Monday 28 December 2009


More Eunoia machines, or Sound boxes, these are for the public performance of site specific, sonic narratives.

Sunday 27 December 2009

Demented Robot from my sister
Carrying on with editing the whole thesis and dumping core chunks of it here for future cramming purposes:
//////////////////

My own practice is grounded in a reconsideration of the ways in which interactivity can be deployed and defined, for example, decentralising the notion of individual user choice in favour of more collective processes and less overt or conventionally individualised forms of ‘user’ control. Wherever possible my own partiality of perspective, in the form of rules embedded in the code, can be examined and, where structurally possible, changed by readers. review 16 - 1.5 spacing



//////////////////////

Writing Machines contains a chapter on Talan Memmott’s From Lexia to Perplexia (2000), a digital work that deploys computation to a significant degree, beyond using the computer as an image processing tool or as an inert vessel for a literalised remediation of the book form. Hayles’s emphasis upon the materiality and embodiment of our engagement with diverse forms of book has been an essential reference and a vindication of the possibilities that computation has to offer the evolving form of the artists’ book. Rev 50

/////////////////////////////////////////////

Words may allow us to articulate and communicate the realizations that happen through material thinking, but as a mode of thought, material thinking involves a particular responsiveness to or conjunction with the intelligence of materials and processes in practice. Material thinking is the magic of handling (Bolt, 2006:1).
the magic of handling review 18

//////////////////////////////////////////////////////////////

This relational convergence and entanglement of events is not the same as claiming that some sort of democratic ‘co-production’ between readers and writers is occurring. Such a notion, is, according to Aarseth, part of the mythology of hypertext fiction, and has been a key factor in claims that hypertextual structures are somehow more suited to human cognitive processes (whatever they may be) than traditional texts review 27//////////////////
Terminal Time uses audience polling and demographic data collection as a way of constructing unintended or atypical histories. But Mateas, Domike and Vanouse are keen to emphasise that Terminal Time is not purely an exercise in cynical commentary but an ‘exploration of some of the unexamined assumptions and unintended side effects of information technology’ (162), including the myth of neutrality or the idea of the computer as a value-free conduit for unfettered user agency. review 31
////////////////////
One of my goals has been to create a reflexive (or, indeed, a diffractive) computational system that can accommodate the type of epistemological shifts discussed in this review. Such a system should be able to reconfigure some of its own standards or presumptions if the agents that monitor it indicate the need for such an adjustment. Through my practice I aim to develop a form of contingent meaning making that is also emphasised, albeit in different terms, in the theoretical writings of Karen Barad (2007). (review 15)
/////////////////


Despite the pluralist discourse of advertisers and technological determinists, the creators of this project ask the crucial question: what is not allowed by a computer system? (163). This question can be applied to any computer system. What types of behaviour and information are excluded beneath the apparently participatory design? What are the built in boundaries of its design? Terminal Time is built in recognition that any system is the ‘active messengers of a worldview’ rev 31
////////

It is interesting to note that Domike, Mateas and Vanouse reject a strongly emergent architecture, which they describe as a ‘form of rootless non-intention’ (168). Their critique is important to my own work, as a decentralised, more widely distributed system is key to my design, but it is not one that I would describe as emergent, for similar reasons. Domike, Mateas and Vanouse rely heavily on a top-down, high-level knowledge base within the Terminal Time architecture. Their choice of such a high level system is confusingly defined by motivations that are similar to Phoebe Sengers’s incentives for making use of less centralized forms of architecture. Mateas, Domike and Vanouse reject ‘strongly emergent’ processes, which, they state, lack authorial ‘hooks’: rev 32

//////

I do not have any commitment to or interest in the notion of emergence for its own sake, rather I have a commitment to systems that (wherever possible) try to eschew predefined boundaries between users and contexts, or society/individual dualisms; a form of contingent meaning-making that is closer to the description made by Sarah Kember (2003), in relation to Artificial Intelligence and Alife, in which emergence is characterised as ‘based on a non-linear, non-deterministic model of connection or communication between multiple rather than individual units’.(Kember 2003: 110). rev 33

///////////////////////

These anthropomorphic value judgements do not always seem logical or necessary but as the authors point out, the anthropomorphising sensibility ‘inflects the whole idea of interactivity by lending it a context of person-to-person connection’ (22). This is perhaps where chat bots from Eliza onwards have failed both in intention and practice, pursuing a mimetic or anthropomorphic goal at the expense of exploring the true strengths of computational forms and structures. rev 34
///////

