Emacs

Emacs

GNU Emacs 22.0.91.1 with multiple buffers and syntax highlighting for LaTeX, C#, and C.
Original author(s)	Richard Stallman
Developer(s)	the GNU project
Initial release	1976, 36–37 years ago
Stable release	23.3 [ edit]  (March 10, 2011 (2011-03-10)) [ ±]
Preview release	Non [ ±]
Written in	C and Emacs lisp
Operating system	Cross-platform
Available in	English
Type	Text editor
License	GNU General Public License
Website	http://www.gnu.org/software/emacs/

Emacs is a class of feature-rich text editors, usually characterised by their extensibility. Development began in the mid-70s and, as of 2008, is still active. Emacs text editors are most popular with technically proficient computer users and computer programmers. The most popular version of Emacs is GNU Emacs, a part of the GNU project, which is commonly referred to simply as "Emacs".

The GNU Emacs manual describes it as "the extensible, customizable, self-documenting, real-time display editor." It is also the most ported of the implementations of Emacs. As of March 2008, the latest stable release of GNU Emacs is version 22.2. The other major Emacs is XEmacs.

The original EMACS was a set of Editor MACroS for the TECO editor. It was written in 1976 by Richard Stallman, initially together with Guy L. Steele, Jr.. It was inspired by the ideas of TECMAC and TMACS, a pair of TECO-macro editors written by Steele, Dave Moon, Richard Greenblatt, Charles Frankston, and others. Aside from GNU Emacs, another version of Emacs in use is XEmacs, which is a fork of GNU Emacs started in 1991. XEmacs has remained mostly compatible and continues to use the same extension language, Emacs Lisp, as GNU Emacs. Almost all of GNU Emacs and XEmacs are written in Emacs Lisp, so the extensibility of Emacs' features is deep.

In Unix culture, Emacs is one of the two main contenders in the traditional editor wars, the other being vi. The word "emacs" is often pluralized as emacsen, by analogy with boxen (itself used by analogy with oxen) and VAXen.

History

Emacs development began at the MIT AI Lab during the 1970s. Before its introduction, the default editor on the Incompatible Timesharing System (ITS), the operating system on the AI Lab's PDP-6 and PDP-10 computers, was a line editor known as Text Editor and Corrector (TECO). Unlike most modern text editors, TECO has separate modes which the user used to either add text, edit existing text, or display the document. Typing characters into TECO did not place those characters directly into a document; one had to write a series of instructions in the TECO command language telling it to enter the required characters, during which time the edited text was not displayed on the screen. This behaviour is similar to the program ed.

Richard Stallman visited the Stanford AI Lab in 1972 or 1974 and saw the lab's "E" editor, written by Fred Wright. The editor had an intuitive WYSIWYG behaviour as is used almost universally by modern text editors, which impressed Stallman. He returned to MIT where Carl Mikkelsen, a hacker at the AI Lab, had added a combined display+editing mode called "Control-R" to TECO, allowing the screen display to be updated each time the user entered a keystroke. Stallman reimplemented this mode to run efficiently, then added a macro feature to the TECO display-editing mode, allowing the user to redefine any keystroke to run a TECO program.

Another feature of E which TECO lacked was random-access editing. Since TECO's original implementation was designed for editing paper tape on the PDP-1, it was a page-sequential editor. Typical editing could only be done on one page at a time, in the order of the pages in the file. To provide random access in Emacs, Stallman decided not to adopt E's approach of structuring the file for page-random access on disk, but instead modified TECO to handle large buffers more efficiently, and then changed its file management philosophy to read, edit, and write the entire file as a single buffer. Almost all modern editors use this approach.

The new version of TECO was instantly popular at the AI Lab, and soon there accumulated a large collection of custom macros, whose names often ended in "MAC" or "MACS", which stood for "macro". Two years later, Guy Steele took on the project of unifying the overly diverse macros into a single set. After one night of joint hacking by Steele and Stallman, the latter finished the implementation, which included facilities for extending and documenting the new macro set. The resulting system was called EMACS, which stood for "Editing MACroS". An alternative version is that EMACS stood for "E with MACroS", a dig at E's lack of a macro capability. According to Stallman, he picked the name Emacs "because was not in use as an abbreviation on ITS at the time." (It has also been noted that " Emack & Bolio's" was the name of a popular ice cream store in Boston, within walking distance of MIT. A text-formatting program used on ITS was later named BOLIO by Dave Moon, who frequented that store. However, Stallman did not like that ice cream, and did not even know of it when choosing the name "Emacs"; this ignorance is the basis of a hacker koan, Emacs and Bolio).

