Keywords

1 Overview

Today, word processing means using a standard application for desktop or mobile computers in order to (a) write and revise any kind of text, and to (b) apply formatting to a text for its output in printed or other form. Hence, word processing software serves as a technology that combines two distinct modes of text production: the composition of a manuscript (from first draft to final version) and the typographic preparation of a document (for publication and distribution in printed or electronic form). By merging the typewriter with the printing shop, word processors have fundamentally changed the process of writing and publishing and have blended the role of the author with those of the typesetter and the graphic designer.

Daniel Eisenberg (1992), Tim Bergin (2006a, b) and Thomas Haigh (2006) have given concise historical accounts of word processing with a strong emphasis on specific PC applications like WordStar, WordPerfect and Microsoft Word. More recently, Matthew Kirschenbaum (2016) has devoted an extensive study to the “Literary History of Word Processing”. More research on word processing will be presented in the next chapter (Kruse & Rapp, “Word Processing Software: The Rise of MS Word”).

While it originated from a very different technological and economic environment, word processing is closely tied to personal computers and their spectacular commercial success since the late 1970s. The widespread adoption of word processors became possible only with the advent of microcomputers and with the rise of the IBM PC platform in the 1980s (Haigh & Ceruzzi, 2021, pp. 227–242). Before that, text editing and formatting tools were confined to time-shared mainframe installations, minicomputers, and dedicated office computers, they addressed narrow, highly trained user groups with specific demands, and they were not open the public. PC word processors, on the other hand, were—and still are—designed for the wider audience and a broad range of purposes. The two domains are historically demarcated by the emergence of a software industry for business and private use of PCs in the late 1970s (Campbell-Kelly, 2003, pp. 201–228). Whereas the first users of early text editors had to program their own custom tools (and many committed hackers and software engineers would continue to do so for a long time), office clerks and PC owners since the late 1970s have been doing their word processing with off-the-shelf, commercially—or freely—available applications.

Conceptually as well as technologically, one of the decisive moments in the evolution of word processing software was the inclusion of the video screen. As electronic displays were uncommon up until the 1970s, early digital text editing usually happened character for character and line by line on hard-copy terminals like teleprinters and customized electric typewriters (Haigh, 2006, pp. 13–15). By putting characters on a real-time video screen, computers turned written text into a ‘malleable’ visual object and opened a new kind of “writing space” (Bolter, 1991) in which individual letters and words, whole sentences or larger textual units could be easily and instantly manipulated. Equally important, bitmapped video screens allowed for WYSIWYG or “What you see is what you get”, i.e., a mode of display that shows all the formatting of a text (with different typefaces, sizes and so on) and its page layout just as it would appear when printed on paper.

Long before video screens for word processing were actually implemented, a few visionaries had already pondered the possibilities and potentials of modern media technology for writing. One of the first authors to do so, and a recurring point of reference in future discussions, was Vannevar Bush. His article “As We May Think” from 1945 established the idea of a mechanized database of documents projected onto ‘translucent screens’ (Bush, 1945, p. 107). Bush’s text exerted a strong influence on two other visionaries, Douglas Engelbart and Ted Nelson. Engelbart expanded on Bush’s ideas during the 1960s with his own concept of “Augmenting Human Intellect”. Displaying text on a computer screen, Engelbart argued, would allow for completely “new methods of manipulating symbols” (Engelbart, 1962, p. 75).

Nelson also continued on Bush’s work. In his treatise on “A File Structure for the Complex, the Changing and the Indeterminate” from 1965, Nelson hypothesized about a computerized ‘dream file’: an electronic text environment that would assist the author with “manuscripts in progress” through all stages of the writing process, and particularly “during the early periods of muddled confusion, when his [or her] ideas are scraps, fragments, phrases, and contradictory overall designs. And it must help him [or her] through to the final draft with every feasible mechanical aid—making the fragments easy to find, and making easier the tentative sequencing and juxtaposing and comparing” (Nelson, 1965, p. 88). Digital computers, in short, would foster the creativity of writers by making written text easily modifiable and re-arrangeable on the screen.

The screen was also instrumental for another decisive shift in writing. For computer displays can act as more than just intermediaries in the digital production of paper documents. Bush, Engelbart, and Nelson all thought about and worked on the possibility of linking together individual documents and fragments of text through mechanical and electronic means—an idea for which Nelson coined the term ‘hypertext’. The concept of strictly digital documents that were not to be printed on paper but would be written and read exclusively on video screens began to take shape with early hypertext systems in the 1960s (Barnet, 2013). From the 1980s on, networked computers with services like bulletin board systems (BBS), Usenet, and, finally, the World Wide Web (WWW), turned this idea into reality. Today, the screen has supplanted paper for many purposes and has become a primary medium for displaying text in its own right. While common word processors are not geared towards creating hypertexts and webpages, they are routinely used to write documents that are meant first and foremost for the screen.