My work aims to place location, sensation and tactile response at the forefront of the site-oriented experience it offers to readers. Spatial fragmentation and psychic disorientation are also part of my site-specific narrative works, in which readers can find narrative threads through tactile experiences or lose themselves and find the work collapsing around them. The South project is committed to an investigation of alternative forms of knowing that are grounded in material practices, such as the generation of meaning through tactile experiences. Readers are invited to construct their own narrative interpretations through touch, sound and movement, to locate themselves within and through the environment. rev 43
/////////////////////////
Readers of the South book and software writings are required to engage with the construction or representation of their own subjectivity. This construction is also enacted through sensory experiments as well as more traditional forms of psychological assessment. If readers attempt to ‘cheat’ or alter the system’s assessment of their personality type, then they are engaging in an unconcealed form of subjective performativity. This notion of performing ones identity has a degree of resonance with Judith Butler’s idea of gender as ‘a regulated system of performances’ (Mansfield 200:77), but Butler’s statement that the body has ‘no ontological status apart from the various acts which constitute its reality’ (Butler 1990: 136) is contentious and has been contested at length (see Kember 2003), not least for the circularity of an argument that equates the total meaning of the body with discourse . rev 44
///////////////////
The ocean is the ground in Polynesia and has the ability to “rise up to decompose the face of identity”’ (Hoete, 2004:216). Refiti and Hoete’s conception of site is challenging and non-literal, providing a valuable practice-based example of extending notions of site-specificity. My own conception of the South Bank site is also paradoxical and mutable. The site is present within my work as a series of both virtual and physical locations, but also as subjective, inter-subjective and situated formations, as complex sites of mutual constitution. rev 45
//////////////////////////to here blog
South proposes a mutable form of both subjectivity and site specificity. The site in South is formulated by specific situations and corporeal sensations, the book therefore emphasises both situated and embodied interaction. Many of the evaluative procedures involve the senses, and indeed the progression of the evaluations through the five senses is part of the underlying narrative of the assessment process. Chapter 3, 8
//////////////////////
My own subjectivity is also posited as a site or meta-location, resonating throughout every aspect of the South project. The book and software therefore aims to understand individual subjects and sites, but in order to work with these concepts I have had to investigate what the notion of a subject and subjective experience meansChapter 3, 8//////////////////////
The South book also limits the mobility of readers dependent on their ‘scores’ and psychometric evaluations. In conjunction with the South software and egg readers can also find their choices limited or expanded along gender and other lines, depending on an ongoing assessment of overall responses.chapter 3, 12
/////////////The paradoxical and borderless Klein bottle, (illustrated overleaf) is an apt representation of the condition Callois describes, and which Ivan Dâr endures, ‘a space where things cannot be put’, because he has no location in space. At times the book deliberately aims to bewilder or undermine its own readers, hinting at the possibility of space as an annihilating agency working against them, generating a hostile space by means of its aesthetic strategies and by asking readers to undertake paradoxical or impossible tasks, lying to them, issuing contradictory instructions or leaving exasperating lose endsChapter 3, 16
////////////////////
Andrew Haslam (2006) describes the way in which the design of books can facilitate ‘emotionally ‘repositioning’ the reader’ (Haslam, 2006 : 26). In South the visual language of the book is deployed to communicate and, indeed to manipulate readers, as much as its written language. Concepts such as ‘chunking’, ‘mimicry’ and ‘self-similarity’, explicated in for example Lidwell, Holden and Butler (2003), and drawing upon the Gestalt theories of Max Wertheimer have been used within the formulation of South to generate meta-narratives through non-verbal means. These are narratives of dislocation and spatial ambiguity that reflect the wider concerns and fictional themes of the South project. chapter 3, 16
//////////////////////////
The verbal and non-verbal stories embedded in South are framed as a form of generative research practice. Readers are invited to extend the themes of these stories through the various exercises and encounters generated by their navigation of the South Bank. Chapter 3, 18
/////////////////
Haraway and Barad (among others) frame reflexivity as problematically directed in orthodox, circular, notions of cause and effect. Haraway writes of the limitation of reflexivity, ‘which seems not to be able to get beyond self-vision as the cure for self-invisibility’ (Haraway, 1997:33). Haraway frames reflexivity as symptomatic of the problem it aims to address, that of the invisibility or spurious neutrality of the scientist. ‘Reflexivity is a bad trope for escaping the false choice between realism and relativism’, Haraway explains (16). Instead she proposes a ‘diffractive’ methodology, in which ‘the production of difference patterns, might be a more useful metaphor for the needed work than reflexivity’ (34). Diffraction suggests a less solipsistic form of knowing than the reflected understanding implied by reflexivity, it suggests the possibility of generating new knowledge without negating our own embedded perspectives and experiences.Chapter 3 22
///////////////////////////////
An artist’s book such as Helen Douglas and Telfer Stokes’s Real Fiction (1987) is cited by Drucker as ‘not merely about its own making, but also about its own conceiving’ (195). Similarly the egg shaped interim object deployed in conjunction with the South book and software may be seen as a means of performing the birth of an artists’ book, readers must physically perform certain functions to complete their own version of the book.chapter 3, 23
///////////////////
The canon of orthodox principles for Western book design is relatively narrow, with a few very clearly delineated (though often unarticulated) systems such as the Golden canon, and the Van de Graaf canon, (also known as ‘the secret canon’), as well as more contemporary divisions of space. The canons are designed to elicit concepts such as harmony and balance in the design of book pages and to emphasize content over form. Chapter 3, 24
///////////////////////
The South project therefore shares formal aspects of earlier works, including Sol LeWitt’s algorithmically generated books, such as Four basic kinds of straight lines (1963), which follow strict generative procedures, but the South project also presents quite distinct features, most notably those features facilitated by the use of techniques from the domain of artificial intelligence, such as data collection and analysis, and the ability to make evaluations and learn from readers and specific situations through its agent based structures. Chapter 3, 27
///////////////////////////// The South system is designed to guide readers through an initial process of subjective evaluation in order to generate appropriate content, both factual and fictional, about the South Bank area of South London. The processes deployed in this evaluation are the result of research into psychometric techniques but also of research into critical thinking around the very notion of obtaining an objective, categorical and stable assessment of an individual’s personality. chapter 4, 3
/////////////////////////
The philosopher Rosi Braidotti (2002) is energetic in her call for ‘more innovative and creative energy in thinking about the structures of subjectivity at a time in history when social, economic, cultural and symbolic regimes of representation are changing very fast ‘(Braidotti, 2002 :73). But Braidotti also asks, is the ‘model of scientific rationality a suitable frame of reference to express the new subjectivity? Is the model of artistic creativity any better? How does it act upon the social imaginary? Will mythos or logos prove to be a better ally in the big leap across the post-modern void?’ (173). chapter 4, 3
mythos and logos - loosely poetry (art, drama) and reason
//////////////
Writers such as Susan Hekman (1990), Evelyn Fox-Keller (1985), Alison Adam, Alcoff and Potter, Lorraine Code and Sandra Harding have written at length about how, in Lorraine Code’s words ‘such beliefs derive from conceptions of detached and faceless cognitive agency that mask the variability of the experiences and practices from which knowledge is constructed’ (Alcoff and Potter: 26). This is not to argue for a relativist epistemology, (described by Sandra Harding as ‘anathema to any scientific project’ (61)) but to argue for the appropriateness of looking at new means of thinking about subjectivity and by extension subjectifying processes. chapter 4, 6
///////////////////////////

