This project is the result of reflections I have made over the course of the last four months, working as a bicycle courier for the restaurant delivery company Deliveroo. The focus is in the broadest sense an interrogation of how technology mediates, and is imbricated in our experience of space. Specifically through my experiences I will touch primarily on the theme of control. Firstly the project will validate, through an analysis of how the schism in approaches to the notion of palimpsest may be resolved by considering Deleuze’s work on time and the virtual, my extension of the notion (following Graham, 2010) to include the technological virtual. Much as the virtual past is variously actualised in the present to interpret a palimpsest, as a non-fixed text, so locative technologies make this relation even more clear, requiring the participation of users as the dynamic virtual is variously actualised in different configurations through user experience and input (de Souza e Silva and Sutko, 2011; Poster, 2001). In an age of the ‘ubiquity and pervasivity’ of location based technologies in everyday life (Pellegrino, 2006, p.133), their study may be able to shed light on, and ‘signal important shifts’ in our present understanding of London through the changes in the ‘political economy, consumption, and regulation of new media services.’ (Wilken, 2012, p.246).
Thoughts on Palimpsests and History
The metaphor of palimpsest has been an enduring one in the field of urban studies, drawing attention to the effects of time on the urban landscape, through allusion to the many layered-ness of the city. The notion can, and has been interpreted in various ways, each with its own implication and theorisation of time and memory.
A palimpsest in its specific sense, is an artefact which is the product of multiple writings and overwritings, as the parchment or vellum on which the writings were inscribed is reused, erasing the previous layer(s) sufficiently so as to make room for the next, but doing so imperfectly, and so traces of what has been written before are retained (Crang, 1996; Carlylye, 2002; De Quincey, 2003; Dillon, 2005). We can see, following de Groote (2014) that there exists a confrontation in the notion of the palimpsest, however I wish to argue that far from being ‘irresolvable’, though they may represent two fundamentally different attitudes towards time, memory and history, through the work of Benjamin and Deleuze we can move towards a more productive notion of palimpsest.
Lauded as the inaugurator of the concept of palimpsest (Dillon, 2005), De Quincey uses palimpsest as a metaphor for memory, proclaiming ‘what else than a natural and mighty palimpsest is the human brain?’ (2003, p.175). De Quincey’s palimpsest is ‘palimpsestuous’, finding neat chronological sequences through its metaphor. ‘Each succession has seemed to bury all that went before. And yet in reality not one has been extinguished’ (ibid.) he writes, thus this metaphorical palimpsest is well able to describe the ‘accretion of historical events and processes’ constituting an urban landscape (Crang, 1996, p.443). In contrast, the palimpsest of Carlyle, which actually predates De Quincey’s usage by some 13 years, proceeds from the material artefact of palimpsest to find fragments of the past, a ‘memorial to compounded losses’ (de Groote, 2014, p.111). The reason, then, for these alternate understandings of palimpsest is that they proceed from fundamentally different perspectives. One considers the process of layering a palimpsest through metaphor, the other the material reality. They are fundamentally incompossible, in De Quincey the past is recognised as fleeting, but this is reversed into the ‘certainty of eternal remembrance’ (de Groote, 2014, p.112), reified to a set of fixed facts, whilst for Carlyle man’s ambitions for eternity are reversed, the past as transitory. A palimpsest cannot be read from a palimpsestuous perspective whilst also read as palimpsestic. The reason is a central contradiction of history, that there is ‘an uncertainty between the determination of a fact and the construction of its history’ (Berg, 2006, p.5). The palimpsestuous reader will ‘attempt to curb the dreadful actions of time… establishing metaphorical connections and circuits’ (de Groote, 2014, p.123), and in doing so constructs a conformist historiography. A palimpsestic reading conversely resists this desire for ‘origin, influence or filiation’ (Dillon, 2011, p.85), a focus on the erasure which is an essential component of a palimpsest meaning that the traces of the past in the present and the potential of pasts lost can be included together. The challenge for a palimpsestic reading however is to continue to refuse a fixation, but without being arbitrary, to avoid a teleological concept of change (Berg, 2006). It is at this juncture that we introduce the ideas of Walter Benjamin on history.
