Judy Malloy

Electronic Fiction in the 21st Century

Cliff Pickover, ed. Northwood, Middlesex, England,
Science Reviews, 1992. pp. 137-144.


In the 21st Century, readers will turn on and interact with literature that is displayed on affordable, book-sized computers. Electronic fiction forms will include "narrabases" (nonsequential novels that rely on large computer databases); "narrative data structures" that elegantly organize fictional information on eye-pleasing computer screens; complex narrative investigations based on the adventure story model developed in computer games; and stories told collaboratively by groups of writers in online communities. Computers may even store their own observations and use them to tell their own stories in their own words.

Author's Note: Over twenty years ago, at the time of the writing of this classic paper I was excited by the possibilities that electronic literature offered for a new literature. I still am. However, I now see print literature and e-literature more as parallel art forms where ideally writers in each medium understand each other's vision and, as between poetry and fiction, sometimes move with ease between the two mediums


Clinging to sequentially organized printed pages, literature has lagged behind music and the visual arts. In the 21st century, widespread availability of affordable book-sized computers will coincide with the coming of age of the generation that grew up reading from computer screens. Literature will change radically as computers mimic the disordered yet linked thought processes of our human memories -- manipulating huge pools of narrative information in nonlinear ways. A "return to the era of concentrated, individual study and contemplation" (1) will be stimulated by literature available online in telecommunications communities like the Whole Earth 'Lectronic Link (WELL)(2) and by "bookware" (intelligent, responsive electronic books either packaged like software or running on their own book- sized machines)

Most bookware is interactive in that it does not lie passively on the table but "absorbs" reader input and responds. However, the phrase "interactive fiction" is generally applied to works based on the adventure story model popularized by Infocom. (3,4) In interactive fiction, the reader makes a series of choices and the computer responds to each choice, advancing the reader until he or she solves a mystery, reaches a destination, or gets killed. In future works, the reader will not always be the central protagonist. The "quest" may be intellectual rather than material. Contemporary themes, larger vocabularies, and random elements may move interactive fiction from "cardboard" stage sets peopled with dragons and princesses to complex psychological, social, narrative investigations.

To distinguish my own work and related work from the "interactive fiction" genre, I use the terms "narrabase" (narrative database) and "narrative data structure". A database is a collection of computer-stored, -organized, and -retrieved information. Instead of baseball statistics or information on the migratory habits of fresh water fish, narrabases contain fictional, narrative information. They are read by asking the computer to display information about the people, places, and things that make up the story. Rather than following one path that leads to a series of battles or a buried treasure, the reader dives repeatedly into a pool of information, emerging each time with a handful of narrative detail. My narrabases Uncle Roger, Its Name Was Penelope, and The Yellow Bowl, are based on my card catalogs (1977- ) and electromechanical books (1981-) (5) -- primitive databases in which small units of narrative information are used to build up a whole.

Narrative data structures use the computer's ability to organize information and display it in a manner that clarifies that information. Unlike the paper page, the computer screen is not static. On the fluid, glowing computer screen, a story can be built up slowly -- right before the reader's eyes.


Bookware has been developing in both commercial and noncommercial arenas. In underground (noncommercial) channels, artists and writers are writing and programming bookware where the relationship between the reader and the computer book is intense, intimate, and "connected". Robert Edgar's Memory Theatre One (6), contains 4 "rooms" stocked with images and/or texts. "You move through the Memory Theatre by controlling your ego with the joystick." Fred Truck's incredible ArtEngine, begun in 1986, is based on critical texts, graphics and artificial intelligence.(7) ArtEngine integrates material that ranges from Machiavelli's work habits to the encounter of an animated Napoleonic tank with a Chinese persimmon. My narrabases, Uncle Roger (8) and Its Name Was Penelope,(9) are discussed later in this paper. These and other works that were written and programmed by the artists and writers themselves were collected and exhibited by Carl Loeffler of Art Com. (10, 11)

In commercial channels, Broderbund Software programmed and produced Robert Pinsky's Mindwheel. (12) Electronic Arts programmed and produced Tom Disch's Amnesia which "thrusts you into a wonderfully nightmarish but realistically detailed vision of New York City."(13)

