home | intro | biography | essay | view | | |

Word limits

Digitisation — the idea, and practice, of converting all media into numeric information — has been a great media equaliser. Digitised media is infinitely ‘translatable’; it can be copied, repackaged and reprogrammed (Mitchell 2003 p.140). Different media manifestations — text, audio, moving image — have become the (enjoyable) surface effects of the sea of data silently accumulating beyond our senses, a textual ontology that is largely unacknowledged and uncritiqued. Instead of existing as a sui generis media artefact, we now understand a media product as a more or less contingent endpoint of a process that is ultimately comprised of data manipulation.

Digital media aims to produce itself as a smooth unfolding experience; an aesthetic that privileges a seamless flow of image and audio dictates most of our art and mainstream media.[1] The fact that text ultimately enables digital media is thus usually obscured. Furthermore, the right of text to exist within the surface play of such media seems questionable. Text disrupts. Various infelicities may occur when it is included. It requires a different type of engagement; a level of concentration and interpretation that is not necessary for the consumption of audio and visual media. Digital media and digital art is more and more filmic, increasingly dominated by the aesthetics of collage and montage (Mirzeoff 1999 p.15).

In general, the digital era seems to treat text as a poor and rather unsophisticated cousin. Among the wealthy and the young — those who are comfortable with this convergent media culture — there appear to be few exceptions to this trend in highly produced digital media and artistic consumption. The high culture lure of the original (Benjamin 1935) seems to have been usurped by an aesthetic of greater and greater sensual immediacy. The increasing ability of the image to represent the real is closely associated with this impetus (Mirzeoff 1999, p.15)[2]. While all media (and particularly digital media) enjoys a tenuous relation to the real, some media can ‘get away with’ its facsimile status much better than text can. Text cannot help but foreground its status as a mediator, and as long as we are obsessed with an aesthetic of greater and greater sensual immediacy, text seems unable to compete.

If text is to remain part of popular culture, how should it respond to the digital ‘mediascape’ (Appadurai 1996 pp.35–36)? The aesthetic disruptions caused by combining text with other media are compounded by aspects of textual media that are unresponsive to digitisation. There is a fundamental level at which text must be presented linearly. Random collections of letters such as ‘ksjhgkajhgka…’ fails to be interesting quite quickly; what interest it has is as much visual as it is textual.

The publishing industry’s faith in text as a viable finished product seems at least under attack. Perhaps text will become the medium used when budgetary or time constraints prevent more fully worked pieces. Its usefulness as a tool is not threatened — for example, in scriptwriting, email, reports etc, but as an aesthetic object it may be.

Digital media is malleable because it can always be translated into it core data. However, there exists a secondary level of malleability. Digital media is ‘chunked’ into items of singular significance and stored in some sort of media database (Manovich 2001 Chapter 5). The items within the database can be combined and recombined in infinite ways. Given that the database offers this second level of malleability, good ways to access and organise digital media become very important. Three major approaches are:

  • interface (design, image) — text and other media is organised via a visual interface which allows users to predict, to a greater or lesser degree, the content they will access if they follow particular links. Hypertext — media in the database are linked to create a hyperlinear (Januzzi and Smith 2000 pp.29–32) path through the media — that is, users create their own sequence among innumerable possible linked paths. algorithms (programming) — text and other chunked, databased media can be organised according to any number of creative programmatic rules, triggered according to user behaviour or some other principle, for example time lapse.

Digital artefacts often combine these three approaches to organising and accessing digital media. The result is what Stuart Moulthrop has called a paranoid textuality, in which all media is inter-connected. All that remains for a user to do is uncover those connections (Moulthrop c.2001).

Writers respond to issues of access in various ways:

  • since text is ‘triggered’ by accessing specific chunks in a database, writers are obliged to create an appropriate interactive architecture. There are various levels of interactivity, the lowest being the act of interpretation, the highest being the ability of the consumer to produce the media they are consuming[3]. Issues surrounding a reader’s power over a text means that writers must adopt a quasi-political position on their relationship to their own texts. since text viewed on any type of monitor (not just computer but also mobile phone) must be visually organised, it must collaborate with image, and perhaps audio. It thus finds itself existing on a continuum with visual media such as computer games, and must learn to compete and integrate. Writers must develop visual skills or collaborate. since databases are increasingly mobile and decentralised, writers must create genres that respond to these environments.

A series of aesthetic conundrums
Various aesthetic issues arise in response to issues of access and organization, not the least of which surrounds the issue of narrative. According to Ricoeur:

…speculation on time is an inconclusive rumination to which narrative activity alone can respond
(Ricoeur 1984, p.6).