After word processing on PCs had become wide-spread and with the revolution of the Internet and the WWW looming at the beginning of the 1990s, writers like Jay D. Bolter (1991) and George P. Landow (1992) again discussed the new electronic ‘writing space’ and hypertextuality from a historical and philosophical perspective. Other notable voices in the debate include Michael Heim (1987), Vilém Flusser (2011) and Jacques Derrida (2005). The consensus of such theoretical analyses seemed to be that word processing had changed writing from the task of producing a fixed, stable, ‘bookish’ text by a single identifiable author to a continual process of creating and revising ever-changing digital documents that constitute a highly dynamic hypertext of multiple and shifting authorial agents.

Notwithstanding the substantial changes brought about by digital hard- and software, our concept of text and even our basic methods of generating letters, words, and sentences have proven remarkably resilient. Most digital texts still largely follow the traditional visual architecture of the “bookish text” that goes back to medieval scholasticism (Illich, 1993, p. 115). And most digital writing is still done by pressing keys on typewriter-like keyboards whose layouts were invented and perfected at the end of the nineteenth century. It is no wonder, then, that the most successful word processing applications still adhere to the model of the printed page.

2 Core Idea of the Technology

The core ideas of technology that led to word processing as we know it today are (roughly in chronological order):

  1. 1.

    The interactive use of computers.

  2. 2.

    Entering and editing text on computers.

  3. 3.

    Using interactive editing tools for “regular” texts (not computer programs).

  4. 4.

    Formatting digital text according to traditional typographic conventions.

  5. 5.

    Putting text on a computer screen.

  6. 6.

    Printing digital text to paper.

  7. 7.

    Computer systems usable by non-professionals.

  8. 8.

    Simulating paper documents on computer screens.

  9. 9.

    Automating clerical work with computers.

  10. 10.

    Computers available to and affordable for everybody.

  11. 11.

    A market for standard word processing software solutions.

The technological foundation of digital word processing is the interactive use of a computer while it is running, i.e. the possibility of a rapid back and forth information exchange between user and system through suitable input/output devices. Interactive computing started around 1960 with the first time-shared installations and minicomputers (Haigh & Ceruzzi, 2021, pp. 109–138). In the beginning, teleprinters were the preferred interface for this new kind of ‘dialogue’ between man and machine. They were well-known from telegraphy, relatively cheap, reliable in operation, and, most importantly, easy to adapt for use with computers: Employing telegraphic character encodings like the Baudot or Murray code, teleprinters already processed writing in digital form. As an additional benefit, they could often read and write texts from and to paper tape, a popular storage medium of early computers.

One of the very first uses of interactive computing was the inspection and debugging of programs. Doing this online was much easier and faster than poring over paper printouts of faulty code and failed runs (van Dam & Rice, 1971, p. 97). It was soon realized that computers could also help with the preparation of program tapes. At the time, computer code was developed using pen and paper, written by hand (sometimes on special coding sheets), then mechanically transferred to paper tape or punched cards, and finally fed to the computer. While faulty cards could be easily swapped, a tape containing an error had to be punched again from scratch. Harnessing the computer for debugging programs and producing corrected tapes would considerably speed up the software development process.

Colossal Typewriter, created in 1960 at Massachusetts Institute of Technology (MIT) for the Programmed Data Processor 1 (PDP-1), the world’s first commercial minicomputer, is arguably the oldest known digital text editor. As the name says, it turned the 120,000 US dollar computer installation into a giant typewriter for the purpose of “tape preparation and tape editing” (McCarthy & Silver, 1960, p. 1). By today’s standards, Colossal Typewriter was extremely rudimentary and cumbersome to use. But it made life much easier for programmers and kicked off a slew of subsequent text editors with ever more advanced capabilities and features. The most important of these is probably TECO from 1962, also initially for the PDP-1 (Murphy, 2009). TECO is the direct ancestor of the Emacs editor which was developed by Richard Stallman in the 1970s and is still used by many programmers and some non-programmers on PCs even today. Again, the name of the program is revealing: While it was later renamed Text Editor & Corrector, the acronym TECO originally stood for Tape Editor & Corrector, pointing to the primary medium of early computing and text editing.

