Playing in Space

This paper proposes a new system for the critique and analysis of video games and other playful artefacts based on Henri Lefebvre’s theory of the production of space. It aims to show that Lefebvre’s work provides an original framework for the analysis of games by dividing playful spaces into conceived space, perceived space, and a lived/ludic space. It also provides new ways of describing a player's actions in playful space through the use of Lefebvre’s concepts of gestures, marks and traces.

Henri Lefebvre’s The Production of Space (Lefebvre, 1991) was a radical attempt to re-centre space in Western thought; to rescue it from the denigrations of the Cartesian tradition, carried on into the mid-twentieth century with the foundation of a myriad of structural linguistic systems. Lefebvre’s work is an attempt to reconcile the “mental space (the space of the philosophers) and real space (the physical and social spheres in which we all live)” (Cover copy, Lefebvre, 1991). The thrust of the work is to show that space is not merely a pre-existing empty void waiting to be filled, but rather something that is actively produced much like any other commodity. Take for example the Spanish pueblos, towns built in the new world to service Spanish imperialism, that have the idea of the hierarchal society built into them through their geometrical layout emanating from a central square. Each society or mode of production produces its own particular space, space that has distinct and unique effects on its inhabitants and visitors. As I will show, if we regard games and other systems of play as spaces akin to those Lefebvre was describing, we can apply aspects of his theory of the production of space to them and open up new avenues for the criticism and analysis of games.

The Spatial Triad

To understand the processes behind the production of space Lefebvre provides us with a heuristic, what he refers to as a “spatial triad”. This is not a rigorous empirical system, a set of instructions that can be followed to accurately read and describe places, but a “dialectical simplification,” (Merrifield, 2006) a squishy and malleable system of moments that overlap and affect one another simultaneously. It consists of representations of space, spaces of representation, and spatial practices. Representations of space means explicitly designed space. Space directed by professionals or technocrats. Space designed by urban theorists, governmental planners or property developers and their architects among many others. This is reified space, where the abstract plans of these people are made manifest, where they are turned into actual “things”. Representations of space are made up not of material, but rather the language and signs employed by these professionals. Lefebvre describes it thusly as “tied to the relations of production and to the ‘order’ which those relations impose, and hence to knowledge, to signs, to codes, and to ‘frontal’ relations.” This is the space of power and capital, and thus plays the largest role in the production of space. “Representations of space must therefore have a substantial role in the production of space. Their intervention occurs by way of construction - in other words, by way of architecture, conceived of not as the building of a particular structure, palace or monument, but rather as a project embedded in a spatial context and a texture which call for ‘representations’ that will not vanish into the symbolic or imaginary realms.” To Lefebvre, this space is a “conception” of the way things should be. Generally, this could be defined as ideology, but Lefebvre argues that in current society, ideology has been replaced by knowledge and (following Marx) is therefore a productive force. As a result, we see that representations of space in our current society are key tools in the reproduction of capitalism and maintenance of its cultural hegemony.

Representational spaces, the second aspect of Lefebvre’s triad, are the spaces of the everyday. This is the world of slang and argot rather than the language of the professional. Lefebvre describes the representational space as “space as directly lived through its associated images and symbols, and hence the space of ‘inhabitants’ and ‘users’, but also of some artists and perhaps of those, such as a few writers and philosophers, who describe and aspire to do no more than describe.” These spaces may be shared but are on the whole uniquely personal, they are the journey from your home to your work, the shop you think is better for buying fresh vegetables, the pub, the first left after you pass under the bridge. These spaces are perceived or experienced, rather than thought - “It embraces the loci of passion, of action and of lived situations…” Contrast this then to representations of space; representational spaces are exposed rather than imposed, therefore they leave little lasting mark on the material world. They mainly inhabit the symbolic realm, rather than having the physical texture that is left by spaces of representation. Lefebvre tells us that representational space “is the dominated - and hence passively experienced - space which the imagination seeks to change and appropriate.” Representational space is hard to pin down, not least because our very thoughts seeks to understand it, and representations of space seek to change it.