The conception of the subject is contentious and has been the focus of many conflicting theories. Philosophers such as Nietzsche and Spinoza challenged Cartesian philosophy. Nietzsche in (1887 ), rejected the presumption of humanist agency and rational intent, and in (1677) Spinoza’s Monism was a direct attack upon Cartesian mind-body dualism . Likewise the processes of subjective evaluation developed from the nineteenth century onwards are the source of ongoing ethical and pragmatic disagreement. By and large I have chosen to work with, and critique dominant conceptions of the subject and the subjectifying practices that have emerged from those notions. Chapter 4 6
//////////////////
Hollway frames the conception of the individual within occupational assessment as a ‘social technology enabling the administration and regulation of employees’ (28). Within institutional assessment practices it is naïve in her terms to look for a straight forward ‘progress towards truth’ (27). Hollway emphasises the historical motivation within what was then called occupational psychology, to aid organizations with ‘the complex problems of maximizing profitability’ (29). It is important to note the connections between personal psychology and commercial interests, and in my work, to make overt the connections between psychological assessment methodologies and market research practices.Chapter 4, 8.
/////////////////////////////////
Using visual practices as an example Haraway emphasises the corporeality of this sensory system which is undeniably embodied and not as she puts it ‘a gaze from nowhere’ (188). Despite this, vision has been used within Western scientism to somehow signify a transcendental and neutral observer, it has been allowed in other words to ‘represent while escaping representation’ (188). But Haraway is keen to point out that her notion of embodiment and particularity also serves as a metaphor for non-human forms of vision, including vision in its varied technological forms.. Chapter 4, 12
/////////////////////////////////////////
My practice engineers, where possible, an end to representing ‘while escaping representation’. An end, for example, to the power of CCTV cameras to escape such representation, throughout the South software CCTV cameras and images act as proxy narrators, (or narrative agents) bringing to the surface a range of desires, beliefs, agendas and motivations, factors that are suppressed in Western scientism and scientific notions of observer neutrality. Institutional CCTV images are deployed as story-telling mechanisms; this is not represented as a novel form of use for such camera lenses but their typically concealed metier.Chapter 4, 14
/////////////////////////////////Within the South system the apparent isomorphism between one CCTV image and the next is subject to a similar form of generative doubt. The South system automatically seeks out differences between images, these differences are interpreted in narrative terms, readers are invited to provide their own interpretation of what these differences mean; readers are also automatically photographed and the results put into a database. According to data I have uploaded that day, (relating to my own state, a subjective index of mood, reactions, finances and other parameters), these I.D photographs will be subject to a range of image processing effects. Images of readers are also overlaid with time-specific data generated in the moments of their interaction with the South system. The result is a visual database of readers defined in overtly located and specific terms by my own shifting categories and responses. A range of categorisations and apparent isomorphisms can be retrieved by querying the database, but the fact that my own partiality is literally embedded in these images denies representation without being represented, which Haraway describes as ‘the cyclopean self-satiated eye of the master subject’ (191). Chapter 4, 16
/////////////////////////
My own methodology and practice applies a conscious requirement to identify structures and assumptions that rely on the idea of an omniscient, idealized and un-represented, conquering eye. Likewise within the evaluation processes I have used there is an overt indexing to my own perspectives. My aim is to create as high a degree of transparency as possible within the South System. The South software deploys charts, diagrams and disruptive photographic processes that are designed to locate and embed a critical self-representation. Chapter 4, 18
///////////////////////
In this vision of collective subject positions the notion of boundaries between objects and subjects becomes far more mutable than in orthodox scientific epistemologies. ‘Boundaries’ Haraway writes ‘are drawn by mapping practices; ‘objects’ do not pre-exist as such. Objects are boundary projects’ (201). The indices to my own subjective evaluation within the South system (produced as instrumental weightings within the software) are also processual in nature and do not exhibit pre-existent states. They are defined by my own mutability and by the mutability of all the parameters within the evaluation processes, as such I would characterise the system as situated, both in the theoretical terms Donna Haraway describes and in Lucy Suchman’s (1987, 2006) conception of situated systems within computer-human interactions. Chapter 4, 18
/////////////////////////////////////////////////////////////
