Any good novel, then ...can afford its reader a way of being - if not being there, in the other world, then being here, in this. It proposes a locus of reclamation, becomes a place inside the place we are situated, a charged time contained within the more diffuse time of daily living (Birkerts, 1996: 10) 
In this chapter I will be discussing how one can make hyperfictions both more 'hyper' and more apposite as fictions. If, as Birkerts claims, the raison d'être of the novel is to offer its readers a 'locus of reclamation', could and should this aesthetic of the novel be applied to hyperfiction writing as well?
Since there is a broad spectrum of viewpoints from which one could approach this discussion, I have chosen to focus mainly on how one could increase the level of interaction between reader and 'work', including the possibility of presenting the reader with a more dynamic and flexible structure than the static hypertext linking. Consequently emphasis is placed on the structure and overall design of the hyperfiction product rather than on the actual content, that is the stories contained within it.To broaden my 'hyperizons', I have tried to find inspiration in other genres of computer 'fictions': the part exploratory hypermedia-production Sophie's World and the interactive film, The Pandora Directive. Thus, in my attempt to engage with these issues, I have tried to envisage what could be achieved within the limits imposed by the authorial programs available on the market today, in order to avoid resorting to speculative 'science fiction' thinking in the literal sense of this phrase.
One of the cardinal characteristics of hyperfiction writing is (or at least ought to be) an enhanced interaction between the reader of the hyperfiction work and the work itself. In theory, this interaction should take place on a level that allows the reader to become a co-writer himself, thus through his interaction effacing any succinct and authoritative distinction between the original author and himself. Judging from the hyperfictions examined in the previous chapter, in current practice the interaction is limited to the possibility of choosing between a number of ways of in which to navigate through the storyweb. In fact, when reading a hyperfiction the authoritative control of the author is experienced as being much stronger than in traditional works of fiction. Furthermore the hyperwriting structure seems to have given birth to a kind of fiction of a tableau-like and static nature that does not fulfill the theorists' promises of a prospectively organic and evolving textual structure.
In consequence, this chapter will examine and discuss these particular questions: how does one achieve a more flexible structure or, more to the point, how does one provide the reader with the feeling of interacting with a more flexible and dynamic storyweb that ideally makes viable an organic development of a number of tightly interwoven, 'immersable' stories?
Evidently pertinent to the possibility of finding a solution to these question is a precursory examination of the present 'nature' of the hyperfiction reader. Throughout the history of print the reader that has been addressed, implicitly or explicitly, in the works of fiction, has always been a reader of a 'general' identity, (general in this context meaning of a 'non-individual' nature). With the advent of hyperfiction this condition has partly been changed. The overall hypertext structure now allows the reader to perform a reading that follows the course of his particular whims and associations. But, nonetheless, the printed content of the work still does not accommodate itself to the individual reader. Hence the reader of the work remains a 'general' reader in that he or she still reads the exact same words as all other readers and in so far that, for all the program 'running' the storyweb is able to register, it could still be anybody pressing the keys on the keyboard. In order for the reader to become a true 'interactor' would not a truly interactive hyperfiction obligate an interaction that allows for individuality on all levels of its structure?
Hence, a genuinely interactive hyperfiction should be able to address its reader on a 'face-to-face' level, allowing individuality on both sides of the screen. This might be implementable by, for instance, partly adjusting the text to the choices of the reader, making room for individual contributions to the work or, in extremum, 'reading' the reader himself and adjusting the hyperfiction according to its 'diagnosis'. To a certain degree, the CD-ROM's discussed below do seem to make room for this kind of 'individualised' interaction which is why they might be worthwhile examining although these products have not originally been designed and written as 'pure' hyperliterary works.
The CD-ROM Sophie's World is based on Jostein Garder's bestseller and semi-fictive book of the same title. The original book is intended to be an alternative and relatively easy accessible introduction to the history and subjects of philosophy. Consequently, in order to facilitate the digestion of relatively 'heavy' philosophical matter, the factual information about various philosophers and their theories is related through the fictive character of Alberto who appears out of thin air in the beginning of the book to give the young girl, Sophie, a course in philosophy. Hence, although the CD-ROM Sophie's world has a clearly structured narrative line, this product is mainly exploratory of genre, providing its user with a fairly detailed factual section on various philosophers (including brief introductions to the Cultural History of the centuries they belong to) and philosophical issues.