Spatial practices are the third aspect of Lefebvre’s triad, and are the practices which “secrete” society's space; spatial practice “propounds and presupposes it, in a dialectical interaction; it produces it slowly and surely as it masters and appropriates it.” What does this mean, however? This question is easier to answer when we look at what Lefebvre had in mind as spatial practices consisting of, “a close association, within perceived space, between daily reality and urban reality (the routes and networks which link up the places set aside for work, ‘private’ life and leisure).” Spatial practices structure reality, they are the networks, routes and patterns that people use for orientation and navigation. They connect “people and places, images with reality and work with leisure” (Merrifield, 2006). We can see how tracks, roads, landmarks and monuments, as well as more natural features such as mountains or rivers help to define a person's sense of location and thus the way they act. Lefebvre is vague however about how exactly these practices relate to representations of space and representational space; these practices simultaneously both combine and separate the two other aspects of space. What he does say though is that lived experiences, representational spaces, always end up dominated by the conception of a space, representations of space. Lefebvre tells us that what is produced then is yet another type of space, abstract space. Abstract in this context does not imply unreality; abstract space is a very real thing indeed, where we see its expression in the design of buildings, places, objects and the actions of states, NGOs and other international organisations. Abstract space imprisons all of us and prevents us from exercising what Lefebvre called our right to difference. This right governs not just our behaviour in a society but also expresses a geography of different rights, so instead of just being an extra source of rights, “it is the source of them”.

Lefebvre intentionally sets up his triad to be confusing. This is to prevent readings of his work that give rise to overly analytical, formal systems. Lefebvre did not intend this triad as a model, indeed the triad “loses all force if treated as an abstract ‘model’. If it cannot grasp the concrete… then its import is severely limited.” Instead “the lived, perceived and conceived realism should be interconnected, so that the ‘subject’, the individual member of a given social group, may move from one to the other without confusion.” For our purposes however, we can simplify it because we do not intend to use it to describe the real world but rather as a tool of the analysis of play and player interaction with video games. As hinted at in the last quote the triad can be reduced to the relations between spaces Perceived (Representational Spaces), Conceived (Representations of Space) and Lived (those emerging from spatial practices). This structure, as I will show, bears a resemblance to the structure of objects which are designed for play. A designer provides a system which contains a conceived notion of how it should be interacted with. A player then brings their own personal, perceived notions on how the same system should be interacted with. These spaces have a mutual effect on one another but both are also governed by a third lived/ludic space that contains the practices that influence a player's and designer's response to all games and the act of play itself.

Why is this triad helpful for the critique of games? And what advantages does it offer us over existing systems? There is a formalist thinking in games design that places too much emphasis on the procedural domain of a game. It holds that meaning can be communicated purely through rules (Bogost, 2006). This method of thinking has given rise to games where players explore the game's procedural domain and uncover meaning through discovering its constraints. Designing in this manner leads to games becoming closed works that lecture the player instead of allowing them room to think critically. According to this type of thinking, games can have meaning that exists before a player begins to interact with them, which diminishes the importance of play and gives players unsatisfactory experiences.

Miguel Sicart seeks to counter this by proposing a new model that introduces a semiotic domain alongside the procedural one. He writes “Games are ethically relevant depending on how the relations between the semiotic layer and the rules are designed. Ethical gameplay design is craft of creating experiences that invoke a player's ethical capacities by manipulating the ways in which the game system is communicated via a semiotic domain that targets the cultural being of a player.” (Sicart, 2013) Sicart is then engaging in a major restructuring of the frameworks we have for analysing games by placing players and player response to the aesthetics of a game back into these frameworks. Using this framework to abstract games into a semiotic domain allows for interpretations of “the meaning of that system as communicated to a player who is interested in the interpretation of the game experience in the context of play.” (Sicart, 2010).

