probing worlds made of words

Sometimes I'm not sure about the merits of the term "electronic literature." I've used it loads of times, including spending some labor in this very blog working on an unfinished taxonomy of "forms of electronic literature." Although even perusing that list will reveal some instances where I refer to certain types of electronic literature as really being more akin to "toys" or "films" then as "literature" per se.

I think about this a lot in relation to the term "interactive fiction" or "interactive narrative," a category which encapsulates what used to be known as the more lowly "text adventure." "Interactive fiction" (IF) seems to have become the commonly-accepted term for these sorts of creative works, and I've stated some of the things I like about the term on record, but I wonder sometimes if it doesn't distort the way people think about the end product. Specifically, I wonder if it's useful to assess IF using the critical tools one would use to assess a piece of literature: I wonder if IF wouldn't be better assessed using the critical tools one would use to assess a video game.

This came up for me recently when I was reading the transcript of "Interactive Narratives Reconsidered," an interesting speech that Ernest W. Adams delivered at last year's Game Developers Conference.

Adams' speech mostly attempts to answer the question of "how can we make interactive narratives better as narratives?" In order to establish the need for improvement in this regard, Adams points out a set of "key problems" that make it "difficult to create interactive narratives."

For instance, "The Problem of Amnesia": "What do we do about the fact that story characters understand the world they live in, but the player is amnesiac about that world? Why does the player have to spend time at the beginning of every game exploring what is supposed to be his own natural environment?"

I think it's a valid point that this maybe doesn't make for a very realistic story, and it may not be compelling as literature, but it's never bothered me very much in the text adventures I've played (or, if you prefer, "in the pieces of interactive fiction I've read"). It's never registered as a "problem," exactly. And in trying to think about why that is, it occurred to me that a big part of what's pleasurable about a video game is this process of exploring and testing the environment. In Steven Johnson's book Everything Bad Is Good For You, he refers to this process as "probing," a concept originally theorized by social scientist James Paul Gee.

Gee is a pretty smart guy (Stanford Ph.D.) who has some things to say on the topic of games as narratives (from this interview):

"Stories in video games work very differently than do stories in books or movies, and we really don't understand well how they work, yet, because we keep treating games like movies. In books and movies, the story is 'top-down,' someone else has made it and you discover it in the order and at the pace the designer has determined. In games, stories are 'bottom-up.' The player picks up bits and pieces sometimes in an order and at a pace determined by the player."

Well, exactly. The sort of "mimetic gap" between the way a player explores a game and the way a heroic protagonist would realistically behave in fiction, seems to me to be a key part of what makes a game a game.

Apply Adams' "Problem of Amnesia" to a classic video game like, say, Defender. I've been playing Defender for approximately 26 years now and I'm still terribly bad at it: the fast gameplay and complicated controls make it a game of near-infernal difficulty. But that's the fun of it. Treating this game as narrative would be laughable: you wouldn't ask "why does this character need to spend valuable time trying to figure out the controls of his own ship?" or "why did this planet choose such an inept defender, thereby insuring their immanent doom?" (It's true that the planet in Defender only has ten humans on it, so their talent pool is pretty limited.)

Similar is Adams' "Problem of Internal Consistency": "What if the player is controlling Superman as his avatar, but wants to do something very unlike Superman: killing people at random, for example?"

This wouldn't make for a very canonical Superman story, agreed, but attempting to do something like that seems to me to be a fundamental part of the process of playing a video game. Part of the fun of a game is figuring out what constitutes its internal consistency, which means that sometimes you're going to do things which don't make consistent or realistic sense were we to be watching the thing as a narrative. Play Shadow of the Colossus and it's only a matter of time before you make the protagonist leap to his doom off a staggeringly high cliff. Taken as story, this makes no sense: why would a protagonist so seemingly driven to complete his goal suddenly opt for suicide? Taken as an instance of "probing"—testing the parameters of a game world—it makes perfect sense.

I just don't think a text adventure / piece of interactive fiction should be held to different standards just because it's made of words and not polygons.

Labels: electronic_literature, game_commentary, narrative

Tuesday, August 08, 2006
11:30 AM
0 comments

twisty little passages II

Early in Nick Montfort's Twisty Little Passages : An Approach to Interactive Fiction, Montfort makes a convincing case for using the term "interactive fiction" to describe the sort of electronic literature that he's writing about. I like the designator, partially because, as Montfort points out, it enjoys widespread usage, but also because it can be precisely located in a hierarchy of related descriptors. Specifically, it is a little bit more inclusive than the term "text adventure" (not all pieces of interactive fiction need to be "adventures") and a little bit less inclusive than my term command-line literature (not all pieces of command-line literature need to be "fictional").