Many have argued for the pivotal nature of narrative to human thought: we are temporal beings, and narrative responds to that. However, the way we consume narrative is influenced by available media technology. Bernstein suggests that the way we write narrative has sometimes been technologically determined (Bernstein 2001). Many have commented on the difficulty of combining hypertext and interactivity with narrative. This may be a slight misunderstanding. There is no reason why chunks of media cannot be linked together hypertextually to create a perfectly good linear narrative. However, the point of doing such a thing seems limited.

Narrative becomes harder to organise when it is accessed via an interface that provides many options (either hypertextually or algorithmically). Principles of AI added to a narrative engine (arguably some computer games achieve this) may mean that narrative outcomes become completely unpredictable. It is also harder to control when several writers (or, indeed, players) are collaborating on the text, a type of writing facilitated by distributed networks. Narrative is facilitated by an author with a grand plan; if you problematise the author by distributing authorial control in some way or another, then you problematise the production of narrative itself.

As Ricoeur suggests, problematising narrative may also problematise our intuitive concept of time. This is interesting when we consider those critics such as Castalls who have pondered the effects of the destruction of time via world-wide distributed networking[4]. The fate of digital narrative might indeed be reflected in the global, networked lives that our technology seems to inexorably march the First World towards.

Intuitively, I have always been more comfortable creating poetry rather than narrative in digitised and databased environments. Two aesthetics result from these structures — repetition and transience — which seem to afford poetry more than narrative.

A standard poetic technique is repetition. Repetition is also an effect of the iteration and re-iteration of databased media items (Darley pp.67–68) which are organised by algorithm, hyperlink and interface. This means of access results in the likelihood that users experience chunks from the media database more than once. Media is combined and recombined.

Chunks of databased media is often experienced as transient phenomena. There is an impetus to always move on with digital media. Often generating a new chunk is more interesting than actually engaging with the specific content in a specific chunk (a mode of interaction decried by critics of the postmodern such as Baudrillard and Jameson, see Darley 2000 pp.69–70). This forward momentum which drives our computer gameplay also infects our interaction with media on the Web or any other highly interactive computer-based medium. This again suits the mainly short and condensed style of poetry.

The aesthetics of transience is compounded in networked media. Media can never been experienced in total isolation on the network. This includes mobile phone networks — messages, images, and increasingly mobile phone videos are forwarded from user to user, possibly with new edits en route. Such work has no end-point or destination, like the communities of users that employ them. This is a quotidian artform; it is not conceived narratively. It is always in transit. It is certainly never finished — or if it is, a conclusion is reached only when circulation is exhausted. Types of networked text, such as joke email, enjoy this mercurial flow through the network.

In the second half of the 1990s, creative writers started exploring the peculiar affordances of digital and networked textuality. Experimental interactive and hypertextual combinations of text with image and audio resulted in complex interactive and time-based media, of varying degrees of potential recombination. Many early explorations of these structures, such as using Storyspace software (which are not actually published online)[5] explore the problematic of narrative and interactivity and remain outside the mainstream. This problematic has been expressed by Ryan:

...narrative coherence is impossible to maintain in a truly complex system of links. We need therefore simpler structures, structures with fewer branches and fewer decision points, so that every path can be individually designed by the author. Once the user has made a choice, the narrative should be able to roll by itself for an extended period of time: otherwise, the system would lead to a combinatory explosion - or fall back into randomness, the deathbed of narrative coherence.
(Ryan 2003 p.11)

This is certainly one approach to the problem of narrative, and it has been used to good effect in online documentaries such as Homeless (c.2002). However, there is another approach to crafting narrative experiences from databased media and interactive environments. It requires a more sophisticated use of rule implementation. In The Princess Murderer (2003), a collaboration with Deena Larsen, we attempted to combine narrative principles with the interactivity of the computer game. In Machine Corporation (2002), a collaboration with Dane, we interwove narrative elements within a quiz functionality. Both used lots of algorithmic rules to determine the way in which the narrative unfolded, as a result of user choice. Both these works borrowed heavily from game principles, which reconfigure the concept of narrative (Ryan, 2001). In particular, we tried to establish environments which draw upon what Ryan has called in ontological mode:

In the ontological mode … the decisions of the user send the history of the virtual world on different forking paths.
(Ryan 2001 p.8)

Thus, we hope that ‘the pleasure of play and the pleasure of reading are … combined’ (Fournel, p.182).

