack worth of pre-integrated circuit electronics. One company announced plans to build a video terminal for $15,000 and attracted a large backlog of orders, but folded when their engineering plans, which included fabricating their own ICs, proved too ambitious. Another approach involved the use of the storage tube, a specialized CRT developed by Tektronix that retained information written on it without the need to refresh.
Early video computer displays were sometimes nicknamed "Glass TTYs" and used individual logic gates, with no CPU. One of the motivations for development of the microprocessor was to simplify and reduce the electronics required in a terminal. Most terminals were connected to mainframe computers and often had a green or amber screen. Typically terminals communicate with the computer via a serial port, often using the RS-232 serial interface. IBM systems communicated over a coaxial cable using IBM's SNA protocol.
Later, so-called "intelligent" terminals were introduced, such as the VT52 and VT100 made by DEC, both of which are still widely emulated in software. These were called "intelligent" because they had the capability of interpreting escape sequences to position the cursor and control the display. Notable non-VT100 computer terminal types include the IBM 3270, various Wyse models (whose Wyse 60 was a best-seller—many are still in use), and the Tektronix 4014.
During the late 1970s and early 1980s, there were dozens of manufacturers of terminals including DEC, Wyse, Televideo, Hewlett Packard, IBM, Lear-Siegler and Heath, many of which had incompatible command sequences.
While early IBM PCs had single color green screens, these screens were not terminals. The screen of a PC did not contain any character generation hardware; all video signals and video formatting were generated by the video display card in the PC. With suitable terminal software PCs could, however, emulate a terminal, if connected to a mainframe computer. Eventually microprocessor-based personal computers greatly reduced the market demand for terminals. Today, most PC  clients provide emulation of the most common terminal—the DEC VT100.
Originally text terminals were electronic computer terminals connected to computers by a serial port, but later computers have built-in system consoles, and terminal emulator programs that work in a graphical desktop environment. Graphical displays have not eradicated the text terminal as it is convenient for computer programmers and appropriate for command-line interfaces and text user interfaces. Most programming languages support standard streams for inputting and printing text, and it is simple to connect the streams to a text terminal.
The System console is a text terminal used to operate a computer. Modern computers have a built-in keyboard and display for the console. Some Unix-like operating systems such as Linux, FreeBSD and Mac OS X have virtual consoles to provide several text terminals on a single computer.
A terminal emulator is a computer program in a graphical windowing system that lets the user operate a text terminal in a window. This lets text terminals coexist with modern graphical user interfaces. Popular terminal emulators include Win32 console and xterm.
The fundamental type of application running on a text terminal is a command line interpreter or shell, which prompts for commands from the user and executes each command after a press of Enter. This includes Unix shells and some interactive programming environments. In a shell, most of the commands are small applications themselves.
Another important application type is that of the text editor. A text editor occupies the full area of display, displays one or more text documents, and allows the user to edit the documents. The text editor has, for many uses, been replaced by the word processor, which usually provides rich formatting features that the text editor lacks. The first word processors used text to communicate the structure of the document, but later word processors operate in a graphical environment and provide a WYSIWYG simulation of the formatted output.
In the simplest form, a text terminal is like a file. Writing to the file displays the text and reading from the file produces what the user enters. In unix-like operating systems, there are several character special files that correspond to available text terminals.
For other operations, there are special escape sequences, control characters and
termios functions that a program can use, most easily via a library such as ncurses. For more complex operations, the programs can use terminal specific ioctl system calls.
For an application, the simplest way to use a terminal is to simply write and read text strings to and from it sequentially. The output text is scrolled, so that only the last several lines (typically 24) are visible. Unix systems typically buffer the input text until the Enter key is pressed, so the application receives a ready string of text. In this mode, the application need not know much about the terminal.
For many interactive applications this is not sufficient. One of the common enhancements is command line editing (assisted with such libraries as readline); it also may give access to command history. This is very helpful for various interactive command line interpreters.
Even more advanced interactivity is provided with full-screen applications. Those applications completely control the screen layout; also they respond to key-pressing immediately. This mode is very useful for text editors, file managers and web browsers. In addition, such programs control the color and brightness of text on the screen, and decorate it with underline, blinking and special characters (e.g. box drawing characters).
To achieve all this, the application must deal not only with plain text strings, but also with control characters and escape sequences, which allow to move cursor to an arbitrary position, to clear portions of the screen, change colors and display special characters — and also respond to function keys.
The great problem here is that there are so many different terminals and terminal emulators, each with its own set of escape sequences. In order to overcome this, special libraries (such as curses) have been created, together with terminal description databases, such as Termcap and Terminfo.
In recent years, the general switching of users to GUI has lessened the attention paid to terminal-handling libraries and to terminal emulation, and almost stalled the debugging efforts.
The specific meaning of the term dumb terminal can vary depending on the context in which it is used.
In the context of traditional computer terminals that communicate over a serial RS-232 connection, dumb terminals are those that can interpret a limited number of control codes (CR, LF etc.) but do not have the ability to process special escape sequences that perform functions such as clearing a line, clearing the screen, or controlling cursor position. In this context dumb terminals are sometimes dubbed glass teletypes, for they essentially have the same limited functionality as does a mechanical teletype. This type of dumb terminal is still supported on modern Unix-like systems by setting the environment variable TERM to dumb. Smart or intelligent terminals are those that also have the ability to process escape sequences, in particular the VT52, VT100 or ANSI escape sequences.
In the broader context that includes all forms of keyboard/screen computer communication devices, including personal computers, diskless workstations, network computers, thin clients, and X Terminals, the term dumb terminal is sometimes used to refer to any type of traditional computer terminal that communicates serially over a RS-232 connection that does not locally process data or execute user programs.
The term dumb terminal sometimes also refers to public computer terminals that are limited to monochrome text-only capabilities, or to terminals that transmit each character as it is typed rather than waiting until it is polled by a host computer.
A vector-mode display directly draws lines on the face of a cathode-ray tube under control of the host computer system. The lines are continuously formed, but since the speed of electronics is limited, the number of concurrent lines that can be displayed at one time is limited. Vector-mode displays were historically important but are no longer used. Practically all modern graphic displays are raster-mode, descended from the picture scanning techniques used for television, in which the visual elements are a rectangular array of pixels. Since the raster image is only perceptible to the human eye as a whole for a very short time, the raster must be refreshed many times per second to give the appearance of a persistent display. The electronic demands of refreshing display memory meant that graphic terminals were developed much later than text terminals, and initially cost much more. Most terminals today are graphical - that is, they can show images on the screen. The modern term for graphical terminal is "thin client". A thin client typically uses a protocol like RDP for Microsoft Windows, or X11 for Unix-terminals. The bandwidth needed depends on the protocol used, the resolution, and the color depth.
Modern graphic terminals allow display of images in color, and of text in varying sizes, colors, and fonts (type faces).
When using a graphical user interface (or GUI) like the X Window System, one's display is typically occupied by a collection of windows associated with various applications, rather than a single stream of text associated with a single process. In this case, one may use a terminal emulator application within the windowing environment. This arrangement permits terminal-like interaction with the computer (for running a command line interpreter, for example) without the need for a physical terminal device.