(An aside: for those of you who aren't really sure what interactive fiction is, this page might give you a basic handle on the form.)

So. Within the first ten pages of Twisty Little Passages, Montfort remarks on the need for "a book-sized resource on interactive fiction's history and implications—one that considers how the form came into being and how it developed through the decades, with basic theoretical discussions of the nature of the form and at least an introductory critical discussion of important works," and it is apparent that the rest of the book intends to fill that need. To quibble with the book's subtitle, one could argue that the different strands in that list do not really constitute a single "approach," but rather several different approaches: there's really enough there to fill a couple of different books. By attempting to tackle each of them in a single concise volume, a certain scantiness ensues (I had no trouble completing the book in a day), but Montfort deserves credit for ambitiously staking out the territory: other scholars of electronic literature will undoubtedly see this book as a valuable starting point to branch off from.

The most successful chapters, to me, are the ones that consider "how the form came into being and how it developed through the decades." The history of Zork and Adventure's development is especially interesting reading, as is the overview of the contemporary IF scene, which has apparently thrived as a non-commercial subculture in the years following the decline of Infocom and other commercial IF publishers. Montfort's critical overview of the major IF works (of the both the commercial and post-commercial era) is pretty condensed—only the most important works get more than a page or two—but valuable nevertheless: I'm hard-pressed to say that I'd trade it for a deeper read into a smaller handful of works. That can come later.

The weaker chapters are the ones that attempt a theory of the form. The first chapter does some decent work establishing a useful terminology with which to discuss IF works: distinguishing between replies and reports, for instance, or distinguishing between which commands are digetic and which are extradigetic. This material, however, is dispensed with in under ten pages, and is forced to share space in this first chapter with the standard "what is interactive fiction?" boilerplate.

The second chapter, probably the book's weakest, unconvincingly attempts to situate the text adventure within the literary tradition of the riddle. Some of the parallels that Montfort attempts to draw have numerous exceptions: for instance, although it is true that riddles are "presented for solution," it is less true that all interactive fiction can (or should) be thought of as doing the same: for instance, notice that many of the IF works available through Adam Cadre's IF page are said to contain "almost no gamelike elements." ("If stuck, just keep exploring," Cadre writes of his latest work, and, in the release notes, he writes "even if you get to an ending, you may have only seen a small fraction of what's possible," neither of which seem like statements that usefully apply to any riddle I know of.)

I take less issue (at least initially) with Montfort's statement that interactive fiction creates a systematic world, but again, the parallel flags for me: is it accurate to say that a riddle also creates this sort of world? Perhaps technically, but the experience of solving a riddle feels to me substantially different from the far more immersive and ludic experience of exploring the world of a work of interactive fiction.

I think Montfort is more on the mark when he touches on the idea of IF as a "literary machine" or what Espen Aarseth would call "ergodic literature." The literary tradition there dates back at least as far as that of the riddle: the I Ching is commonly cited (including by Montfort) as a "literary machine" that dates back to antiquity.

Quibbles aside, I'd highly recommend this book to anyone studying electronic writing, and I even think that most of it is accessible enough to be of interest to people who remember the old Infocom games fondly and might have an interest in seeing what's new in the field.

Labels: book_commentary, electronic_literature, interactive_fiction

Tuesday, January 13, 2004
5:03 PM
0 comments

I haven't added to my "forms of electronic literature" list since July. At this rate it'll be a year-long project. Actually maybe not, as I only have one more big category left to deal with after this one.

Command-Line Literature

Command-line literature is a textual system that you navigate by entering prompts at regular intervals. Usually you and the system "take turns," creating a kind of dialogic effect. These sorts of systems were popularized by a series of interactive games produced by companies like Infocom, of which Zork is the best well-known.

Many pieces of command-line literature, including the Infocom works, are essentially hypertextual: that is to say, they are primarily composed of large units of pre-existing text (lexia) that the user navigates their way through. One key difference, however, is the nature of the user interface: most widely-known conceptions of hypertext take the link as the primary means of naviagation, rather than the address-driven mode that characterizes command-line literature.

An additional difference is that most command-line works use at least some generative element. No writer can possibly compose responses to cover every possible user input, and so generally certain computer subroutines are employed to handle statements that the system reads as nonsensical, yielding responses like the familiar "I see no X here."