For Benjamin the past is always understood in terms of the concerns of the present day. ‘The past can be seized only as an image that flashes up at the moment of its recognisability and is never seen again’ writes Benjamin in On the Concept of History (2003, p.390). It is a confusion of the past and of time which complicates accounts of history and of palimpsest which I hope to clarify. The constructive act of interpreting evidence in the present as facts about the past implies a ‘superposition of time’ (Berg, 2006, p.4). ‘To articulate the past historically does not mean to recognize it “the way it really was”… it means to seize hold of a memory as it flashes up at a moment of danger’ (Benjamin, 2006, p.392).
This way of thinking is given increased clarity in my opinion by the work of Deleuze which demonstrates a remarkable accord with that of Benjamin. In his works on cinema, Deleuze distinguishes two main categories of image, the movement-image and the time-image (1986, 1989). The first conception is an understanding of time as continuous, as the ‘movement-image’, which maintains a linear trajectory, is chronological, part of a chain of causality, so that it is entirely possible to identify a first cause and an origin. This linear trajectory of the movement-image is presumed to be true, evolving in one specific way. In opposition to the movement-image is the time-image, which is derived from combining the Deleuzo-Guattarian ‘rhizome’, a labyrinth without a centre, to the Bergsonian concept of duration (Martin-Jones, 2006), effectively amounting to multiple virtual parallel universes. Time is considered as a virtual whole, which is constantly in the process of becoming actual, ‘as a labyrinth in the process of becoming a line’ (ibid., p.24). Here, we can begin to grasp the importance of the two images of time. The movement image is actualised time thus giving the impression of there being just one path through the labyrinth of the virtual time-image. The rhizomatic labyrinthine nature of the time-image means that there exist two ‘incompossible presents’ in every situation, which always exist virtually, but only become actualised depending on their alignment with the present.
The ‘oscillation’ between the actual and the virtual is at the core of a critical reading of palimpsest. The two palimpsests as the two times, do not exist independently of each other, ‘rather they are different manifestations of the same time’ (Martin-Jones, 2006, p.24). Therefore I propose to use the notion of palimpsest put forward by Eiland (2010) in which he suggests that the notion refers the idea of interpreting reality as a script, but specifically as a many layered and essentially ambiguous script whose surface writing partly reveals and partly conceals a virtual expanse of stratified underwritings. We can see here the past as virtual, labyrinthine, and discontinuous, and palimpsests as able to give rise to many different histories without one being ‘correct’.
Virtuality and locative media
The notion of palimpsest has traditionally been used to evoke the multitude of present and primarily past discursive and physical layers used to interpret place, however as Graham (2010, p.423) notes, ‘with the advent of the Internet, an entirely new dimension of layers has begun to be added to the palimpsest of place’. Akin to the way that that the past is always brought into the context of the present moment, through technology our relationship with digital and physical space is mediated, on an individual level requiring participation in constructing the world which is experienced (Poster, 2001; de Souza e Silva and Sutko, 2011). The notion of palimpsest is here extended, considering the virtual as potential and as part of the real.
Graham suggests that there is ‘often a reflexive relationship’ (2010, p.423) between digital representations and their physical counterparts, and whilst these digital representations were tied to access through desktop computers, the theorisation of the virtual which it implies, one as a simulation of the real, remained tenable, however in an age of location aware technologies it is clearly inadequate.
Where the theorisations of the virtual as simulation or copy break down, the virtual as potential excels. It is able to cope with the ways in which location based applications ‘interface between physical space and digital information’ (de Souza e Silva and Sutko, 2011, p.33) [For full explication, please refer to Appendix I]. There is in the contemporary world, an increasing interpenetration of material objects by patterns of information. The virtual as potential, as becoming, is therefore much more applicable, as location based applications actualise connections between the physical environment and users (ibid.). We are all (like Benjamin) immersing ourselves in reality as in a palimpsest (pace Adorno, 1955), through the usage of location aware technologies, as space, digital information and users are ‘processually bound up’ (Hayles, 1999 in de Souza e Silva, 2011, p.37).
Having argued in general terms that we should understand our experience of reality as a palimpsest, that a core component of this reality is the virtual, with this virtuality as ‘the cultural perception that material objects are interpenetrated by information patterns’ (Hayles, 2000, p.69), and that locative technologies make this relation between information and materiality abundantly clear, let us turn now to how this assemblage of device, application and user informs, expands and limits the other’s capacities (ibid.; de Souza e Silva, 2011), reflecting on my personal experience working for Deliveroo.