As Colossal Typewriter, TECO and their successors spread throughout computer labs and facilities in parallel to the rise of time-sharing systems and minicomputers during the 1960s, programmers realized that these tools could be used to write not just code but regular texts in prose as well. Soon they also created technical documents, office memos, lab reports, and other pieces with the same programs they used for editing code (Brock, 2018, p. 9). In the process, text editors were gradually extended and enhanced for the new tasks. And because regular texts were read by humans from pages of paper (not by computers from paper tapes), they needed to be organized accordingly for printouts with proper line, paragraph, and page breaks, headers and footers, page numbers etc. Consequently, the first methods and instruments for digital text-formatting were invented.

The common way to do this was, and still is, for the user to put special control characters or commands like .BR or .CENTER into the text at the right places (what is called “markup” today). When a text was printed, the control characters or commands in the text were processed by the formatting program and effected the desired typographic results like page breaks, centered lines, indented paragraphs etc. One of the earliest tools, Type Justifying Program 2 (TJ-2), again developed for the PDP-1 at MIT in 1963, already made use of the computer’s electronic display and light pen for hyphenating words (Massachusetts Institute of Technology, 1963). More influential would become the RUNOFF program, also created at MIT in 1964 for the time-sharing system CTSS (Saltzer, 1964). Not only did its control commands allow for more complex formatting of texts and page layouts than before, RUNOFF also served as the main inspiration for most other formatting programs and languages to follow and is the direct precursor to the basic text processing tools at work in every Unix operating system (including macOS computers) even today.

At the same time as text editors and formatting tools were developed and refined in university labs, a few visionaries and outsiders of the computer industry began to build “free form text editors” (van Dam & Rice, 1971, p. 105) that were meant to enable wholly new ways of thinking and writing. Chief among them was the aforementioned Douglas Engelbart at Stanford Research Institute (SRI) who sought to “augment human intellect” through computer-aided symbol manipulation on electronic displays (Engelbart, 1962), a project that was funded by the US Air Force, NASA, and ARPA (Advanced Research Projects Agency). Together with his team, Engelbart created the oNline-System (NLS), a time-shared computer installation for collaborative work which he famously demonstrated to the public at the Fall Joint Computer Conference in San Francisco in 1968—an event that has become known as ‘The Mother of all Demos’. While NLS boasted many ‘firsts’ (including the computer mouse, linked hypermedia and document version control), it was essentially a screen-based word processor technologically and conceptually far ahead of its time (Bardini, 2000). The NLS, though never successfully commercialized, had a profound impact on computer culture. Probably the most important contribution to digital writing was that it showed to the world what were the possibilities for working with text when it was displayed on a computer screen.

The failure of NLS was its enormous technological and structural complexity and the resulting steep learning curve. Non-specialists found the system nearly impossible to work with (Ittersum, 2008, pp. 156–157). Making computers useable for ordinary people was a big challenge that the industry had to confront in the 1970s. Some of the most important contributions in this regard were made at the Palo Alto Research Center (PARC), founded in 1970 by photocopier giant Xerox in order to invent the ‘office of the future’ (Hiltzik, 1999).

One of the major conceptual breakthroughs at PARC was the enforcement of modeless editing. Simply put, this means that pressing a key on the keyboard when editing a text should always result in the corresponding letter being inserted, never in something else (like, say, the current line being deleted or two paragraphs being transposed). This was obvious to a secretary at PARC who was asked by the software engineers how she imagined editing text on the screen was supposed to work (Perry, 2005, pp. 50–51). But it was news to the programmers who had invented digital writing tools and were accustomed to operate within multiple modes (the aforementioned TECO, for example was actually more of a programming language than a text editor). The insight gained from this ethnographic study was, in short: For word processing, the computer keyboard should serve just a like regular typewriter, not like the control console of a computer.

Probably the biggest of PARC’s contributions to computing was its advancement of the mouse-driven graphical user interface (GUI). With their experimental Alto computer, developed from 1972 on, PARC pioneered high-resolution bitmapped graphics that turned the screen into a digital canvas able to display all kinds of visual information: pictures, tables, drawings, diagrams, and, of course, letters (Haigh & Ceruzzi, 2021, pp. 245–250). Not incidentally, one of the first major applications that made good use of the Alto’s GUI capabilities was a word processor called Bravo, created in 1974. Not only did Bravo show all the details of a text’s graphic formatting on the screen, i.e. the various looks of typefaces, styles, sizes, and so on. It was also the earliest WYSIWYG application—a text editor that let the users see what they were writing on the screen just as it would appear in printed form (Kirschenbaum, 2016, pp. 125–126). With Bravo, the text on the computer screen visually matched the text on the page produced by a laser printer—which was another one of PARC’s ground-breaking inventions.