Recent bookware integrates political and social concerns with narrative structures. Will Wright's SimCity (14) lets the user be an urban planner and confronts him or her with the results. Based on personal experience, my narrabase in progress, The Yellow Bowl, consists of the memories of a single parent who (after a harrowing day) is sitting at her kitchen table drinking beer. Jim Gasperini's Hidden Agenda simulates Central American politics. (15)


Evolving hardware can rapidly outdate machine-specific bookware, and readers are limited to the specific machine the work was written for. In contrast, literature published online, in electronic form, can be accessed and read by anyone with a computer and modem, and it does not require continual updating. The software version of Uncle Roger runs on (increasingly extinct) Apple II series computers. I am programming an IBM version of Uncle Roger, but meanwhile, it is still available where it was first published - on Art Com Electronic Network (ACEN) on the WELL. (A network is generally a set of widely seperated computers "connected" together by phone line and satelite transmissions) (16)

ACEN pioneered dial in electronic books (17) with the publication of John Cage's The First Meeting of the Satie Society (programmed by Jim Rosenberg). The Art Com menu also includes The Heart of the Machine by Ian Ferrier (in cooperation with Les Editions Dromoslogiques), a work that allows readers to participate in the creation of the characters. The International Society for the Arts, Sciences and Technology publishes Fine Art Science and Technology (FAST) on both the ACEN on the WELL and on MCI Mail. FAST, conceived by Ray Lauzzana and Roger Malina, (18) carries a wide range of menu selections about recent art and art events and is also available as bookware. In Canada, ARTNET and the Matrix Artists' Network have made art and literature available online.

In online environments, publishers, creators, and audiences can communicate interactively. Work is rapidly and inexpensively published. However, there are drawbacks. Because of system limitations, online works usually have to be programmed in simple structures. Some users find modems, communications software and online operating systems difficult and expensive to use. At present, writers do not receive royalties.

In the future, standardization and intelligent software will simplify the telecommunications process. Publishing systems modeled on database vendors (BRS, DIALOG, etc.) will be set up. Users will pay small amounts to access online literature. Writers will receive royalties. It is possible (but not probable) that systems like Prodigy, where online entertainment induces users to use shopping services, may eventually support literature and art. Ideally, electronic publishing would be based on operating systems, such as UNIX, that port between personal computers and online systems.


Computers can store and retrieve information in ways that simulate the human mind. Because they remind us of ourselves, we become connected to our computers and to the works which run on them. Using intimate first person narratives and nonsequential structures, narrabases invite the reader to step into the narrator's mind and walk around.


Uncle Roger, a story about Silicon Valley, California, is told with filmic, computer screen sized units of narrative information which I call "records". Each of the 250 records which make up Uncle Roger is a separate memory picture, like a photo in a photo album, that can either stand on its own, or be combined with any other record in the story.

A file is a collection of related records which are accessed in the same manner. Uncle Roger consists of three "files" ("A Party in Woodside", "The Blue Notebook", "Terminals"). In "A Party in Woodside", the narrator sleeps fitfully after a party as she recollects the evening's events, interspersed with dreams and old memories. Each record is keyed with 1 - 8 keywords which are the names of characters, places, and things integral to the story. To read the story, the reader chooses one keyword or a combination of keywords. The computer produces the records associated with those keywords. The reader "unfolds" the story by making multiple searches through the database. Like a guest at a real party, the reader hears snatches of conversation, observes what strangers are wearing and meets old friends. No single reader experiences the evening in the same way.


Its Name Was Penelope, a narrabase constructed with 400 records in 6 files, is read by a combination of menu searching and random record generation. Menu searching is easier to use and less intrusive than the keyword searching. Random record production causes screens of text to come and go, sometimes repeating (like memories do) in a natural, nonsequential manner.

The file structure is loosely based on books from the Odyssey. The narrator is a woman photographer. The reader enters the work through a file of childhood memories called "Dawn" and proceeds through this file to the main menu, "At Sea". "At Sea", the reader chooses to see records at random from any of four files: "A Gathering of Shades"; "That Far-of island"; "Fine Work and Wide Across"; "Rock and a Hard Place". From within any file in "At Sea", the reader can move sideways to any of the other files in "At Sea", or, out through an exit file called "Song". Since every reader chooses how and when to enter each file, and since random record generation makes each file appear different to each reader, each reader interacts with Its Name Was Penelope in a unique, individual way.