Stallman saw a problem in too much customization and de-facto forking and set certain conditions for usage. He later wrote:

"EMACS was distributed on a basis of communal sharing, which means all improvements must be given back to me to be incorporated and distributed."

The original Emacs, like TECO, ran only on the PDP line of computers. Its behaviour was different enough from TECO to be considered a text editor in its own right. It quickly became the standard editing program on ITS. It was also ported from ITS to the Tenex and TOPS-20 operating systems by Michael McMahon, but not Unix, initially. Other contributors to early versions of Emacs include Kent Pitman, Earl Killian, and Eugene Ciccarelli. By 1979, Emacs was the editor used by most people in MIT's AI lab and its Computer Science lab.

Other early implementations

Many Emacs-like editors were written in the following years for other computer systems, including SINE (Sine is not EMACS), EINE ("EINE Is Not EMACS") and ZWEI ("ZWEI Was EINE Initially", for the Lisp machine), which were written by Michael McMahon and Daniel Weinreb. ("Eine" and "zwei" are German for "one" and "two") Weinreb's EINE was the first Emacs written in Lisp. In 1978, Bernard Greenberg wrote Multics Emacs at Honeywell's Cambridge Information Systems Lab. Many versions of Emacs, including GNU Emacs, later adopted Lisp as an extension language.

The first Emacs-like editor to run on Unix was Gosling Emacs, written in 1981 by James Gosling (who later invented NeWS and the Java programming language). It was written in C and, notably, used a language with Lisp-like syntax known as Mocklisp as an extension language. In 1984 it was proprietary software.

GNU Emacs

In 1984, Stallman began working on GNU Emacs to produce a free software alternative to Gosling Emacs; initially it was based on Gosling Emacs, but Stallman replaced the Mocklisp interpreter at its heart with a true Lisp interpreter, which entailed replacing nearly all of the code. It became the first program released by the nascent GNU project. GNU Emacs is written in C and provides Emacs Lisp (itself implemented in C) as an extension language. Version 13, the first public release, was made on March 20, 1985. The first widely distributed version of GNU Emacs was 15.34, which appeared later in 1985. Versions 2 to 12 never existed. Earlier versions of GNU Emacs had been numbered "1.x.x", but sometime after version 1.12 the decision was made to drop the "1", as it was thought the major number would never change.

GNU Emacs was later ported to Unix. Since it had more features that Gosling Emacs, in particular a full-featured Lisp as extension language, it soon replaced Gosling Emacs as the de facto Emacs editor on Unix.

Until 1999, GNU Emacs development was relatively closed, to the point where it was used as an example of the "Cathedral" development style in The Cathedral and the Bazaar. The project has since adopted a public development mailing list and anonymous CVS access. Development takes place in a single CVS trunk, which was at version 23.0.60 as of mid-2008.

Richard Stallman remained the maintainer of GNU Emacs through most of the time, but took several breaks. The latest one persists and started 2008, when maintenance was handed over to Stefan Monnier and Chong Yidong.

Release history

Version	Release date	Significant changes
22.2	Mar 26, 2008	New support for the Bazaar, Mercurial, Monotone, and Git version control systems New major modes for editing CSS, Vera, Verilog, and BibTeX style files Improved scrolling support in Image mode
22.1	June 2, 2007	Support for the GTK+ graphical toolkit Support for Drag-and-drop on X Support for images, toolbar, and tooltips Unicode support Support for Mac OS X
21.1	October 20, 2001	Support for displaying colors and some other attributes on terminals Built-in horizontal scrolling Sound support Wheel mouse support Improved menu bar layout
20.1	September 17, 1997	Multi-lingual support
19.28	September 15, 1997	Support for multiple frames using the X Windowing System New interface for version control systems, called VC New hexl mode, which is a hexadecimal editing mode
18.24	October 2, 1986
17.36	December 20, 1985	Backup file version numbers
16.56	July 15, 1985
15.10	April 11, 1985
13.0?	March 20, 1985

XEmacs

Beginning in 1991, Lucid Emacs was developed by Jamie Zawinski and others at Lucid Inc., based on an early alpha version of GNU Emacs 19. The codebases soon diverged, and the separate development teams gave up trying to merge them back into a single program. This was one of the most famous early forks of a free software program. Lucid Emacs has since been renamed XEmacs; it remains the second most popular variety of Emacs, after GNU Emacs.

Other modern forks and clones

Many other derivatives of the GNU Emacs have been made, such as a Japanese version for MS Windows called Meadow Steve Young's fork of XEmacs called SXEmacs, and a version which focuses on integrating with the Apple Macintosh user interface called Aquamacs.