Deliveroo is a food-delivery start up premised on the idea of bringing restaurant quality food to customers in their homes or at their workplaces. Focused on the premium end of the market, they partner with a network of restaurants, many of whom would otherwise not offer a takeaway service, and allow them to expand their customer base without increasing the number of covers in-house. One of the central aspects of Deliveroo’s business model is that it directly employs ‘drivers’ (motorcycle and bicycle) to add a consistency to the process. Deliveroo is projected to soon join the ranks of other ‘unicorn’ companies, worth more than $1 billion dollars (FT, 2015). In recent years many of the start-ups which have experienced such a meteoric rise have been pioneers of the ‘gig’ or ‘sharing’ economy. Comprised of companies such as Uber, Lyft and Airbnb, many, like Deliveroo, are app-based, and rely on a workforce of independent contractors, rather than full employees (Rutkin, 2015).
In this ‘gig economy’, there is a new articulation of organisational control through technology, which expresses itself in especially spatial ways for those workers whose jobs require mobility. In order to demonstrate and question this control, let us consider the notion of assemblage in relation to Deliveroo. Assemblages can be defined as ‘processes of on-going becoming in order to achieve certain accomplishments’ of various duration (Introna, 2013, p.333). Hence, in the context of Deliveroo, I want to sketch an ideal type, a ‘Roo’, an unattainable standard to which all workers are compared. A Roo is a heterogeneous assemblage of driver, phone, ‘Driveroo’ application software, branded clothing, bicycle/motorcycle, etc., which perfectly accomplishes the tasks set. A Roo, following Nietzsche, has no existence beside its ‘doings’, since ‘there is no being behind doing, acting, becoming… the doing [becoming] itself is everything’ (1996, p.29 in Introna, 2013).
Contrary to this ideal type, an average, fallible, human worker has many becomings. The worker is an assemblage of the same components as a ‘Roo’, though their effecting of commands varies in efficiency, for a number of reasons. The central conflict in working as a Deliveroo driver is between the more dynamic aspects of the assemblage, the software based on a matching algorithm, and the human enaction of the software’s instruction. In the literature on algorithms, one of the major themes in recent years has been their power and agency (Neyland and Möllers, 2016).
As Lash (2007, p.71) argues, ‘power is increasingly in the algorithm’, and when working, my behaviour is presumed to be controlled (Spring, 2011). Let me outline here the functioning of the `application and ways in which it may be considered to have agency and power. My ‘Driveroo’ application will not allow me to login until I am inside of the central zone, denoted by the blue circle in Figure 2, and instructs me to move to the centre of the zone at all times while waiting for a delivery. A delivery will be allocated according to the algorithm, and I then ride to the restaurant, pick-up the food, alert the application, and then cycle to the customer, depositing the food, and in declaring the order delivered, am made available for the next order to be assigned (see Figure 3).
Algorithms are ‘central to the distribution of consequences’ in our society (Neyland and Möllers, 2016, p.2), and yet often ‘technically and intellectually inaccessible’ (Wagenknecht et al., 2016, p.537). In the context of Deliveroo there is no incentive for the company to disclose to its employees how its algorithm for allocating orders works, rather the new recruits are given an extremely cursory explanation of how their application functions, and told to always move back towards the centre of the zone in order to be allocated a delivery, which for East Central (EC), where I work, is the surroundings of Spitalfields Market (see Figure 2). In a software update on the 5th April 2016, this was formalised, drivers explicitly reminded when they don’t have an order to wait in the centre of the zone, rather than it being more of a norm than a rule (Figure 4).
What this serves to do is remind us that however much power we ascribe to algorithms, they are but one component of an assemblage, and do crucially rely on human work (Wagenknecht et al., 2016). In contrast to the ‘Roo’ outlined earlier, which would carry out the deliveries as quickly and efficiently as possible, humans have various preferences and employ strategies in order to accomplish aims which are not necessarily commensurate with those of the company.