I have found that narrative has become increasingly possible for me as my programming skills have improved. The ability to establish complex rules for what and when users can read gives us the ability to create nonlinear structures which nevertheless permit plot progression. Hypertext, in the narrow sense that I use the term, can not create these interesting structures. It must be combined with algorithmic interactivity to enable interesting and permutating linking.

Generative poetry
As can probably be inferred, my true love remains poetry. My Oulipan experiments explored the principle of random recombination of textual media, exemplified by Tristan Tzara (1920), Raymond Queneau (1961) and others. Florian Cramer (1996–2000) used this principle to program texts for delivery via the computer and Web. My use of the principle expanded the effects of the recombinatory algorithm to visual and audio elements as well as text. Using software called Director, I have attempted to create environments that would work on a variety of aesthetic levels. In my Oulipan works, the application of aesthetic principles like repetition and transience foregrounds the role of the algorithmic rules in text assembly and creates a fugue type of experience.

Writing text for random recombination is very challenging, particularly if, as with me, you are less interested in the surreal impetus of Tzara and William Burroughs(1978). In concatenation I wanted to address a real issue — interracial conflict and the politics of despair. I tried to use the recombination to suggest that whatever side of a particular conflict you were on, the issues were the same — trauma, homelessness, injury, confusion. Of course, I had to set up many rules to constrain the recombination. Total freedom would have resulted in the ‘ksjhgkajhgka’ mentioned above.

In semtexts I wanted to set up an engine which would recombine meaningful ‘semes’ (with apologies to Richard Dawkins[6]) — what I am calling meaning-laden, or at least suggestive syllables — in unexpected ways. I wanted to create a Joycean word-engine which would take the stress out of neologism manufacture. Hopefully unexpected combinations of syllables prompt users to new concepts.

The politics of text
As a result of digitisation and networking, many collaborative, opportunistic and interactive textual practices have developed. Textual production and communication is now a mobile phenomenon, and this has implications not only for form but for content. For example, SMS text does not require geographical specificity. As a result, a lot of SMS text is about location — ‘In mshpt, where r u?’ The swarming mentality of SMS-enabled teenagers (Mitchell 2003, p.32), and approximeetings (McCamish 2004 p.3; Mitchell 2003 p.157) results from unshackling textual communication from location.

The developing phenomenon of moblogging, which enables people to use their mobile phones to publish text, images and video direct to their internet website, opens further directions in self-publishing, combining the joy of permanence with the mobility of the network. Mitchell believes these developments constitute a ‘fundamental shift in subjectivity’ in which we no longer have a fixed identity (Mitchell 2003 p.62). Bernstein envisages hypertexts which are ‘extensible and recombinant’; they are modular and potentially ever-expanding, diffuse and decentralised. The reader ‘will always want to do things that nobody (and no computer) could anticipate’ (Bernstein 2001)[7]. How hypertextual and algorithmic structures will play themselves out on the networked mobile devices of the near future is certainly an interesting field for the imagination.

Text is circulating in different ways. It is decentralized, informal and personal. The centre truly is not holding, and when it tries to reassert itself — for example, by republishing texts that originally were only available on the web, strange hybrids result which do not wholly make sense (for example Graham and Burton, 2004). Text has also become unstable. Not only is it malleable, but various forms of writings (styles/signs) and creoles exist side-by-side.

 Who is your audience?
Issues of copyright aside, networking and digitisation facilitates the circulation and reiteration of all media, including text. While mainstream, big-budget media productions seem to sideline text, this endless P2P circulation of text confirms its pivotal position as a communications medium. However, its very mundane-ness and ubiquity may well make its value harder to see: while we are all being extremely prolific, there are no longer authoritative voices or styles of writing to give value to our SMS messages. Communities of writers share text on a sometimes minute-to-minute basis, but they are more or less closed communities from which ‘art’ rarely escapes. Furthermore, questions of audience and reception fade when capitalistic incentives for writing have been taken out of the equation, so there is an issue about who would be defining such activity as ‘art’ in the first place.

The politics of such decentralised, peer-to-peer communications has not escaped the notice of professional media companies. Increasingly fearful of being left out of the informal communications loop, mainstream media is attempting to muscle in on networked media making. Big Brother is indicative of a paradigm shift for mainstream media production which trades on ‘a commonplace of postmodern mass media that we are all insiders now’ (Mirzeoff 1999 p.99); that is, we can all play producer (although there are clear limitations to what mainstream media corporations think it is appropriate for us to produce). Another example is Ninemsn.com, whose business strategy is aimed at becoming the arbiters of P2P communication rather than the kings of content (AustralianIT 2004).