When the generative element is foregrounded, the hypertextual element is correspondingly reduced, and the experience becomes less like navigating a text (less "literary," one could argue) and more like having a conversation. The key text to think of here is Joseph Weizenbaum's notorious ELIZA, and the generation of chatbots born in her wake.

Forerunners of command-line literature: the Turing Test, early AI research, early command-line-driven computer langauges, Will Crowther's ADVENT.

Labels: electronic_literature, hypertext

Saturday, October 19, 2002
11:43 AM
0 comments

I just renewed the Imaginary Year domain for five years.

I must remember to apply for a Community Arts Assistance Program Grant from the City of Chicago Department of Cultural Affairs, to see if they will help fund the project. (Applications are available in October, due in December.)

Related piece: No Visible Means of Support, by Sue Thomas, which I link to, although the revelation that most new media writers aren't making any money shouldn't be news to anyone, and Talan Memmott's notion of "licensing" doesn't sit well with me at all.

I'm happy to give away the work for free, but it would also be great if it could generate some income, somehow. I wonder if there's a market out there for Imaginary Year T-shirts, or coffee mugs.

Labels: electronic_literature, personal

Monday, September 09, 2002
7:34 PM
0 comments

Hypertext & Hypermedia.

Okay, hypertext isn't exactly a "new trend," but it is a form that's key to thinking about the other types of electronic writing, so it's worth discussing here if only to round out my taxonomy. To many of you this will be old hat.

A hypertext is a text broken up into several chunks. Traditionally the various "chunks" are textual in nature, although they can be images or sounds as well (only then you'd properly use the term hypermedia). George Landow, probably hypertext's most important academic booster, uses Roland Barthes' term lexia to describe these chunks, although they're also referred to frequently as "nodes," and the Deleuze and Guattari fans out there might prefer "rhizomes."

These chunks can be traversed in a variety of ways, and the traversing mechanisms are generally refered to as links. The various connections between the nodes and links can be represented as a web. If you're reading this page, all of these terms are no doubt familiar to you, and one could argue that the World Wide Web is in fact a single gigantic multi-authored hypermedia work.

To qualify as hypertext, there should ideally be a degree of interactivity involved, and the work should ideally have a certain nonlinear dimension. After all, one could argue that a traditional novel, broken into chapters, is organized into lexia, but since those lexia are designed to be read in a linear manner (links are notably absent) any interactive element is essentially negligible.

Some people treat hypertext and electronic literature as essentially interchangeable. Early attention given to hypertextual electronic authoring tools such as StorySpace or HyperCard helped to cement this conception. But I think it's inaccurate, and certainly some of the other types of electronic writing that I've outlined in the past, such as computer-generated narrative or language films lack one or more of the key elements of hypertext.

Examples of hypertext abound, but well-known hypertext narratives include Michael Joyce's afternoon, a story, Shelly Jackson's Patchwork Girl, and Stuart Moulthrop's Victory Garden. (All three of these were written using Storyspace, and require a CD-ROM to experience, although an abridged version of Victory Garden has been adapted for the Web, here).

Forerunners include: Choose Your Own Adventure books; Robert Grenier's Sentences; B. S. Johnson's The Unfortunates; Robert Coover's nonlinear short stories (such as "The Babysitter"); Cortazar's Hopscotch; Borges' "Garden of Forking Paths;" Ted Nelson's Literary Machines, Dream Machines, and Project Xanadu; Vannevar Bush's "As We May Think"; tools for non-linear access of traditional texts, such as indexes, page numbers, concordances; early ergodic or non-linear texts such as the I Ching.

Labels: electronic_literature, hypertext

Wednesday, July 03, 2002
3:35 PM
0 comments

Temporal writing.

The ability to change Web texts through time (to create "time-shifted" media, to borrow a phrase from this essay) is one of the fundamental, unique properties of Web-writing, and thus it should be a key consideration for current practicioners of electronic writing. And yet I find myself surprised that so few of the current crop of electronic writers are producing work that makes use of the temporal dimension. I'm also surprised that none of the web journals that publish electronic writing make allowances for time-shifted work (to the best of my knowledge).