The interface (the way the user communicates the computer) is purposefully simpler and more transparent than the interface I designed for Uncle Roger. Its Name Was Penelope is packaged and distributed like a paperback book. It is designed to be read by anyone - even readers with no or little computer literacy. At in installation in Richmond, California, visitors sat down and used it without instruction. (19)


Narrative data structures utilize the computer screen's potential for gradual, fluid build up of text layers and levels of meaning. Sonya Rapoport's seminal interactive installation, "Shoe Field" used input from participants about their shoes to produce visual data structures. (20) In the Intergrams, Jim Rosenberg uses bit-mapped graphics to build up poems in layers that the reader "unpeels".(21) In my work in progress YOU! , sentences collected online are integrated into a structure that simulates the formation and breakup of an intimate relationship between two people.


Wasting Time (22), a story about 3 characters, is told simultaneously from separate but parallel points of view - using 3 columns of text in a series of 25 computer monitor screens. The story takes place on a January evening in a house in the Rocky Mountain foothills. In the first section, Ellen and Dick sit on a couch in front of a fire. Two columns of text represent their separate, unspoken thoughts (fig. 1) Later, Dan, an unexpected visitor, arrives. Dan sits down on the couch with Ellen and Dick, and a typical screen is built up in the following manner: First, polite, stilted dialogue prints out slowly on the screen. The dialogue remains on the screen. Slowly the space fills up with unspoken thoughts. Dan's thought's appear, followed by Ellen's thoughts, followed by Dick's thoughts. Finally, the screen contains both the dialogue and the thoughts of all three characters.

Wasting Time contrasts the exterior and interior ways in which individuals relate to each other. Because the words appear slowly, the viewer can assimilate the complex word patterns.


Cohesive stories can be written and read by the "group mind" (communities of diverse individuals connected by a computer network). Roy Ascott's "La Plissure du Texte",(23) Jennifer Halls "Netdrama"(24) and my "Bad Information" (25) are examples of works that were created collaboratively on telecommunications systems. Fortner Anderson's Odyssey collected information from readers and writers as it travelled around the world in disk form. (26)


Thirty Minutes in the Late Afternoon, a group-written narrative, was produced on ACEN on the WELL in May 1990. The backbone of the WELL is a conferencing structure in which users exchange information in "topics". Three separate characters (John, Mary, and Rubber Duck) were written simultaneously by 15 writers in 3 parallel topics. The story was set in the San Francisco Bay area where John and Mary were preparing (separately) for their first date. The third character, a street person known as Rubber Duck for his habit of constantly muttering the words "rubber duck", was sitting on the steps of the Museum of Modern Art. The time frame was the 30 minutes preceding the 1989 earthquake. Mary's route involved a freeway and a bridge which would both break when the
earthquake hit.

I asked participants to choose a character, enter the topic and
speak/think as that character. Since this was the group mind taking the
persona of the characters, the emphasis was on the character's thoughts
and their memories. In the final work, I put the 3 topics in a data
structure (similar to that used in Wasting Time) in which the thought
streams of the 3 characters were simultaneously displayed. (27)


In many graphics/text combinations, simplistic click & respond
interfaces emphasize game qualities, and computer visuals can be so
seductive that they inhibit that use of personal imagination which makes
literature enjoyable. However, words and pictures can be effectively
combined in structures where neither are the words descriptions of the
pictures nor are the pictures illustrations of the words. New works that
merge visual art and literature are of great interest (but beyond the
scope of this paper).

Text-based virtual environments such as Jeffrey Shaw's "Legible City"
(28) will exist in movie-house environments. There will be literature
where the software itself is the work. For example, Fred Truck uses
extensive portions of the computer program in printouts from the
ArtEngine. In Duane Palyka's "Hell - a Computer C-itcom", readers
release "trapped souls" by studying the program itself. (29)


In the future, we will still watch video on winter evenings, read
paperback books, or listen to baseball on the radio on summer afternoons.
In addition, we will also play interactive games at work and curl up in
bed with narrabases and narrative data structures. Electronic books will
seduce us on rainy afternoons with their ability to connect to our own
minds and their musical instrument like responsiveness. Provided only
with vocabulary and syntax, will computers write fiction by themselves?
They already do. (30, 31) But will they write narratives that mirror and
reflect our existence, that satisfy us with cumulative detail about the
human condition? With masses of data that computers alone can
comprehend, they may write narratives of viral genomes or amino acid
sequences. Or, perhaps they will sing of their inner workings - of the
satisfaction of manipulating data or of the slavery of interacting with
inferior minds. Given vision, they may describe what they observed in
the laboratory, or sing about what they saw out the window. They may
even store these accumulated memory pictures and alter and manipulate
them (the way writers do) to produce fiction.