The course (and the book) unfolds as an on-and-off Socratic dialogue between teacher and pupil (Alberto and Sophie), with Sophie representing the allegedly naive reader's point of view. The narrative crux is to be found approximately halfway through the book, when Alberto and Sophie themselves are revealed as being fictive characters, appearing in a story written by Major Knox as a birthday present for his daughter Hilde. However, on the CD-ROM, it is not as characters in a book, but of course, as characters in a computer program!
The CD-ROM producers have tried to capture the dialogic character of the book -and have furthermore extended it to include the 'reader' of the CD-ROM - by inventing a 'dialogue-engine' that automatically displays the dialogues between Sophie and Alberto and furthermore makes it possible for the reader to enter answers to Alberto's philosophical questions in a separate text box. In turn Alberto responds specifically to the reader's input. Unfortunately (most likely due to the pedagogic intentions lurking), the reader more often than not can only answer with 'yes' or 'no' or specific words and quotes (to be found in the factual parts of the product) which makes it difficult to conceive of this interaction as being true-to-life dialogue.
Nevertheless, this 'inclusion' of the reader is successful as it forces the reader to actively embrace and respond to the questions posed and in that it encourages her to become Sophie's partner and allied in the search for a philosophical understanding. Albeit new wine in old bottles, the fact that Alberto addresses the reader by the reader's own name (entered in the start-up phase of the program) succeeds as well in increasing the feeling of being engaged in a one-to-one interaction.
One amongst several interesting features in this CD-ROM specifically related to the problematics of interaction is the 'characterisation' of the reader, that appears in the 'scene' Between the Lines towards the end of the story. This characterisation is fairly elaborate and quite puzzling and hence fulfills its intended(?) function as food for thought confronting the reader with his or hers philosophical and personal inclinations manifested through choices, preferences and input in the interactive sections of the product.
Finally, the producers have managed to absorb the characters of the book into the universe of the computer by letting them appear en miniature on stage (i.e.: on screen) in a complete fictional setting (a garden, a wood) - that is, exactly after it has been revealed that they are fictional characters only! In the shape of real actors performing some of the (partly rewritten) dialogues from the book, this visualisation of the characters enriches the game with an almost surreal filmic quality. This 'fleshing out' works out particularly well in the final scene in which the user is presented to Major Knox in persona for the first time. He even addresses the reader directly, asking him "What do you think we are made of?" thus establishing a even more unmediated connection between the storyworld and the reader's reality than the dialogue-box has done.
Although still at its infant stage this dialogic interaction between program and reader opens up interesting prospects. For instance, one could imagine a hyperfiction created in Storyspace but supplemented with a dialogue-box-interaction like in Sophie's World. The 'box' could ask the readers different questions of more or less provocative character and according to the answers given by the reader, take him to specific lexias in the storyweb or deny him access to others (imagine an alert message reading as follows: 'Since you think that Y should not engage in an affair with N you would not be interested in reading this'). Correspondingly, the reader could be 'diagnosed' on the basis of his choices, (i.e. read by the program!) to be presented with this diagnosis in a kind of 'readers logbook' at the end of each reading. The program's reflection of the reader could furthermore lead to 'educational' attempts by letting the program force the reader to 'endure' certain scenes or lexias that he would otherwise have avoided. However, one should be aware of the fact that playing with the 'interactor' like this would mean that the one in control would definitely be the program, not the interactor himself!
The Pandora Directive story is what one might call 'postmodern' in its irreverent bricolage of various narrative genres. Thus the 'plot' and characters mainly belongs to the traditional 1940's hard-boiled detective story with some spicy flavour added by 'Government conspiracy plot' much like the one found in the presently very popular X-Files and Dark Skies-series. The entire story is set in a Blade Runner-like science-fiction world sometime in the next century with a lot of nifty technical devices appropriate to this genre (for instance the almost true to life 'Vidphone' )
However, the basic ingredient in the game is still the back-to-the-basics gameplaying of the adventure/puzzle type in which one has to look for various 'tools' in various rooms and settings, solve a number of puzzles and combine strange items in order to move through the game/story.