With such incomplete knowledge of the allocation algorithm and proximity/closeness of any surveillance and discipline mechanism, initially I stuck very tightly to the instructions given, and my prime concern was being as good as possible at my job. This meant riding as quickly as possible, and always returning to wait in the centre of the zone. Over time however, as my knowledge increased through ‘sense-making’ I developed a number of ‘strategic workarounds’ (ibid., p.539). My intuition of the algorithm is that it weights distance to restaurant and driver speed in order to compute the best driver allocation. I came to this conclusion through my accumulated experiences of receiving on average orders close to my location, and because that is intuitively efficient. It also became clear through talking to other Deliveroo drivers that I was receiving more orders than many of them on less busy evenings. This, I believe, is due to the algorithm being composed at least in part of an average speed of delivery component, as I tend to rider somewhat faster than many other bicycle drivers.
Over time my knowledge of the extent (read lack) of scrutiny of individual drivers also increased. There has to date been absolutely no performance related feedback; no metric, no incentive, nothing, and so the only feedback I receive is purely monetary. The wage earned is comprised of an hourly rate of £7, and then a subsequent £1 per delivery, and any tips. This breakdown, and lack of stimulus to work hard, led me to target certain restaurants which are expensive, and thus more likely (in my experience) to give tips. Viewing my location history data requested from Google, I have been able to map my movements over time as I have employed these strategies, to give them some (limited) empirical support. Figure 5 shows 5 maps, each plotting a heatmap of my movements. Initially, in the first two we can see a close adherence to the advice to stay in the central zone, however as my knowledge improves, we can see a drift, towards certain restaurants further from the central zone, such as Clutch chicken, Busaba Eathai and Bodean’s. In the latest maps we can see a much greater distribution in location, as when a delivery takes me away from the central zone, I will often ride back towards the centre in the direction of a favoured restaurant and loiter in the vicinity. This is a strategy which is far less effort than the highly efficient ‘Roo’-like one. It allows for other activates to be undertaken in the time which would otherwise be spent waiting for another allocation or in cycling, singular-mindedly towards the centre of the zone. I am able to sit, on pleasant evenings, and read a book, write, or merely explore the city (see Figure 6).
Coming back to the question of power and the assemblage, let us draw some conclusions. Following Wagenknecht et al. (2016), we may consider algorithms as ‘iterative socio-technical performance’ (p.538). Were they to have the power some authors claim, (pace Spring, 2011 in Neyland and Mollers), ‘algorithms trap individuals and control their lives’, then my experience of
Deliveroo would be one where I adhered as tightly as possible to the ideal ‘Roo’; as it is the agency is much more mine than it is the application or the algorithm. Considering the work as the prolonged arm of computation, and as performance, we can assert that as such it is normatively governed (Introna, 2013). My becoming-roo is governed by prior performative outcomes we might suggest. The way in which the heterogeneous elements of the assemblage are associated by my agency belies a choice; ‘to extend further, risking dissidence and dissociation, or to reinforce consistency and durability, but not go too far’ (Latour, 1988, p.198). Gradually through iterative risk-taking I have established as an individual norm a situation where I locate myself more strategically, and use the ‘dead time’ between orders or while waiting for a restaurant to cook the food to pursue other endeavours. These are lines of flight within the apparatus of capture and control of the algorithm (cf. Deleuze and Guattari, 1987, p.225). They do not ‘go too far’, but offer new modes of existence contrary to the dominant way of being. Within the constraints of the application there is room for manoeuvre, for the exercising of individual agency, but only so long as this manoeuvre is within the realm of reasonable action. For instance, there is really little difference between my obstinacy in not moving to the centre to the zone, and forcing the application to close when I have a delivery which has been assigned to me, but which I don’t wish to carry out. The difference is that one (the assignment of deliveries) is given much more weight in my mind, and so while I may wish to exit the app (as has happened on many an occasion, see Figure 7), I do not, and continue to work far later than I wished to. The fear of ramifications for doing so keeps me in check, and so there are certain boundaries which despite the performative nature of the assemblage, I remain within.
Through my experience of working for Deliveroo I have had extensive exposure to an algorithm and an application which are location aware, and been able to reflect on the manner in which they have shaped my actions. Whilst this was specifically in a workplace context, I believe that many of the findings are of general importance.
Firstly, our relationship with space is becoming increasingly mediated as location aware applications are becoming a ubiquitous and essential component of everyday life for their users. The situation of a cyborgic assemblage of user and device able to interface with the virtual information embedded in space is a [more] accurate description of interaction with locative media today (Hayles, 2000; de Souza e Silva and Sutko, 2011).