Though there is contention within the domain of cognitive science as to the exact nature of these interactions, there is, according to Suchman, a general belief that this model of intelligent human agency does not merely have a resemblance or resonance with computational processes, but that it ‘literally is computational’ (37). The notion of human intelligence as tantamount to the management of symbolic representations is exemplified in rule based computation such as expert systems and factory floor robotics. But as Suchman points out, these are narrow domains with a high degree of containment and predictability. Chapter 4, 20
///////////////////////////////
As an alternative to the cognitivist emphasis upon the unitary Cartesian cogito, (and the specific forms of rationality associated with it) Suchman, like Haraway, suggests that we instead pay attention to thespecificities of knowing subjects, multiply and differentially positioned, and variously engaged in reiterative and transformative activities of collective world-making (Suchman 2005: 3). Suchman challenges and inverts the cognitivist idea of similarities between humans and computers and instead asks how we might usefully understand, and instrumentally deploy, the significant differences between machines and people.Chapter 4, 20
///////////////////////////////////////Weizenbaum also observed that these anthropomorphized models reduced and disembodied human processes and that in their reduction to rule based formalisations, even in the case of something as seemingly abstract as grammars (or admissible sentences) there was an irreparable loss of meaning:
They know them (to use Polyani’s word) tacitly, that is, in the same way that people know how to maintain their balance while running. In both speaking and running, by the way, performance once mastered, deteriorates when an attempt is made to apply explicit rules consciously.(Joseph Weizenbaum, 1976:137-138)In other words our knowledge of the word, in this case of grammars, is in practice rarely reducible to a simple set of explicit, context-free rules.Chapter 5, 10
//////////////////////////////// My own work is mindful of these criticisms, but unlike Phoebe Sengers or Andrew Leonard I have been concerned with using some of the embedded cultural artefacts, (if I may speak of the historical and structural signs buried within agent design in such terms) as qualities that can act as protagonists, provocations and re-mediations of subjectivity. The use of medium specific qualities, whether they are problematic, accidental or ‘helpful’ within my work locates my processes back to the continuum of practices engaged upon by the makers of artists’ books, in which, as Katherine Hayles emphasises, the medium specificity of the book form has traditionally been revealed, challenged and acknowledged extensively. South, as I will demonstrate throughout the rest of this chapter, exploits and exposes fragmentation, rather than obliterating it. Chapter 5, 12
////////////////////////
The interweaving of voices, including a robotic proxy voice that represented myself both as the author of the software and the author of the fictional narratives, was an important facet of the interface, one that opened up notions of ‘performing’, or underscoring the form and identity of the book. But the interactions were still essentially static and deterministic, the so-called subject-orientation of the software was top-down, defined within my own fixed data structures and interactive buttons that (albeit unintentionally) epitomised the linear model of cause and effect outlined by Slack and Wise, in which core factors are framed as 'the same under any-and–every-circumstance', (Slack and Wise, 2005:116)Chapter 5, 20////////////////////////////
Though Lent is an extremely simple agent I would differentiate ‘him’ from an even simpler reactive agent, (which reacts in a way that is almost reflexive to its environment) in that Lent maintains an internal state relating to ‘his’ levels of alcohol consumption. Lent is consistent with the requirements for a deliberative agent and with Wooldridge’s (2002) requirements for an intelligent agent, in that ‘he’ or it is:
• Situated – ‘he’ is embedded in an environment • Goal directed – ‘he’ has goals that ‘he’ tries to achieve • Reactive – ‘he’ reacts to changes in ‘his’ environment • Social – ‘he’ can communicate with other agents (including humans)
Chapter 5, 22////////////////////////////////////
Although there are aspects of this proto-type that I have found useful, such as the authoring of a less deterministic deliberative agent, the greater value of creating this proto-type has been in enabling me to identify its weaknesses, and the weaknesses inherent in the conventions I have followed in producing the program. Although I take Michael Mateas’s point that believability is not the same as realism, I would still identify the central weakness of Road as an overtly anthropomorphized construction of an agent entity. Chapter 5, 24
///////////////////////
Why, I asked myself, was I forcing this program to exhibit purely human traits such as cravings, impatience and aggression? What would a program be like that attempted more profoundly to explore the asymmetries of machinic knowledge generation, the way for example, that machines reason and process language, instead of covering up errors and asymmetries of understanding between computers and humans? What would it be like to cultivate those qualities as cultural traits and medium specific, distinct materialities, of the agent medium? Chapter 5, 24
///////////////////////////////