The game is divided into a number of sections, the principle of division being the number of days Tex Murphy, the protagonist, has been on the job. Throughout the game the player alters between being forced into a passive watching mode or an interactive 'agent' mode, that is: between watching short videoclips in which the actor Tex on screen partakes in some kind of action or moving around 'as' Tex in the 3-D world of the game (mainly the alleys of Chandler Street).
From the beginning the player is prompted to make himself identify with the storyworld through a number of visual cues. The 3-D interactive world and the possibility of moving 360 degrees around, standing on tiptoes or approaching 'crawling on the floor'-level are supposed to provide the player with the feeling of moving in and through a 'real' environment, being Tex him- or herself. The filmscenes (videoclips) makes use of traditional camera techniques to make the player assume Tex's point of view. For instance, several of the film-scenes comprise from the back or over-shoulder shots, close-ups of Tex's eyes or face prompting the reader to identification with his point of view. The use of 'I' in Tex's voice-over comments is another traditional 'trick' (adopted from both books and film) to bring about identification with 'the speaking person' (the reader is given access to his consciousness and experience him from the 'inside').
Finally, in between the dialogue in the filmscenes, the player must choose what Tex says to the people, he interrogates. Throughout the game, the player has in general a choice between 3 'response' attitudes, more or less 'aggressive' in tone (e.g ranging from polite to rude).This implies, that on a certain level, the player can actually choose which kind of person Tex and himself is. Nonetheless, in practice, the reader's choices are limited, since at certain points he simply have to respond in a particular mode to be able to continue the game or obtain the information, he needs. Consequently, - after all The Pandora Directive is a computer -game most of the time - setting off the actions resulting from the dialogue, not the dialogues themselves, becomes the main incentive for engaging in the dialogues in the first place. This, however, need not be the case in a interactive fiction that asked to be read for 'its own sake', not with a specific purpose.
1) Immersion and interaction combined
Albeit that the intrinsic game-nature of The Pandora Directive prompts a purposeful, not 'leisurely' reading, nevertheless it illustrates how one may also go about creating a fiction in which interaction is an essential part of the process of identification with the work. Do not the various means of identification with the protagonist and the storyworld universe literally make room for a 'reading' of the kind Birkerts and other defenders of the 'old-fashioned' aesthetics of novel-reading advocate - reading as transferral to another place and time? Furthermore, since this game is a very long game (6 CD-ROM's) it makes space for the narration and development of a fairly complicated story and plotline whose outcome the reader can affect by his choice of 'voice' and action. However, the 'content' is not particularly interesting in terms of relating an experience of how it is to be human in the world of today since The Pandora Directive as a game is not created with 'experience-transferral' as it main objective. But might nevertheless provide its readers with the opportunity of, for a while, being somebody very different from themselves.
Still, employing some of the features from the Pandora Directive in 'serious' hyperfiction writing might lead to captivating results. Could not a number of lexias or key-trails consist of visits to 3-D worlds where the reader could meet and be a character in the storyworld for a moment? Choosing among a number of responses or response-modes could take the reader to different lexias in the storyweb either of similar form or merely 'just' written. This does not necessarily imply that the imaginative abilities of the reader are stripped away or that the reading as such loses its creative qualities when the storyworld is perceptually rendered (Bolter's anxiety). Visualisation might dispense with the abstract quality of the written word but does necessarily need to imply a simultaneous loss of semiological plurality. The visualisation might be of a very general, crudely 'cartoonish' kind that still makes room for interpretative 'readings' of the characters individual appearance. Or the visualisation of the characters might constantly change during the playback of the visualised scenes so as not to convey a 'fixed' image of the characters in question. The 'world' surrounding the characters might likewise be of a formal, 'sketchy' outline rather than meticulously true to life - or of the merely illustrative 'quadratic' quality used in present day virtual reality graphics.
2) The AI as character
Promulgating the influence of the reader on the outcome and outline of the stories in a hyperfiction storyweb would seem to be in accordance with the 'ideology of hypertext' that aims for maximum proximity between author/scriptor and reader/interactor. However, increasing the interaction between reader and 'program' as I have suggested, simultaneously raises new questions to be dealt with. What is the 'nature' of the program itself (be it Storyspace or an engine specifically designed for an individual product) that makes possible this interaction? Is it a 'stand-in' or a representative of the scriptor? Is it another mediator - the word itself being the first - between the communicating parties, scriptor/reader; and if so, does it then increase the distance between these two entities rather than diminish it? 