As these applications which allow us to interface with the virtual are by and large informed by algorithms we may think of them on the one hand as a kind of performance, and also as infrastructure (Wagenknecht et al. 2016). Through a specific application and concomitant performance we have been able to consider the question of agency in relation to the control that apps and algorithms may be considered to hold. As performance we have seen that our interaction with applications is normatively governed and malleable, adherence to the internal ‘rules’ of the application predicated on our understanding of what is possible within its function. As infrastructure, the unseen ways in which algorithms are changing the experience of space through sorting the information which becomes actualised within them may be a cause for concern. My work for Deliveroo is a demonstrative case, as at every moment in time I have a set of attributes which are matched via an algorithm and provide me with a binary response, I have been allocated an order, or I haven’t. This matching based on a set of attributes determined the visibility of the virtual, it determines what becomes actualised. This principle is equally valid when applied to applications with greater complexity. Consider the case of Google Maps/Places (cf. Barreneche, 2012). As individuals we build up a locational profile, and based on that (and other information Google may have access to) a sorting algorithm determines the visibility when one were to, for instance, search ‘pubs’ according to this profile. As location aware applications are now so pervasively used, we should be reminded that they too are not neutral.
Finally, however, we should recognise that the virtual is, after all, potential. As such, while it is constrained within certain boundaries in certain applications, it does provide the possibility, the potential, for many new kinds of urban practice. The imbrication of information and materiality is clearer now than ever before, and while we have argued that history should be understood in an analogous manner, the way in which reality appears as palimpsest is most vivid from the perspective of the technological virtual.
Adorno, T. (1988). Introduction to Benjamin’s Schriften. In: G. Smith, ed., On Walter Benjamin: Critical Essays and Reflections, 1st ed. Cambridge: MA: MIT University Press.
Barreneche, C. (2012). Governing the geocoded world: Environmentality and the politics of location platforms. Convergence: The International Journal of Research into New Media Technologies, 18(3), pp.331-351.
Baudrillard, J. (1994). Simulacra and simulation. Ann Arbor: University of Michigan Press.
Benjamin, W., Eiland, H. and Jennings, M. (2006). Walter Benjamin. Cambridge, Mass.: Belknap Press of Harvard University Press.
Benjamin, W., Tiedemann, R. and Schweppenhä user, H. (1972). Gesammelte Schriften. Frankfurt am Main: Suhrkamp.
Berg, O. (2006). Benjamin und Deleuze: Ansätze für eine kritische Geschichtswissenschaft in Filmbildern. Zeitschrift für kritische Theorie 12, (22-23), pp.68-98.
Carlyle, T. and Vanden Bossche, C. (2002). Historical essays. Berkeley [Calif.]: University of California Press.
Chesher, C. (2012). Navigating sociotechnical spaces: Comparing computer games and sat navs as digital spatial media. Convergence: The International Journal of Research into New Media Technologies, 18(3), pp.315-330.
Crang, M. (1996). Envisioning urban histories: Bristol as palimpsest, postcards, and snapshots. Environ. Plann. A, 28(3), pp.429-452.
de Groote, B. (2014). The Palimpsest as a Double Structure of Memory. Orbis Litterarum, 69(2), pp.108-133.
De Quincey, T. and Milligan, B. (2003). Confessions of an English opium-eater and other writings. London: Penguin Books.
de Souza e Silva, A. (2006). From Cyber to Hybrid: Mobile Technologies as Interfaces of Hybrid Spaces. Space and Culture, 9(3), pp.261-278.
de Souza e Silva, A. and Sutko, D. (2011). Theorizing Locative Technologies Through Philosophies of the Virtual. Communication Theory, 21(1), pp.23-42.
Deleuze, G. (1986). Cinema. Minneapolis: University of Minnesota.
Deleuze, G. (1989). Cinema 2: the time image. London: Athlone.
Deleuze, G. and Guattari, F. (1987). A thousand plateaus. Minneapolis: University of Minnesota Press.
Dillon, S. (2005). Reinscribing De Quincey’s palimpsest: the significance of the palimpsest in contemporary literary and cultural studies. Textual Practice, 19(3), pp.243-263.
Dillon, S. (2007). The palimpsest. London: Continuum.
Drakopoulou, S. (2010) A Moment of Experimentation: Spatial Practice and Representation of Space as Narrative Elements in Location-Based Games. Aether: The Journal of Media Geography, 6(1).