The PlayBots program, which, in its current form, is no more than a proof of concept, clearly has much room for improvement, particularly in terms of situating the agents and giving them more sophisticated goals, beliefs and desires. A long term goal would be to embed this program within two hardware robots and link the speech to environmentally situated physical movements. In the short term, the experience of developing this work has revealed the possibility of an alternative form of believable agent, one that did not mimic a construction of human culture or human-like behaviour, but was medium specific, allowing the particularities of the agent medium to come forward. Chapter 5, 30
//////////////////////////////
Throughout this chapter, and indeed throughout this thesis as a whole, I have emphasised investigating the specific materiality of computational media, asking what are the medium specific strengths (aside from anthropomorphic, zoomorphic or biomorphic projections to paraphrase Penny) of this technology and how can they be meaningfully deployed within my own work? The following section aims to look in more detail at these questions while acknowledging, as Simon Penny puts it, that ‘artistic solutions are often highly contingent and specific to a certain scenario, and may not generalise to general principles for a class of scenarios’ (1999), such generalisation is not the purpose of my work and indeed it would be at odds with my methodological commitment to a contingent, mutable and situated practice to try and generalise the solutions I have created for South. Chapter 5, 32
/////////////////////////////
'Both interactionist AI and GOFAI share research goals that are at odds with the goals of those using AI for cultural production. Artists are concerned with building artefacts that convey complex meanings, often layering meanings, playing with ambiguities and exploring the liminal region between opaque mystery and interpretability. Thus, the purpose of, motivation behind or concept defining any particular AI-based artwork will be an interrelated set of concerns, perhaps not fully explicable without documenting the functioning of the piece itself. ' (Michael Mateas, 2001)While I am wary of adhering to such an emphatic separation of artistic practices from those of AI practitioners, his identification of goal orientation or task competence as the unifying feature of different AI practices is useful, and certainly in relation to my own practice represents a meaningful distinction and a relevant point of contrast from the traditional focus in AI, which Mateas goes on to define as being focused on task competence..
Chapter 5, 33
//////////////////////////////////////////////
The problem therefore is of navigating seemingly conflicting intentions, on the one hand to acknowledge and work with contingency (including cultural contingencies) and on the other to maintain a coherent narrative experience for readers, which might be characterised as in keeping with the more goal oriented orthodoxies of artificial intelligence. Chapter 5, 35
//////////////////////////The end part of this section will show, via a case study or user scenario, how these qualities have specifically been deployed in South and also how the South project stages the relationship of AI practices in relation to the subjective experience of a specific reader, generating activities that investigate how that reader, and the computational agents activated on her behalf, try to act within and upon their respective worlds.Chapter 5, 38
/////////////////////////////
Though I concur with some of the broader points of Leonard’s statement, there are, however, significant structural and epistemological differences between these forms. The South software utilizes and in some ways stages these differences as overt frames of cultural reference, much as one might contrast Cubism with Futurism. Chapter 5, 38
/////////////////////////
A central facet of expert or problem solving systems, and indeed of all explicitly programmed knowledge representation schemas deployed in AI, is the notion of brittleness, domain dependence or the fact there is a limit to the extent and generalizability of knowledge contained within these systems. Users of such systems cannot be sure of the limitations embedded within a knowledge base or rule set, and therefore cannot fully trust the answers that a system provides. In South the extent of its own brittleness becomes a form of both narrative and embodied adventure. Readers are invited to explore the extent of the system’s brittleness and the absurdities that may arise from its inherently incomplete knowledge. Chapter 5, 39
/////////////////////////
By inviting readers to test the boundaries of the South system in both its expert and agent bases, the notion of disembodied expertise is also critically challenged, particularly in relation to the concept of personality. Continuous sensory inputs, in the form of CCTV images, reader interactions, temperature and other inputs or percepts, open the system up to a more situated, contextually aware form of knowledge generation and, as outlined in the user scenario below, to broader notions of agency than those found in most expert systems of the 1980s. In describing the broader agencies at play in South I am not intending to idealise the South system or to make bogus claims for its superior efficacy in understanding human subjects and human situations, I am instead emphasising the materialities at play within this system and suggesting the significantly creative ways in which artificial intelligence and agent technologies can be deployed and re-conceptualized by artist practitioners.Chapter 5, 40
////////////////////////////
a map in the South book works in conjunction with the egg content, the software has an internal model of the map and can therefore make meaningful instructions for readers to put themselves in particular locations:
Chapter 5, 55
/////////////////////////