There are still limits to our media making. Perhaps the future of text is circumscribed by the 160-character limit of my mobile phone: poetry rather than the ‘great convergence novel’. This ‘poetry’ is one of praxis. It facilitates fluid, democratic, nomadic activities in public and private space (Mitchell 2003 p.159) including democratic protests (CNN 2004; Rheingold 2002)[8].

There would appear to be a high level of play which motivates a teenager to send 50 SMS messages every day and develop an SMS creole to rival the graffiti on our subway walls. Written text is changing to suit the medium, but the fundamental idea of text as a creative medium has not. While it may seem that utility has replaced art, I wonder whether, in our children’s bedrooms, they might be redefining a mobile version of art for art’s sake in a technologically informed reprise of early romanticism (Williams 1963 chapter 2 p.169).

Perhaps text will change its status; there will be lots of it, but no-one publishing commercially. Instead, text will be the underground artform of teenage bedrooms and surreptitious classroom messaging. The aesthetics of transience will dominate our engagement with text. The more important this aesthetic becomes, the less there is room for stable, longterm, ‘finished’ textual objects.

RMIT University


Appadurai, Arjun (1996) Modernity at large Minnesota: University of Minnesota Press.
AustralianIT (2004). ‘Ninemsn books maiden profit’ AustralianIT: http://australianit.news.com.au/articles/0,7204,10574085%5e15306%5e%5enbv%5e,00.html.
Benjamin, Walter (1935) The work of art in the age of mechanical reproduction http://www.student.math.uwaterloo.ca/~cs492/Benjamin.html (accessed 27/04/02).
Bernstein, M (2001) Card shark and thespis: exotic tools for hypertext narrative Markbernstein.org: http://www.markbernstein.org/talks/HT01.html.
Burroughs (1978) ‘The cut-up method of Brion Gysin’ Reprinted in The new media reader (2003) Edited by Noah Wardrip-Fruin and Nick Montfort. Cambridge, MA: MIT Press.
Castalls, Manuel. ‘An introduction to the information age’ In The media reader: continuity and transformation Edited by Hugh Mackay and Tim O’Sullivan London: Sage, pages 398–410 (Originally an address to the Conference on Information and the City, Oxford, 1996)).
CNN (2004) The latest protest tool: 'texting' CNN.com: http://www.cnn.com/2004/TECH/ptech/09/02/textmessaging.protest.ap/
Cramer, Florian (1996–2000) Permutations http://userpage.fu-berlin.de/~cantsin/permutations/index.cgi.
Darley, Andrew (2000) Visual digital culture: surface play and spectacle in new media genres London: Routledge.
Dawkins, Richard (1990) The Selfish Gene 2nd edition. Oxford: Oxford University Press.
Fournel, Paul (1961) ‘Computer and writer: the Centre Pompidou experiment’ Reprinted in The new media reader (2003) Edited by Noah Wardrip-Fruin and Nick Montfort. Cambridge, MA: MIT Press.
Galloway, Alexander (2004). Protocol. Cambridge, MA: MIT Press.
geniwate (2003) concatenation http://www.idaspoetics.com.au/generative/generative.html. 
geniwate (2004) semtexts http://www.idaspoetics.com.au/generative/generative.html.
geniwate and Larsen, Deena (2003) The Princess Murderer. Iowa Review: http://www.uiowa.edu/~iareview/tirweb/feature/june03/larsen_geni/prin.html.
geniwate and Dane (2002). Machine Corporation http://www.eatmydata.co.uk/MachineCorporation/index.html.
Graham, Alan and Burton, Bonnie (2004) Never threaten to eat your co-workers — best of blogs New York: Apress.
Graham, Trevor; Hesp, Rose and Wellington, Rob (c.2002) Homeless ABC: http://abc.net.au/homeless/.
Januzzi, M and Smith, R (2000). Dotlinepixel: thoughts on cross-media design. Switzerland: Gabriele Capelli Editore.
Manovich, Lev (2001) Chapter 5: ‘The database’ In The language of new media. Cambridge, Mass: MIT Press, pp.218–137.
McCamish, Thornton (May 15, 2004) ‘Never Alone’ The Age Review pp.2–3.
Mirzeoff, Nicholas (1999) An introduction to visual culture New York: Routledge.
Mitchell, William J (2003) Me++: the cyborg self and the networked city Cambridge MA: MIT Press.
Moulthrop, S (c.2001) Polymers, Paranoia, and the Rhetorics hypertext The new media reader: http://www.newmediareader.com/cd_samples/WOE/Moulthrop_Polymers.html (accessed 08/08/03).
ninemsn http://www.ninemsn.com/.
Queneau, Raymond (1961) ‘Six mille milliards de poèmes’ Reprinted in The new media reader (2003). Edited by Noah Wardrip-Fruin and Nick Montfort. Cambridge, MA: MIT Press.
Rheingold, Howard (c.2002) Smart Mobs: The Next Social Revolution book summary Weblog: http://www.smartmobs.com/book/book_summ.html.
Ricoeur, Paul (1984). Time and narrative. Volume 1. Translated by Kathleen McLaughlin and David Pellauer. Chicago: University of Chicago Press.
Ryan, Marie-Laure (July 2001) ‘Beyond myth and metaphor — the case of narrative in digital media’ Game Studies, volume 1, issue 1. http://www.gamestudies.org/0101/ryan/.
Southern Star-Endemol and Channel 10 (2001–2004). Big Brother: http://bigbrother.ten.com.au/holding_post/index.html.
Tzara, Tristan (1920) ‘To make a dadaist poem’ http://userpage.fu-berlin.de/~cantsin/permutations/tzara/poeme_dadaiste.cgi (Originally published as ‘Dada manifeste sur l'amour faible et l'amour amer, VIII’ In Oeuvres complètes, vol.1, Paris, 1975, p.382).
Williams, Raymond (1963) Culture and Society 1780–1950. UK: Pelican.