There's certainly a huge populist explosion of this type of writing, however. Tens of thousands of people use content management systems like Blogger and LiveJournal: these systems essentially automate the organization of material into chronological streams. I wouldn't generally consider most weblogs or LiveJournals to be "electronic literature," but there's a contiunuum here, with the hyperlinked brevity of Robot Wisdom on one end. On the other end we find the literate, essay-like entries of, say, Michael Barrish's Oblivio, Steve Cook's Snarkout, or Paul Ford's FTrain (this last also includes things that apear to be personal essays but which are actually fiction). Are these sites still weblogs? Arguable, but the important point is that these sites make use of temporality: the sites are not static, and this makes us read them differently from how we would read, say, a book of short nonfiction pieces.

When applied to fiction, temporal writing often takes the form of serialized narrative, such as Phantomnation or my own Imaginary Year. A key related text is Michael Stutz's piece, "Episodic Writing."

Forerunners of temporal writing: content-management applications, syndicated newspaper columnists, periodical narratives (such as comic books), Thomas Wolfe's episodic novels, film serials, Victorian-era serializations / series novels.

Labels: electronic_literature, narrative, time, weblogs

Tuesday, June 11, 2002
1:03 PM
0 comments

I've been trying to decide how to classify this Storyland website, and I've decided that it falls neatly at the intersection of two different categories of electronic literature.

It is part computer-generated narrative, of the sort assembled by programs like James Meehan's Tale-Spin. But its frivolousness aligns it with sites like They Have Blogs!, or the Random Bar Joke Generator. I think of these sorts of sites language toys: they're entertaining enough, but they don't, to my mind, produce anything that takes on the status of literature. (The possible exception being the sites that are dedicated to the William Burroughs' cut-up technique, a technique which arguably possesses both a cultural critique and an occult potency.)

Forerunners of computer-generated narrative: Vladimir Propp's Morphology of the Folktale, Georges Polti's The 36 Dramatic Situations, S. Klein's plot generating software (1973).

Forerunners of language toys: Mad Libs, random number generators, Raymond Queneau's 100,000,000,000,000 Sonnets, Tristan Tzara's Dada cut-ups, surrealist games.

Labels: electronic_literature

Monday, May 20, 2002
11:11 PM
0 comments

OK. I'm going to attempt, over the next few weeks, to get a grip on some of the current trends in electronic literature, in part by reading my way through the index of work presented at the Electronic Literature State of the Art Symposium.

The first grouping that I can clearly identify encompasses short animations, generally designed in Flash, that use the written word as a central component, and have little or no interactive element (beyond possibly clicking "start" or performing some other basic navigational tasks, like scrolling). I'd call these "kinetic poems," except kinetic poems is too broad a designation: for instance, a kinetic poem can be interactive or non-interactive. Since the language-based content is the only thing that significantly distinguishes these from other short experimental web-based films, I'm going to call them language films.

Examples include: Miekal And's SEEDSIGNS for Philadelpho, or The Dreamlife of Letters by Brian Kim Stefans.

Forerunners: concrete poetry, language poetry, experimental film, Sesame Street animation segments.

More to come.

Labels: electronic_literature

Tuesday, April 16, 2002
4:32 PM
0 comments

I really should consider going to this conference on New Media Poetry this fall in Iowa City.

There are often conferences that I'd like to attend, but can't, because of the expense involved in travelling halfway across the country. The most recent one I missed was the Electronic Literature Organization's State of the Art Symposium, out in LA, where I in fact presented work even though I wasn't there.

But Iowa City is closer, just a few hours away by car. Of course, there's still the expense of renting a car for a few days, which is probably about the same as the cost of airline tickets...

Labels: academia, electronic_literature

Monday, April 15, 2002
10:46 AM
0 comments

I'm currently reading Sven Lindqvist's History of Bombing.

The subject matter is obviously relevant, but it is the book's form that fascinates me the most. The book is organized into 399 fragments, organized roughly chronologically, but within that chronology, several different narrative threads exist: you are not intended to read the book in a straight-through linear fashion, but rather to jump from fragment to fragment, depending on which narrative chain you are following. (The book opens with twenty-two starting points, which you can choose from freely.)

The book functions like a more structured version of Cortazar's Hopscotch. Or, more generally, it works as another good example of "ergodic literature," where "nontrivial effort is required to allow the reader to traverse the text," a category which also applies to a great deal of electronic writing. It not entirely unlikely that part of the reason I respond so favorably to the idea of separate narratives within a chronological superstructure is because it is similar to my own project, Imaginary Year.

Lindqvist describes his book as "a labyrinth with twenty-two entrances and no exit."

Labels: book_commentary, electronic_literature, writing

Monday, March 25, 2002
12:40 PM
0 comments