In working with agents South has also emphasised cultural situatedness as well as the physical, sensory or environmental situatedness conventionally implied by computer scientists when they refer to ‘situated agents’. By situating themselves in a dynamic network of environmental, cultural and inter-subjective forces, and thus establishing a less linear foundation for interactivity, the agents deployed in South are able to destabilize and meaningfully interrogate its core themes of subjectivity, epistemology and agency. The reasons for engaging in such an interrogation are both pragmatic and ethical.
Chapter 5, 58
////////////////////////////
My rationale in looking for alternative subjectivities, epistemologies and agentive distributions has been rooted in the premise that electronic literature is in a significant crisis (as outlined in the introduction to this thesis) and that this crisis is symptomatic of electronic literature’s provenance in rigidly symbolic, a priori knowledge representations. These rigidly top-down and pre-determined representations have resulted in stilted and predictable works that do not deploy contemporary, networked computation optimally. At the same time, broader, more relational framings of agential and epistemological processes are part of a methodological commitment to a multiplicity of perspectiveChapter 5, 58
////////////////////////////////
Key outcomes and inferences:
• Medium specificity also applies to AI practices: Such practices are specific mediums, not unlike dramatis personae with their own unique traits, capabilities and flaws. Many of these qualities would, in anthropomorphic terms, be framed as ‘mal-adaptations’ by agents to conditions beyond their ‘ecological niche’. But such mal-adaptation is arguably an inherent feature of all agents, whether human or non-human, and as such a potentially creative resource.• South’s Agents are also proxy researchers: The agents embedded in South engage in research on my behalf, we might therefore be described a structurally coupled.• South agents are non neutral: South and its deliberative agents deny the assumed objectivity of the Cartesian ‘man of reason’; the embedding of non-neutral values is revealed via post-hoc explanation systems and access (where they exist) to knowledge bases.• The South system is supported by an ethical methodology: South aims to establish a plurality of practices and to accommodate a similar plurality of agencies (these will be explained in the next chapter).• A ‘situated agent’ is also a culturally situated agent: As Mateas and Sengers have argued (2003) AI systems are cultural artefacts. They also argue that agents perform on behalf of their authors; these ideas are investigated throughout South.• South’s subjects are framed as environments: There is a meaningful degree of what might be called structural coupling or mutual influence, between the South agents and their environments (readers), but this mutuality is intended to extend beyond the binary, linear relationship, implied by the word coupling.• South’s agents are allowed to be graceless: Agents generate subjective weightings upon each new execution of the program, supported by situated and ‘sensory’ inputs that can alter the agent’s evaluation of the reader, this has some resemblance to simplified machine learning algorithms, in which agents observe situations and interactions and attempt to detect patterns in order to act upon them. Janet Finlay and Alan Dix write that ‘issues of control and grace of interaction between agent and user are common to any system that involves learning user actions’ (Janet Finlay and Alan Dix, 1996: 233). However, South is also interested in contesting orthodox notions of interactivity, and thus of exploring what a ‘graceless’ system might be like, whether in fact there are power struggles or creative asymmetries at play. • South’s agents are embodied: But to talk of embodiment in relation to wholly software agents is complex and contested.
Chapter 5, 59/////////////////////
I would therefore venture to define these agents as both physically and artificially embodied by virtue of a meaningfully dynamic relationship to their environment. Computer scientists might characterise this as a form of structural coupling. Such a structural coupling, however, as outlined for example by Maturana and Varela (1987), suggests a loop of agencies, in which the artificial agent effects the environment (a reader) and the environment, or human agent effects, in turn, the artificial agent. The idealised conception of the South system aims to be looser, less linear then the notion of structural coupling or loops of cause and effect might imply. South attempts to open its structural relationships to a wide and distributed network of agencies, including human agents, such as readers and myself, artificial agents and other types of non-human or hybrid agencies, from forces of nature to economic forces that have complex agential foundations.
Chapter 5, 60
///////////////////////////////////
“Mašinka, you need to look after your egg”
The agents that find and generate data are contained in a program called Web Gather, Mašinka is encouraged by the evaluative software to maintain this program by ‘looking after it’, this means regularly executing the program so it can find and analyse data. The model for this comes from Japanese Tamagotchis, or handheld digital pets. Tamago, the root of this word comes from ‘egg’. The main character in South, Ivan Dâr is an oologist, or egg expert. Mašinka is also encouraged to regularly update the egg_modes_film program; to keep a record of how she looks each time she interacts.
Chapter 5, 47
////////////////////////