But semiotics alone are not enough to fix the problems of the early models. Rules and codes developed for the analysis of literary texts applied to a playful space only allow us to remain on a descriptive level, where we do not begin to understand the act of play or why people respond the way they do: “Any attempt to use such codes as a means of deciphering social space must surely reduce that space itself to the status of a message, and the inhabiting of it to the status of a reading.” (Lefebvre, 1991). I seek not to develop a code to just read games, but one that also allows them to be constructed.

Spaces in Games

Let us now take a closer look at how the aspects of Lefebvre’s triad relate to video games and how the whole system could be used for a deeper analysis of them. Representations of space (conceived space) in the context of a game is the space designed by the developer. Into it go all their biases and notions about how a player should react and play with the systems and aesthetics they have conjured. It is key now to remember that a games space does not only refer to virtual worlds, but also the space of play that all games define. Candy Crush (King, 2012), a “match-three” puzzle game, defines a space consisting of 2D icons of sweets, while Skyrim (Bethesda Game Studios, 2011) defines a large fantasy world for the player to occupy and explore. Both of these are spaces in the context assumed by this paper, space here means nothing more than the area that is available to a player for play in any game. This is not the same as saying the breadth of actions available to a player or the set of variables that make up the game state. A game with a dominant strategy, that is, a strategy that is considered better than any other, has a small space for play no matter how complex the games rules or no matter how large the set of actions available to the player.

Representational space (perceived space) is the space that the player immediately occupies when they play a game. It is the space they construct as they continue to interact with a system. In this space, players conjure their own preferred strategies, routes and systems. Crucially, they do not bear a one-to-one relationship with the conceived space laid out by the developers, rather each player has their own way of playing that is never individually catered for by the designers. As mentioned above, these spaces are exposed rather than imposed. This space is hard to rationalise, and when asked why they did things in a particular way, players will respond with an answer such as they did it because it felt “natural” or that it was simply the first thing they considered for example.

Lefebvre wrote of a similar kind of space when he described theatrical space, “To question whether such a space is a representation of space or a representational space, the answer must be neither - and both. Theatrical space certainly implies a representation of space - scenic space - corresponding to a particular conception of space (that of the classical drama, say - or the Elizabethan, or the Italian). The representational space, mediated yet directly experienced, which infuses the work and the moment is established as such through the dramatic action itself.” This passage could easily be translated to consider games. Game space also implies a representation of space which follows a particular conception of space, thinking of the differences in style between German-style and American-style board games, or between Japanese and Western video games, for instance. Each of these different traditions implies a different conception of space, in fact it is the idea of space that they convey that could be said to define them as a genre. Just as Lefebvre writes that representational space is defined in the moment through dramatic action in theatre, we can say that representational space is established in games through the act of play. Playing works to create a particular space that players individually inhabit while interacting with a playful artefact.

No Wrong Way To Play

Breath of the Wild (Nintendo, 2017) (BotW) is an open world RPG developed by Nintendo Entertainment for the Nintendo Switch. In it players take on the role of Link, a knight of Hyrule, tasked with exploring the large open world kingdom. BotW is defined by its nonlinear gameplay where players are purportedly left to their own devices and allowed to decide what to do themselves. In reality, the space of BotW is produced by strong conceptual notions introduced by the designers which I will seek to describe here.

At Japan’s Computer Entertainment Developers Conference in 2017, Game Director Fujibayashi Hideyuro and Senior Lead artist Makoto Yonezu (Hideyuro and Yonezu, 2017) discussed some of the design choices they had employed to guide the player through the game. Such a large open world as the one in BotW means a way had to be designed to encourage the player to move through it. The solution was the employment of lots of triangles in the environment. This served two purposes, it presents a choice to the player to either climb it or to go around it, as well as breaking up and obscuring the environment to encourage exploration and progression in the player. This slide shows the extent to which this rule was employed.