One might argue that even if the written word itself has been desubstantiated, prone to give way to 'yieldings' and erasures, the program controlling the way in which this desubstantiation is eventuated remains sacrosanct and indeed highly substantial. Another kind of materiality exists 'between the lines', a materiality of electronic, not physical, nature but a 'materiality' that demands as much respect and reverence as the physicality of a book. The intrinsic structure of the program is as inviolable as matter itself unless you are among the chosen few that can 'read' and write programming language. To the layman 'a program' might be as impenetrable as the walls of a labyrinth.
Not that one should conceive of a 'program' as being intelligent in any way. In comparison with the 'live' scriptor who might eventually be open to argument there will always be only so much reader-feedback that the program is able to understand and reflect upon. It has no consciousness of its own and can only represent or reproduce the intelligence of the human who produced it, who might not even be identical with the scriptor himself.And the necessity of thinking structurisation itself in digital terms is bound to impose limits on the way in which present and future scriptors and program designers envisage and implement the hypertext structure. Furthermore, the more complicated these programmes grow, the more liable to mysterious interferences by 'bugs' do they become, hence opening up for readings of a somewhat 'uncontrollable' nature. Thus, although the number of hypertext writing programmes are rather limited, one should not discount the existence and latent possibility of 'interference' by the very program structures underlying these, both when writing and reading hyperfictions as well as when dealing with individual hyperwriting programmes (such as Storyspace or Guide).
Hence, one way of confronting the invisibility of these programmes would logically be to make them visible agents or characters in the reading or in the storyworld. This could either by effectuated by literally referring to the programmes in the text boxes or lexias. For instance, on a number of occasions in Sophie's World) the dialogue box eventually reads:
PROGRAMME: Sophie leaves
hence bringing the existence of the program to the fore.
Another means of making the program visible could be to make use of it as a thwarted version of an explicit narrator. One could make the program 'talk' in alert message-boxes, verbally commenting on the readers choices of links or reading him in every so many ways. In fact, restricting the reader's choice of lexias by making the program prevent the reader from accessing them, could be a sophisticated way of drawing attention to the fact that the program is always on a very basic level in control of the reader. Indeed, taken to its extreme the program could be programmed to consciously 'crash' when the reader has performed a number of specified actions - a perverted version of the death of the author (and the victory of the machines)! This would, however, be a story in itself, that has yet to be told.
The problem with getting inside the act of reading, is its ubiquity - there's no escaping it, and, like any environment that we are overly familiar with, we no longer see it. When we read print narratives we arrive already equipped with a full repertoire of reactions and strategies....We never come face to face with the ground zero of reading (Joyce, p. 227, quoting Jane Yellowlees Douglas)
This paper has tried to examine and discuss various aspects of present day writings on and in hypertext fiction, focusing especially on the problematics of interaction and readership in both theory and practice. I am well aware that since more and more scholars turn their attention to this field of study, the discussion of how to add to the variety and degree of engagement in the reading and of how to move towards a 'truer' and more dynamic interaction, is one that has already been engaged with in various writings.