[1] Galloway argues that media likes to conceal the way it is produced (2004 p.64); an ideal of ‘smoothness’ (2004 p.67) exists, exemplified by the browser, which is an apparatus to hide the apparatus (2004 p.75).
[2]Mirzeoff argues that the final stage in the history of the image is the copy with no original (then it can't be a copy) (Mirzeoff 1999 p.28).
[3] I define levels of interactivity according to the power the use has over the text. The highest level of interactivity is when a user can permanently change a text.
[4]Castalls suggests: ‘As with all historical transformations, the emergence of a new social structure is necessarily linked to the redefinition of the material foundations of life, time and space’ (Castalls 1999 p.405). New technologies participate in a relentless effort to annihilate time, ‘including past, present and future in the same hypertext, thus eliminating the “succession of things” that, according the Leibniz, characterizes time, so that without things and their sequential ordering there is no longer time in society (ibid)’. The network society constructs ‘all dominant processes … around timeless time’ (Castalls 1999, page 406); this competes in various unfortunate ways with biological and clock time (ibid). Thus ‘...a fundamental struggle in our society is around the redefinition of time, between its annihilation or desequencing by networks, on one hand, and, on the other hand, the consciousness of glacial time...a battle undertaken, in my view, by the environmental movement’ (Castalls 1999 p.406).
[5] Bernstein wonders about the extent to which ‘hypertextual’ works are actually a response to software rather than an exploration of the concept of hypertext per se. He suggests that we need to plan our works in terms of constraints and assertions, rather than simply linking unconnected nodes. This is a rule-based (ie, algorithmic) narrative environment. ‘…we need plenty of ways to assert and retract conditions; otherwise, we’ll soon “get stuck”. When we give readers a sense of control over the algorithms they set off, the reader becomes ‘a dramatic co-conspirator’ transferring agency to the automata established in the script/scripting (Bernstein 2001).
[6]As defined by Richard Dawkins in The Selfish Gene (1976): ‘a unit of cultural transmission, or a unit of imitation. Examples of memes are tunes, ideas, catch-phrases, clothes, fashions, ways of making pots or of building arches. Just as genes propagate themselves in the gene pool by leaping from body to body via sperms or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain via a process which, in the broad sense, can be called imitation’.
[7] Bernstein agues that a system of constraints and preconditions for what text chunks can appear is hypertext. I use hypertext in a more limited sense; algorithmic behaviours enhance hypertext by uncoupling the necessary connections that traditional hypertext operate upon. Perhaps the distinction between traditional hypertext and algorithmically enhanced “hypertext” will one day be unnecessary, however it serves the purpose of my paper to maintain the distinction. Although Bernstein goes on to say that such systems are not games, I would suggest that the distinction between game and narrative is increasingly problematised and may not, in the long run, be particularly useful.
[8] Mitchell mentions other examples (Mitchell 2003 p.209).