Emacs clones, not based on GNU Emacs source code, are more numerous. One motivation for writing clones was that GNU Emacs was initially targeted at computers with a 32-bit flat address space, and at least 1 MiB of RAM. At a time when such computers were considered high end, this left an opening for smaller reimplementations. Some notable modern clones include:

• MicroEMACS, originally written by Dave Conroy and further developed by Daniel Lawrence, which exists in many variations. The editor is used by Linus Torvalds.
• mg, originally called MicroGNUEmacs (and later mg2a), a public-domain offshoot of MicroEMACS intended to more closely resemble GNU Emacs. Now installed by default on OpenBSD.
• NotGNU , a small, fast, freeware implementation for DOS, Win16, Win32 and Linux by Julie Melbin.
• JOVE (Jonathan's Own Version of Emacs), a non-programmable Emacs implementation for UNIX-like systems by Jonathan Payne.
• Freemacs, a DOS version with an extension language based on text macro expansion, all within the original 64 KiB flat memory limit.
• MINCE (MINCE Is Not Complete Emacs), a version for CP/M from Mark of the Unicorn. MINCE evolved into Final Word, which eventually became the Sprint word processor from Borland.
• Zile
• Climacs, an Emacs-variant implemented in Common Lisp and more influenced from Zmacs than GNU Emacs.

Licensing

The source code, including both the C and Emacs Lisp components, is freely available for examination, modification, and redistribution, under the terms of the GNU General Public License (GPL).

For GNU Emacs (and GNU packages in general), it remains policy to accept significant code contributions only if the copyright holder executes a suitable disclaimer or assignment of their copyright interest to the Free Software Foundation, although one exception was made to this policy for the MULE (MULtilingual Extension, which handles Unicode and more advanced methods of dealing with other languages' scripts) code since the copyright holder is the Japanese government and copyright assignment was not possible. This does not apply to extremely minor code contributions of less than 10 lines or bug fixes. This policy is in place so that FSF can defend the software in court if its copyleft licence is violated.

Older versions of the GNU Emacs documentation were released under an ad-hoc license which required the inclusion of certain text in any modified copy. In the GNU Emacs user's manual, for example, this included how to obtain GNU Emacs and Richard Stallman's political essay The GNU Manifesto. The XEmacs manuals, which were inherited from older GNU Emacs manuals when the fork occurred, have the same license. Newer versions of the documentation use the GNU Free Documentation License and makes use of "invariant sections" to require the inclusion of the same documents, additionally requiring that the manuals proclaim themselves as GNU Manuals.

Features

Emacs is primarily a text editor, not a word processor; it is geared toward manipulating pieces of text, rather than manipulating the typeface (the "font") of the characters or printing documents (though Emacs can do these as well). Emacs provides commands to manipulate words and paragraphs (deleting them, moving them, moving through them, and so forth), syntax highlighting for making source code easier to read, and "keyboard macros" for performing arbitrary batches of editing commands defined by the user.

GNU Emacs is a "real-time display" editor in that your edits are displayed on the screen as you make them. This is standard behaviour for modern text editors.

Customisable

Almost all of the functionality in the editor, ranging from basic editing operations such as the insertion of characters into a document to the configuration of the user interface, is controlled by a dialect of the Lisp programming language known as Emacs Lisp. This unique and unusual design provides many of the features found in Emacs. In this Lisp environment, variables and even entire functions can be modified without having to recompile or even restart the editor.

There are three primary ways to customize Emacs. The first is the customize extension, which allows the user to set common customization variables, such as the colour scheme, using a graphical interface, etc. This is intended for Emacs beginners who do not want to work with Emacs Lisp code.

The second is to collect keystrokes into macros and replay them to automate complex, repetitive tasks. This is often done on an ad-hoc basis, with each macro discarded after use, although macros can be saved and invoked later.

The third method for customizing Emacs is using Emacs Lisp. Usually, user-supplied Emacs Lisp code is stored in a file called .emacs, which is loaded when Emacs starts up. The .emacs file is often used to set variables and key bindings different from the default setting, and to define new commands that the user finds convenient. Many advanced users have .emacs files hundreds of lines long, with idiosyncratic customizations that cause Emacs to diverge wildly from the default behaviour.

Extensible

As a result, the behaviour of Emacs can be modified almost without limit, either directly by the user, or by loading Emacs Lisp code known variously as "libraries", "packages", or "extensions".