Eiland, H. (2010). REALITY AS PALIMPSEST: ON BENJAMIN’S ARCADES PROJECT. [podcast] The Program in Critical Theory, UC Berkeley. Available at: http://criticaltheory.berkeley.edu/events/event/reality-as-palimpsest-on-benjamins-arcades-project/ [Accessed 1 Apr. 2016].
Financial Times. (2016). Takeaway start-up Deliveroo nets $100m in latest fundraising – FT.com. [online] Available at: http://www.ft.com/cms/s/0/319413e0-8eee-11e5-8be4-3506bf20cc2b.html [Accessed 10 Apr. 2016].
Graham, M. (2010). Neogeography and the palimpsests of place: web 2.0 and the construction of a virtual earth. Tijdschrift voor economische en sociale geografie, 101(4), pp.422-436.
Hayles, K. (1999). How we became posthuman. Chicago, Ill.: University of Chicago Press.
Hayles, K. (2000). The Condition of Virtuality. In: P. Lunenfeld, ed., The digital dialectic: New essays on newmedia, 1st ed. Cambridge, MA: MIT University Press.
Jennings, M. and Eiland, H. (2014). Walter Benjamin: A Critical Life. Harvard University Press.
Lash, S. (2007). Power after Hegemony: Cultural Studies in Mutation?. Theory, Culture & Society, 24(3), pp.55-78.
Latour, B. (1988). The pasteurization of France. Cambridge, Mass.: Harvard University Press.
Lee, M., Kusbit, D., Metsky, E. and Dabbish, L. (2015). Working with Machines. Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems – CHI ’15.
Leibniz, G. and Morris, M. (1934). The philosophical writings of Leibniz. London: J.M. Dent & Sons, Ltd.
Leszczynski, A. (2014). Spatial media/tion. Progress in Human Geography, 39(6), pp.729-751.
Martin-Jones, D. (2006). Deleuze, cinema and national identity. Edinburgh: Edinburgh University Press.
McDonagh, J. (1987). Writings on the mind: Thomas De quincey and the importance of the palimpsest in nineteenth century thought. Prose Studies, 10(2), pp.207-224.
Neyland, D. and Möllers, N. (2016). Algorithmic IF … THEN rules and the conditions and consequences of power. Information, Communication & Society, pp.1-18.
Nietzsche, F. and Smith, D. (1996). On the genealogy of morals. Oxford: Oxford University Press.
Pellegrino, G. (2006). Ubiquity and Pervasivity: On the Technological Mediation of (Mobile) Everyday Life. In: J. Berleur, M. Numinen and J. Impagliazzo, ed., IFIP International Federation for Information Processing, Volume 223, Social Informatics: An Information Society for All?, 1st ed. Boston: Springer, pp.133-44.
Poster, M. (2001). What’s the matter with the Internet?. Minneapolis: University of Minnesota Press. Rutkin, A. (2015). App comrades unite. New Scientist, 228(3051), pp.20-21.
Turkle, S. (1995). Life on the screen. New York: Simon & Schuster.
Vaujany, F. and Mitev, N. (2013). Materiality and space. [Basingstoke]: Palgrave Macmillan.
Wilken, R. (2012). Locative media: From specialized preoccupation to mainstream fascination.Convergence: The International Journal of Research into New Media Technologies, 18(3), pp.243-247.
Appendix I – Theories of the Virtual
Allow us to expand on this point to more thoroughly consider the utility of various theories of reality and the virtual in exploring the ‘increased imbrication of information, technology, space and experience’ (de Souza e Silva and Sutko, 2011, p.31).
The idea of virtual reality as a representation, a simulation or illusion of reality was the dominant paradigm in the 1980s and 90s, as it was conceived of a being a futuristic technological curiosity, a ‘metaphorical space that arises only through interaction with the computer, which people navigate by using special hardware’ (Turkle, 1995, p.181). This position is predicated on a Platonic or Baudrillardian conception of the virtual. In the first instance, Plato considers reality in a hierarchical model of worlds; ideal, sensible and represented. The virtual in a Platonic schema has less value than the real, as it is but a copy, a representation of sorts. In the case of mapping software, they retain far less information than their real counterpart, and are inferior copies of the sensible (de Souza e Silva and Sutko, 2011). Crucially this Platonic logic ‘ignores the dynamic connection between the physical location of a place and the information available’ to be interfaced with (ibid., p.29). Baudrillard’s model of simulation and simulacra allows for interaction of physical space and information. Location aware technologies are able to ‘overshadow’ the referent with their simulation of space, as the trope of car GPS navigation systems ignoring their surroundings to follow (sometimes erroneous/irrational) directions suggests (ibid.). However, theories of the virtual as a representation or simulation are unable to come to terms with the dynamic connectivity and interdependence of physical and digital space which locative media is able to produce. It is bearing this in mind that, following de Souza e Silva and Sutko, we consider the virtual from an Aristotelian/Deleuzian perspective of potentiality.