Thursday 24 December 2009

Left, a mock contents page from my artist's book South. The chart traces the associations generated by the word ’book’. The example destabilizes canonical notions of linear and logical visual hierarchies as navigational aids for readers. Though the chart is still hierarchical it is arguably not spatially indicative of where readers can find further information, it is more like a map of undiscovered territories than a useful representation of significant divisions within the book. Disorientation is a key theme within the fictional writing and algorithmic procedures of South.

Wednesday 23 December 2009

Editing the whole document now and storing here what seem to be core passages and references, as a sort of condensed crammer for my own thesis


see 27/12/09
links to some code experiments:
change style site (sketch)
chageable site (sketch)
saves changes to php:
cookie goes to php page

Monday 21 December 2009



I've just finished the first draft of my introduction, which means I've finished the whole thesis - at least in draft form. I've pasted in an excerpt from the introduction to the introduction, I expect it will be dismantled and condensed even further, though at 3043 words it was (for a change) pretty spot on with my intended target of 3000 words. Now I'm wondering if that is too short? There's always a neurotic compulsion to adhere to some mystical true thesis form, or ludicrous normative model...I hope its as long as it needs to be, which is as long as a piece of very long string, and hope that's ok..


Excerpt - strange how reading it on the web makes me read it more critically then reading it in .doc format on my own computer:


South, a Psychometric Text Adventure is an artists’ book and a set of software programs that re-conceptualise the artists’ book form. The South project represents a significant overlap between the artists’ book and literary works, hence the allusion to both forms throughout this thesis. In relation to both artistic and literary forms the project fosters a creative sensitivity to the temporally and socially entangled agencies that are always at play, but often subsumed, in complex systems of human-computer communication. South in both its analogue and digital book forms is designed to work with a physical location, the South Bank area of South London, but it is also designed to work with the subjectivity and ‘personality traits’ of individual readers, and to wider situating forces.