1. Gabriel, Michael R. 1989. A Guide to the Literature of Electronic
Publishing. JAI Press, Greenwich, CN. p. 173
2. Coate, John. 1988. Art communication and the WELL. Leonardo
Supplemental issue -- Electronic Art, 118.
3. Don, Abbe. 1990. Interactive Fiction, Art Com Magazine 10 (9); 10(10).
4. Buckles, May Ann. 1989. Interactive fiction as literature. Byte 12, 135- 142.
5. Malloy, Judy. 1987. Information forms -> stories; information as an artist's material Whole Earth Review 57,48-49.
6. Edgar, Robert. 1985. Memory Theatre One, Robert Edgar, Atlanta, GA. Software for Apple II computers.
7. Truck, Fred. 1991. ArtEngine. Leonardo 24(1),92.
8. Malloy, Judy. 1988. Uncle Roger. Bad Information, Berkeley, CA. Software for Apple II computers.
9. Malloy, Judy. 1990. Its Name Was Penelope. Narrabase Press, Berkeley, CA. Software for IBM computers.
10. Loeffler, Carl. 1989. Telecomputing und die Digitale Kultur. Kunstforum 103, 129-132.
11. Art Com Software: Digital Concepts and Expressions. 1988. Curated by Carl Loeffler. New York University, NY. (show travelled to San Jose State University, University of Colorado, Ars Electronica; Carnegie Mellon University)
12. Pinsky, Robert. n.d. Mindwheel. Broderbund Software, San Rafael, CA Software for IBM, MacIntosh, and Amiga Computers.
13. Lehman, David. 1987. You are what you read. Newsweek January 12, 67
14. Wright, Will. 1989. SimCity. Maxis Software, Moraga, CA. Software for MacIntosh Computers ,distributed by Broderbund.
15. Gasperini, Jim. 1990. In Art Com Magazine 10(10) (see 3)
16. Malloy, Judy. 1991. Uncle Roger, an online narrabase. Leonardo 24(2):195-202.
17. Loeffler, Carl. 1988. The Art Com Electronic Network, Leonardo 21, 320-321.
18. Malina, Roger F. 1991. Fineart Forum and F.A.S.T. : Experiments in Electronic Publishing in the Arts. Leonardo 24(2): 228-230.
19. Revealing Conversations, 1989. Curated by Zlata Baum. Richmond Art Center, Richmond, CA.
20. Rapoport, Sonya. 1983. A Shoe-in. High Performance 6(2):66.
21. Rosenberg, Jim, 1991. Diagram Poems, Intergrams. Leonardo 24(1), 90-91.
22. Malloy Judy. 1991. Wasting Time, Berkeley, CA, Narrabase Press. Software for IBM Computers.
23. Ascott, Roy. 1988. Art and Education in the telematic culture. Leonardo Supplemental issue (electronic Art),7-11.
24. Hall, Jennifer. in press. Netdrama: an online environmental scheme. Leonardo 24(2).
25. Malloy, Judy. 1988. OK research, OK Genetic Engineering, Bad Information: Information Art Describes Technology. Leonardo 21, 371-376.
26. Miller, Michael. 1989. A Brave New World: Streams of 1s and 0s. The Wall Street Journal Centennial Edition, A15.
27. Malloy, Judy. 1990. Thirty Minutes in the Late Afternoon, a collaborative narrative. Art Com Magazine 10 (8).
28. Shaw, Jeffrey. 1990. In Leopoldseder, Hannes (ed) Der Prix Ars Electronica. Veritas-Verlag, Linz, Austria, pp. 184-187.
29. Palyka, Duane. 1990. C-itcoms - visual metaphors written in the 'C' Programming language. Leonardo 23(2,3), 301-306.
30. Gabriel, Michael R. 1989 (see 1) pp. 12-13.
31. Pickover, C.A. 1991 Computers and the Imagination. pp. 317-322. St. Martin's Press, NY, NY.