Emacs contains a large number of Emacs Lisp libraries, and more "third-party" libraries can be found on the Internet. Emacs can be used as an Integrated Development Environment (IDE), allowing programmers to edit, compile, and debug their code within a single interface. Other libraries perform more unusual functions. A few examples are listed below:

• Calc, a powerful RPN numerical calculator
• Calendar-mode, for keeping appointment calendars and diaries
• Dunnet, a text adventure
• Ediff, for working with diff files interactively.
• Emerge, for comparing files and combining them
• Emacs/W3, a web browser
• ERC, an IRC client
• Gnus, a full-featured newsreader and email client (and early evidence for Zawinski's Law, along with Rmail and VM)
• VM (or 'View Mail'), a full-featured email client
• MULE, MultiLingual extensions to Emacs, allowing editing text written in multiple languages, somewhat analogous to Unicode
• Info, an online help-browser
• Planner, a personal information manager for Emacs

Many third-party libraries can be found on the Internet; for example, there is a library called wikipedia-mode for editing Wikipedia articles. There is even a Usenet newsgroup, gnu.emacs.sources, which is used for posting new libraries. Some third-party libraries eventually make their way into Emacs, thus becoming a "standard" library.

Performance

The downside to Emacs' Lisp-based design is a performance overhead resulting from loading and interpreting the Lisp code. On the systems in which Emacs was first implemented, Emacs was often noticeably slower than rival text editors. Several joke backronyms allude to this: Eight Megabytes And Constantly Swapping (from the days when eight megabytes was a lot of memory), Emacs Makes A Computer Slow, Eventually Mallocs All Computer Storage, and Eventually Makes All Computers Sick. However, modern computers are fast enough that Emacs is seldom felt to be slow. In fact, Emacs starts up quicker than most modern word processors. Other joke backronyms describe the user interface: Escape Meta Alt Control Shift.

Platforms

Emacs is one of the most ported non-trivial computer programs. It runs on a wide variety of operating systems, including most Unix-like systems (GNU/Linux, the various BSDs, Solaris, AIX, IRIX, Mac OS X, etc.), MS-DOS, Microsoft Windows and OpenVMS. Unix systems, both free and proprietary, frequently provide Emacs bundled with the operating system.

Emacs runs on both text terminals and graphical user interface (GUI) environments. On Unix-like operating systems, Emacs uses the X Window System to produce its GUI, either directly or using a "widget toolkit" such as Motif, LessTif, or GTK+. Emacs can also use the native graphical systems of Mac OS X (using the Carbon or Cocoa interfaces) and Microsoft Windows. The graphical interface provides menubars, toolbars, scrollbars, and context menus.

Editing modes

Emacs adapts its behaviour to the type of text it is editing by entering editing modes called "major modes". Major modes are defined for ordinary text files, source code for many programming languages, HTML documents, TeX and LaTeX documents, and many other types of text. Each major mode is an Emacs Lisp program that extends the editor to behave more conveniently for the particular type of text it covers. Typical major modes will provide some or all of the following common features:

• Syntax highlighting (called "font lock" in Emacs): using different fonts or colors to display keywords, comments, and so forth.
• Automatic indentation: maintaining consistent formatting within a file.
• "Electric" features, i.e. the automatic insertion of elements such as spaces, newlines, and parentheses which are required by the structure of the document.
• Special editing commands: for example, major modes for programming languages usually define commands to jump to the beginning and the end of a function, while major modes for markup languages such as XML provide commands to validate documents or insert closing tags.

Additional features can be added by using "minor modes". While only one major mode is associated with a buffer at a time, multiple minor modes can be simultaneously active. These may affect documents directly. For example, the major mode for the C programming language defines a different minor mode for each of the popular indent styles. Or, they may affect the editing environment instead. For example, "Winner mode" adds the ability to undo changes to the window configuration, while "MMM-mode" adds the ability to use more than one major mode in a single file, as required in common cases such as editing an HTML document with embedded CSS and JavaScript.

Self-documenting

The first Emacs included an innovative help library that can display the documentation for every single command, variable, and internal function. (It may have originated this technique.) Because of this, Emacs was described as "self-documenting" in that it presents its own documentation, not only of its normal features but also of its current state, to the user. For example, the user can find out about the command bound to a particular keystroke simply by entering C-h k (which runs the command describe-key), followed by the keystroke. Each function included a documentation string, specifically to be used for showing to the user on request. The practice of giving functions documentation strings subsequently spread to various programming languages such as Lisp and Java.

Further, through Emacs's help system, users can be taken to the actual code for each function - be it a built-in library or an installed third-party library.

Emacs also has a built-in tutorial. When Emacs is started with no file to edit, it displays instructions for performing simple editing commands and invoking the tutorial.