Having already briefly rehearsed in relation to time the Deleuzian schema of the virtual, actual and of becoming, let us elaborate its intellectual lineage. The idea of potentiality on which Deleuze’s actual and virtual are founded was developed by Aristotle as a way to think about a change in a being’s state (Aubenque, 1962). For Aristotle the virtual is a potential state, which through the act of becoming is actualised, and so the actual is realised potential. This framework was developed further by Leibniz and applied to all reality, describing ‘a world composed of many incompossible worlds’, all possible but unable to coexist (Liebniz, 1934, in de Souza e Silva and Sutko, 2011, p.32). Drawing on an idea from a short story by Borges, Deleuze suggests that there ‘exist’ multiple parallel universes which exist in a virtual state and ‘become actual in the present along a series of infinitely bifurcating paths’ (Martin-Jones, 2006, p.23), a rhizomatic structure of possibility/potentiality (See Figure 1). The virtual is ‘always ready to emerge, to have actual existence’ (de Souza e Silva and Sutko, 2011, p.33). With every fork in the path there are two ‘incompossible presents’ (Deleuze, 1989, p.131), however we are only able to perceive the actual so giving the impression of linearity and singularity.
Whilst this Deleuzian model is of the virtual is a little difficult to envision in terms of its application to time, history and reality in general terms, it has enormous utility
when considering location based applications and the ‘imbrication of information and materiality’ which they imply (and rely) (de Souza e Silva and Sutko, 2011, p.33). There is in the contemporary world, an increasing interpenetration of material objects by patterns of information. The virtual as potential, as becoming, is therefore much more applicable, as location based applications actualise connections between the physical environment and users (ibid.)
Appendix II – Interesting features of Deliveroo Driver Contract
2.1 You are a self–employed supplier and therefore acknowledge that you are neither an employee of Deliveroo, nor a worker within the meaning of any employment rights legislation.
2.2 You further warrant that neither you nor anyone acting on your behalf will present any claim in the Employment Tribunal or any civil court in which it is contended that you are either an employee or a worker.
3.1 Deliveroo is not obliged to make available any minimum level or amount of work to you, nor are you obliged to perform any minimum level or amount of work.
3.2 When applying to join Deliveroo’s supplier pool and at regular intervals thereafter you will provide an indication of the time periods during the week in which you typically expect to be available to work. Deliveroo places reliance on such indications provided by suppliers in planning to meet customer demand. We accordingly expect you to inform a member of the Operations Team if this changes materially, and reserve the right to terminate this Agreement if you are no longer able to work at time periods which meet Deliveroo’s needs.
5.2 Deliveroo reserves the right, upon notice to you, to discontinue the arrangements set out in 5.1 above and move to a delivery fee model. In such event, you will no longer have any entitlement to an hourly fee but will instead be entitled to a delivery fee in an amount to be notified to you for each delivery you complete (a delivery, for these purposes, being the collection of Order Items from a Partner and delivery to a customer of Deliveroo).
SERVICE DELIVERY STANDARDS
You will be expected to meet the following minimum Service Delivery Standards.
During a time period in which you have registered to perform Services, you will log into the rider app and will promptly accept any orders in your zone which you are available to perform. You will promptly answer calls from members of the Operations team or, if you are unable to answer them for any reason (for example, because it would not be safe to do so), you will return them as soon as reasonably practicable.
Upon collecting an order from a Partner, you will click on the rider app to confirm that you have done so. You will then click again when you have successfully delivered the meal to the customer.
You will be expected to meet certain minimum delivery times from Partner to customer. These shall be as notified to you by a member of the Operations team. Persistent failure to meet these requirements shall be considered a serious breach of the Agreement.