A later version of Bravo that combined the GUI/WYSIWYG display with modeless editing arguably counts as the world’s first word processing software in the modern sense. Its ease of use and graphic text editing capabilities made it an instant hit—not only with PARC engineers and employees but also with their families and friends who would come in to create personal documents like newsletters, resumes, and school reports on the Xerox Alto machine. And although Xerox failed to capitalize on the many conceptual and technological innovations concerning personal computing at PARC (Hiltzik, 1999, pp. 389–398), the Alto computer and Bravo program would exert a lasting influence on the further evolution of the personal computer and word processing. In 1979, a team of Apple’s engineers were given tours of PARC and demonstrations of the Alto. Their subsequent work on the Macintosh, the first commercially successful GUI computer released in 1984, was heavily inspired by what they had seen. And in 1981, Charles Simonyi, the lead programmer of Bravo, left PARC to join Microsoft where he would oversee the development of productivity applications and become the chief architect of Microsoft Word (Lohr, 2002, pp. 135–136).

The beginnings of commercial word processing outside of research labs like SRI and PARC in the 1970s were much more modest than what Engelbart’s NLS or the Xerox Alto had to offer. Thanks to advances in semiconductor technology and falling prices for memory chips, video terminals as computer interfaces were becoming more common. But they were mostly meant for input of and access to structured data in large companies and public offices, not the editing of regular texts. At the beginning of the 1970s, computers were still too costly and too difficult to operate for untrained clerks and secretaries. Paperwork in offices (and in private homes) was still done almost exclusively on mechanic or electric typewriters. Fittingly, IBM began to use the term “word processing”—an invention by one of its German typewriter division managers (Heilmann, 2012, pp. 141–155)—to promote all of their office products, typewriters, copiers, and dictating machines alike.

Computer-based word processing for the office was championed by other, much smaller companies than IBM like Wang Laboratories. Although it is mostly forgotten today (as are other competitors in the business like Lexitron, Vydec, and Linolex), Wang Labs actually dominated the market for office word processing systems during the second half of the 1970s (Haigh, 2006, p. 22). Their dedicated word processors were, essentially, ‘micros’ like the first PCs released by MITS, Apple, or Commodore at the time, i.e. computers based on 8-bit microprocessors by Intel, MOS Technology, or Zilog. Unlike PCs, however, they were marketed to businesses, came with all the necessary peripherals (keyboard, screen, printer), and were not freely programmable but designed to do one thing, and one thing only: text editing for clerks and secretaries. In fact, Wang Laboratories were very careful not to advertise their word processors as ‘computers’. Instead, they pointed out the similarities to familiar office equipment: “Just type as on a normal typewriter” (quoted in Heilmann, 2012, p. 172). Wang word processors could in no way match the GUI and WYSIWYG capabilities of the Xerox Alto and Bravo. But they were actual products available on the market and quickly garnered a reputation for being easy to use and speeding up paperwork. History has it that the architects of Wang’s initial word processing system wrote the user manual first and only then set about to design the required hardware and software (Haigh, 2006, p. 18).

Despite their early and spectacular success in the word processing market for businesses, Wang would ultimately not survive the vast expansion and consolidation of the microcomputer landscape through the dominance of the IBM PC platform in offices and homes in the 1980s. With personal computers, word processing solutions were transformed into off-the-shelf applications for everybody. In the face of an ever-growing market for PC hardware and software, expensive single-task workstations like Wang’s dedicated word processors had no future.

The story of the PC has been told many times (Bergin, 2006a, b; Campbell-Kelly et al., 2014, pp. 229–251; Ceruzzi, 1999) and need not be recounted here. A few short remarks on the relation of personal computers to word processing have to suffice.

Word processing for PCs was not a revolution—neither in technological nor in conceptual regard. Rather, the problem was one of re-implementing known concepts and techniques as a commercial software product for a novel hardware platform, the ready-assembled microcomputer for home and business users. It is not surprising, then, that the development of PC word processing applications reiterated seminal moments in the larger evolution of digital writing since the 1960s in fast-forward.