Manuals

Apart from the built-in documentation, Emacs has an unusually long, detailed and well-written manual. An electronic copy of the GNU Emacs Manual, written by Richard Stallman, is included with GNU Emacs and can be viewed with the built-in Info browser. Two other manuals, the Emacs Lisp Reference Manual by Bill Lewis, Richard Stallman, and Dan Laliberte, and Programming in Emacs Lisp by Robert Chassell, are also included. Apart from the electronic versions, all three manuals are also available in book form, published by the Free Software Foundation. XEmacs has a similar manual to the GNU Emacs Manual, which forked from the GNU Emacs Manual at the same time as the XEmacs software.

Internationalization

Emacs supports the editing of text written in many human languages. There is support for many alphabets, scripts, writing systems, and cultural conventions. Emacs provides spell checking for many languages by calling external programs such as ispell. Many encoding systems, including UTF-8, are supported. Emacs 22 has full unicode support, however, it uses Emacs-specific encoding internally, necessitating conversion upon load and save. UTF-8 will become the Emacs-internal encoding in Emacs 23.

However, the Emacs user interface is in English, and has not been translated into any other language, with the exception of the beginners' tutorial.

For visually impaired and blind users, there is a subsystem called Emacspeak which allows the editor to be used through audio feedback only.

Using Emacs

Commands

In the normal editing mode, Emacs behaves like other text editors: the character keys (a, b, c, 1, 2, 3, etc.) insert the corresponding characters, the arrow keys move the editing point, backspace deletes text, and so forth. Other commands are invoked with modified keystrokes, pressing the control key and/or the meta key/ alt key in conjunction with a regular key. Every editing command is actually an invocation of a function in the Emacs Lisp environment. Even a command as simple as typing a to insert the character a involves calling a function--in this case, self-insert-command.

Alternatively, if a user would prefer IBM Common User Access style keys, "cua-mode" can be used. This has been a third-party package up to, and including, GNU Emacs 21, but is included in GNU Emacs 22.

Note that the commands save-buffer and save-buffers-kill-emacs use multiple modified keystrokes. For example, C-x C-c means: while holding down the control key, press x; then, while holding down the control key, press c. This technique, allowing more commands to be bound to the keyboard than with the use of single keystrokes alone, was popularized by Emacs, which got it from TECMAC, one of the TECO macro collections that immediately preceded Emacs. It has since made its way into modern code editors like Visual Studio, and is even used to some extent by some consumer word processors such as Microsoft Word.

Some Emacs commands work by invoking an external program (such as ispell for spell-checking or gcc for program compilation), parsing the program's output, and displaying the result in Emacs.

Minibuffer

The minibuffer, normally the bottommost line, is where Emacs requests information. Text to target in a search, the name of a file to read or save and similar information is entered in the minibuffer. When applicable, command line completion is usually available using the tab and space keys.

File management and display

Emacs keeps text in objects called buffers. The user can create new buffers and dismiss unwanted ones, and several buffers can exist at the same time. Most buffers contain text loaded from text files, which the user can edit and save back to disk. Buffers are also used to store temporary text, such as the documentation strings displayed by the help library.

In both text terminal and graphical modes, Emacs is able to split the editing area into separate sections (referred to since 1975 as "windows", which can be confusing on systems that have another concept of "windows" as well), so that more than one buffer can be displayed at a time. This has many uses. For example, one section can be used to display the source code of a program, while another displays the results from compiling the program. In graphical environments, Emacs can also launch multiple graphical-environment windows, known as "frames" in the context of Emacs.

Emacs Pinky

Because of Emacs' dependence on the modifier keys, in particular, how the control key is pressed with the little finger ("pinky"), heavy Emacs users have experienced pain in their pinky fingers (see repetitive strain injury and fat-finger). This has been dubbed the "Emacs Pinky", and vi advocates often cite it as a reason to switch to vi. To alleviate this situation, many Emacs users transpose the left control key and the left caps-lock key or define both as control keys. To put the control key back where it was (usually the leftmost key on the middle row of characters) when Emacs key combinations were chosen, before the era of IBM-style PC keyboards. Others use viper-mode, a feature built into Emacs that allows the use of the vi keys for basic text editing and the Emacs keys for more advanced features. Others use special keyboards such as Kinesis's Contoured Keyboard, which reduces the strain by moving the modifier keys so that they are in a position to be easily pushed by the thumb, or Microsoft Natural keyboard, which has large modifier keys placed symmetrically on both sides of the keyboard so that they can be pressed with the palm.