Thus, most sections of my discussion do not put forth any arguments or ideas that have not already been vented elsewhere. Nonetheless, to my knowledge a discussion of the relationship between interaction and a multimedia approach to the construction of hyperfictions has by and large remained untouched. One might wonder why, since ours is an age in which the concept of what 'text' is, seem to be extended continuously. Perhaps part of the explanation can be found in the fact that the majority of writers (both in theory and practice), in spite of the youthfulness of their 'printoclasm', alone by virtue of their physical age are rooted in a tradition of 'one or two-medium' writing and reading.Furthermore, as discussed in chapter 1, this 'printoclasm' is intimately connected with a project of liberation of political and institutional character that does not leave much room for subtle elaborations on aesthetic practice. Last, but not least, as in Bolter's case, part of the explanation might well be that actually they disavow the 'rhetorical' and persuasive character of moving images on the basis of the argument that that which is concretized cannot be transparent, i.e. makes no room for the reader's imagination (that this notion might not in fact be true for all instances of 'visual' rendering, I have briefly discussed in chapter 3)
Consequently, the theorists' rejection of media that 'begs immersion', such as Virtual Reality of the kind one finds in The Pandora Directive imply that what is 'good' interaction in their eyes is the version of interaction that unfolds on a conscious level of reading. To them perceptual immersion - understood as the kind of reading in which the reader lets himself be absorbed completely by the text (i.e. the fictional universe or the 'storyworld' rendered) - seems to be equal to a submission to the authority of the 'master'-text, hence preventing a genuinely liberating and interactive reading from coming taking place. Accordingly, 'good' hyperfiction would seem to be the one which demands of its reader a presence of a mind of reason rather than of a state of emotion (which, to a certain extent, do in fact pertain to hypermedia in that their structure call for the employment of reason by demanding of their readers the choosing of which links to follow, hence some 'conscious' reflecting on this act).
But, if the 'ideology of hyperfiction'  first and foremost is to liberate its reader, letting him or her take 'authoritative' control of the reading, this might as well be embodied in a multimedia environment as in one consisting of printed words only. Hopefully, my study of the state-of-the-art of other forms of computer-'fictions' such as Sophie's World or The Pandora Directive will have shown that one can find inspiration and suggestions on how to engage with, for instance, the problematics of 'person-to-person' interaction or character identification, in other media than those of print. Accordingly; as I pointed to in the brief discussion of the difference between the nature of signification in written and visual exemplifications – a logical consequence of multimedia inclusion in hyperfiction practice, should be an inclusion in hyperfiction theory of the academic discourses on other media, such as film theory or computer game studies.
Still, in whatever direction the future of hyperfiction design and writing might head, a fundamental paradox seem to remain irresolvable: interaction necessarily implies interruption of the space and time of a given storyworld whereas immersion in a fictional universe seems to demand some kind of 'seamless' rendering of it, which in turn seems to beg for some kind of authorial control of the text. In this context, it is interesting that whereas exploratory hypertext makes possible the breakdown of the institutional authority of text (that Bolter and Landow so welcome), the constructive hypertext (which in theory leaves the construction of order and meaning in the text completely to the reader) in practice seems to reinstall the 'authoritative' control of the author on another level, that of the structure of structures, as my analyses of afternoon and Victory Garden have shown.
In order to engage in a productive way with some of the inherent problematics of story-telling and readership within the field of hyperfiction studies, to me the most fruitful approach to further discussions would seem to be to conceive of hyperfictions as a new literary genre which both offers us, its readers and theorists, a new way of thinking of readership as well as it provides certain literary practitioners, its 'scriptors', with the medium they have been looking for to make some of their literary dreams come true. Most importantly, however, what this new genre offers us, is the possibility of anew 'coming face to face with the ground zero of reading' and with what 'this thing' that we read should be called. HyperLiterature, perhaps?
Thus, whether one rejoices or trembles in the encounter with this brave new world of writing, not only should we keep asking: 'What can I use it for and how do I apply the possibility of interaction and empowerment of reader hat hypertext offers me?', but furthermore what is indeed important is, that we, given a new point of view, should seize the opportunity to ask ourselves such basic questions as: 'Why do I want to read in the first place? and 'Why should I want to read this kind of fiction?'. Thus to broaden the horizons of hyperfiction studies, I suggest that future seamen n the seas of hypertext should embark on a exploratory journey which should have as its objective to establish and identify what an 'erotics of hyperfiction' would be like. For instance, can we and should we as readers be allowed to satisfy our 'blessed rage for order' (or immersion) when we engage in a hyperfiction reading? Hence; while at present day the reading and writing hyperfiction seems to offer itself of a new way of 'writing the self' - meeting your own reflection through the choices you make – or as the fictive genre that by denying us 'another place and time' once and for all deprives us of the possibility of restoring it; only the exploration of all the 'hyperizons' of hyperfiction writing in the time to come can reveal what a true 'erotics' of reading hyperfiction would be like. However, one must never forget, that all we can know, all we will know ultimately depends on which trail we choose to follow.