PC word processors grew out of homemade editors to program the new machines for which no software existed at first—beginning with Michael Shrayer’s Electric Pencil from 1976 (Freiberger, 1982; see also Bergin, 2006a, pp. 33–35 for WordMaster from 1978 and EasyWriter from 1979). They spread on the back of mass-marketed micros and in turn served as one of the ‘killer applications’ that helped introduce the new hardware paradigm to the general public (together with games and spreadsheets). They quickly differentiated into a myriad of competing solutions on the growing range of personal and home computer systems, most of which are forgotten today (Bergin, 2006a, b, p. 44). Due to the limited resources of early PCs, the programs were text-based at first—like the popular WordStar (1978) and WordPerfect (1979) applications; but as computing powers increased, they gained the GUIs and WYSIWYG capabilities demonstrated by the Xerox Alto—most notably with Word for Mac (1985) and Word for Windows (1989). And although there were some experiments towards a ‘purely’ digital writing on and for the screen with systems like Storyspace and HyperCard (Bolter & Joyce, 1987; Williams, 1987), the imperative of printed paper would dominate word processing (along with desktop publishing pioneered by Aldus PageMaker from 1985 and Adobe’s PostScript and PDF technology) even after PCs had become networked through the WWW in the mid-1990s.

According to Bergin (2006a, b), the history of word processing for PCs unfolded in three overlapping stages: an initial phase of ‘origins’, beginning in the mid-1970s with early microcomputers like the MITS Altair 8800 and the very first rudimentary PC applications like Michael Shrayer’s Electric Pencil and John Draper’s Easy Writer; a second phase of ‘proliferation’, beginning at the start of the 1980s with the introduction of the IBM Personal Computer and more sophisticated word processors, most notably MicroPro’s WordStar and SSI’s WordPerfect; and a third phase of ‘consolidation’, beginning around 1990 with the rise of Microsoft Windows and the eventual monopoly of Microsoft Word for Windows.

The three phases of PC word processing described by Bergin coincide with major shifts in the ecology of microcomputer hardware and software: the first phase (ca. 1975–1980) was characterized by the initial diversity and mutual incompatibility of machines, reconciled only by the popularity of the CP/M operating system; the second phase (ca. 1981–1989) brought a massive standardization of technology through the homogenizing forces of the IBM PC hardware platform and the MS-DOS software environment; finally, the third phase (since ca. 1990) saw the breakthrough of the GUI paradigm for PCs and completed their standardization through the hegemony of Microsoft Windows. Thus, the evolution of word processing software followed the trend of the PC platform as a whole: from a variety of competing but incompatible products to a single, ‘universal’ solution; and from simpler, text-based products to an elaborate graphical system within a common GUI framework. From this perspective, the success of Microsoft Word can been seen not only as the result of Microsoft’s ruthless business practices but also as the culmination of a larger technological and commercial process of increasing standardization and integration in personal computing. While most essential word processing features had already been implemented by other programs in the mid-1980s, the addition of true WYSIWYG capability and the seamless interaction with the Windows framework was the unique factor that helped Microsoft Word conquer the market at the beginning of the 1990s. (On the Apple Macintosh, Word possessed WYSIWYG capability since 1985; Microsoft’s main competitor WordPerfect only got it more than a year after Word for Windows and never really played well with Windows).

In total, PC word processing differs from the earlier digital writing tools and systems from the 1960s and 1970s discussed above by four main facts:

  1. 1.

    It consists almost exclusively of commercial off-the-shelf products (with a few exceptions like OpenOffice or LibreOffice Writer).

  2. 2.

    While there was a very lively and diverse market for word processing applications in the beginning, the field has been monopolized by the de facto standard of Microsoft Word for Windows since the early 1990s.

  3. 3.

    Since the mid-1990s, word processing has stretched beyond narrow user groups and reached the general population (at least in so-called developed countries) where it has mostly replaced the typewriter.

  4. 4.

    Today, the scope of word processing covers almost any field of writing, from personal notetaking to the preparation of legal documents.

3 Functional Specifications

Word processing applications for PCs typically offer the following four sets of essential functions:

  1. 1.

    Editing of text (entering and deleting text, copy-pasting and search-replacing strings etc.).

  2. 2.

    Handling of documents (creating, saving, deleting files).

  3. 3.

    Formatting of text and documents (choosing different fonts, text sizes, paragraph alignments, page layouts etc.).

  4. 4.

    Displaying and printing of documents (with video screens and laser or inkjet printers, especially in WYSIWYG mode).

As shown in the previous section, the first and second set of functions are historically derived from the text editors used by computer programmers since the 1960s. The third set stems from the text formatting and document processing tools invented for time-sharing and minicomputer installations in the 1960s. Finally, the fourth set goes back to experimental computer systems like NLS and the Xerox Alto from the 1960s and 1970s.

While word processing applications are most commonly used by authors for composing their own texts, the four sets of functions actually address them in different roles: The first set treats the author as editor, the second set as secretary and the third and fourth set as typesetter and graphic designer. Addressing the author as a creative and a collaborative writer was not an integral part of word processing until the 1990s. More information on the corresponding technological functionality will be offered in the following chapter (Kruse & Rapp, “Word Processing Software: The Rise of MS Word”).