The software and book represent a practice-based hypothesis that subjectively and environmentally situated intra-active software can enliven digital literary works and artists’ books while significantly developing previous notions of ‘the interactive’. The software also presents the case for bespoke works while acknowledging and nurturing collective meanings and shared experiences. Such works challenge linear, humanist, conceptions of agency that might characterise the ‘bespoke’ as a solipsistic and individualist construction.
The South project evolved in relation to the continuum of my practice as a writer and fine artist engaged with making artists’ books. The project was also developed in the context of a critical examination of previous digital literary works. Throughout this thesis I propose that such works have not lead to the death of the book and that digital literature has not met the hyped expectations proclaimed by some commentators in the late 1980s and 1990s. Similarly, within the fine arts the artists’ book has, with a few exceptions, failed to engage significantly with computation. This thesis argues the case for these statements and for a material engagement with digital technology, and more specifically, for an engagement with interactive programming that extends the philosophical and critical involvement many artists have historically exercised in relation to the book form.
This thesis links the absence of significantly computational artists’ books to the putative ‘failure’ of digital literary works (again, acknowledging some important exceptions). It also associates the lack of significant material or processual impact on literary and artists’ books to a lack of critical engagement with both the philosophical meaning and material capacities of computers and ‘interactivity’. The rationale for my approach is rooted in a critical evaluation of key historic and contemporary digital literary works, for example, Talan Memmott’s From Lexia to Perplexia (2000). This work is notable for its engagement with computation beyond, as Katherine N. Hayles puts, it reducing computers to a matter of ‘hardware and software’(Hayles, 2000), however, I am weary of falling in with technologically deterministic outlooks on human-computer-interactions. Such outlooks are in danger of blankly suggesting that ‘computation is fundamentally altering the ways in which humans conceive of themselves and their relations to others’ (Hayles, 2000). Instead I argue for a more nuanced understanding of human-computer relationships, one that does not presuppose a discontinuity in the conceptual foundations of programming and computers from other cultural and philosophical artefacts. My thesis frames computational constructions within a historical continuum, in which both the subject and the subject’s generation of knowledge are linked to enlightenment and positivist philosophical positions, and therefore to wider cultural and historical movements. At the same time I have sought to confront or re-frame the separation between computers and humans, or indeed the ready made separations that we project between subjects and objects (including readers and books). An important aspect of my work has been to identify significant features of computational knowledge generation, while acknowledging that computers are not clearly separable from ourselves, but, like all human artefacts, are of us. This methodological position is supported by writers such as Donna Haraway (1991), Henri Bergson (1896, 1907), Rosi Braidotti (2006) Karen Barad (2007) and Lucy Suchman (1987, 2005, 2006). These writers shore up the proposition that human beings are entangled with their technologies and with complex, relational and temporally bound systems of agency.

Purpose of the research
This research was conducted in order to support the generation of significant new forms of digital and artists’ books; the research was funded by a full studentship from the Engineering and Physical Sciences Research Council. The research critically evaluated and examined the status and role of digital literature, artists’ books and the broader area of digital interactivity. My research explored the extent to which a priori structures should define computational literary works and wider conceptions of the interactive. The role, benefits and drawbacks of using the subjectivity and situatedness of individual readers was also investigated. Likewise the expectations and opinions of potential readers in relation to interactive works were also solicited and constructively embedded within the research design.
The research specifically highlights the importance of engaging with the cultural and philosophical significance of programming practices. Such an engagement enables us to recognise significant differences of kind between computational and analogue forms as well as meaningful differences in degree. This recognition can help us to focus on the strengths computation has to offer as well as protecting practitioners and theorists from making exaggerated claims for digital works. The results of this research are likely to benefit both the producers of interactive works and their users. This claim is supported by a number of new projects and collaborations I am involved in. These individual projects and collaborations deploy the methodological and technical strengths generated by this project, encompassing diverse areas from film-making, site specific performance, curation, website design, teaching and collaborative fiction writing.

Sunday 20 December 2009

Catching up on PHP & JQuery skills this weekend for another project. I've also re-phrased my thesis question: it should now be:
Is it possible to create an intra-active artists’ book that is subjectively and environmentally situated? It just makes more sense in light of the whole document.

Saturday 19 December 2009


This Christmas break is an opportunity to get immersed in finishing my final PhD chapter - in time honoured fashion - the introduction. Now that I am supposed to fully grasp what it is I am introducing. Then the really challenging bit, which is to make the whole thing into a coherent whole. I'm living on crackers and mince pies, which will run out anyday now, what with being virtually snowed in (yes there are at least 3 millimetres of ice on my front garden path), and buckets more on the way. I think I'm going to miss it though. I've really enjoyed being able to immerse myself in answering one question over three years! The biscuit is for Jane Skead - enjoy...

Wednesday 16 December 2009

The other journal was so huge it was getting a bit slow to edit, so I've relocated here for a smart new, rapid future. At the moment I'm working on a device for beautiful thinking as part of my Goldsmiths transfer and incubation proposal, the idea is to make recidivist students into fully functioning members of society, or something like that.
The old journal is still there, frozen in time >>>> http://www.dareinteractive.com/blog2.html

My prototype Eunoia Machine, audio book and heuristic device..December 16th 2009