This method was also fractal. Larger triangles drew the players towards more critical areas and acted as gates between major areas of the map. Medium-sized triangles hide smaller, non-critical areas such as enemy camps or buildings. The smallest triangles, sometimes just made up of a rock on the ground hide the smallest rewards such as individual items or the games collectable known as "kurok seeds." The following slide shows examples of the three scales of triangles within the game.

The conceptual space employed by the designers does not only guide the movement of the player. Weapons found in the game degrade extremely quickly compared to other games with similar mechanics. Weapons, shields and bows will permanently break after even a small amount of use. No item in the game is immune to this effect, even the sword and shield elevated to divine status by the game's narrative will be unusable after a certain period of use. Why is this mechanic present in the game? It is there to build a certain representation of space, it is there to encourage a certain type of play. The designers believed that the players not being able to rely on their weapons would encourage them to play with the other systems in the game such as the physics system or the “Sheikah slate” powers (a collection of magical abilities the player possesses) that allow other actions to the player such as magnetism or time-freezing.

Player reaction to the low weapon durability was mixed. Many players were unhappy with what they saw as an unwelcome intrusion into their enjoyment of the game by regularly taking away the tools they were using to have fun. Other players recognised the intentions of the designers and argued that this decision encouraged more creative types of play. Who is to say who is right in this case? The players that complained about this mechanic were surely being sincere about their belief that they would get more enjoyment out of the system if they were allowed to keep weapons for longer. Meanwhile its supporters were surely right about the mechanic's intent and its potential to guide players to more interesting types of play. However, if this guiding was indeed successful in its aim then there would not have been the same degree of complaints about the mechanic in the first place. Here then, for some players at least, there is combat between the representation of space as conceived by the designer and the representational space as played by the player. Some players have been lost to abstract ludic space.

No Wrong Way to Play (Burch, 2013 - 2017) is an online collection of the different methods, other than those the designers intended, that players have found of playing their favourite games. For example, a method is provided for turning Rocket League (Psyonix, 2015), described as "soccer, but with rocket-powered cars" into a version of shuffleboard where “Essentially, players flip on their back with full turbo equipped, and another player comes by and hits them forward. You want to land in the center of the field, where the game would normally spawn the ball that everyone’s after. The other player is allowed to use turbo (once!) to position themselves on the board, but when the turbo runs out, that play is over.” (Klepek, 2015) These methods are a testament to the radical ways that players find their own fun in games. While real representational space never reaches this kind of formality. it does create the conditions for these types of experiments to be born.

Player-made modifications to games also perform this radical reorganisation of who has control over a game's conceptual space. They generally fall into two categories of modifications that “improve” the game by providing a slight tweak to pre-existing mechanics, and larger modifications that introduce whole new features to a game or “total conversions” that construct whole new games out of old ones. On the popular “Nexus Mods” site that acts as a download hub for these mods, players can make small changes to their copy of Fallout 4 (Bethesda Game Studios, 2015) by downloading mods that make the mini games present in the game easier through changing the amount their character can carry or making their armour more durable. If they were seeking a larger change they could add a cover system to the game or advanced survival mechanics that oblige them to eat and drink regularly. Modifications sometimes expand on aspects of the base game that the original developers never could expect. “Surf” maps for source engine games, especially Counter Strike: Source (Valve and Turtle Rock Software, 2004) exploit a bug in the engine that allows players to “surf”, that is move forwards without player input, along surfaces with a certain gradient. Players have made maps that facilitate this type of movement and ask players to traverse the environment in a very different way to what they would normally expect.

Ludic Practices

With a better understanding of what spaces exist in video games, we now can turn our attention to defining some of the practices that govern the relationship between representations of space and representational spaces. In The Production of Space, Lefebvre leaves scant writing on how individuals experience space, focusing instead (and perhaps understandably) on how societies have constructed particular spaces. One clear idea he does leave for us however is the idea of gestures, marks and traces. Thankfully much simpler than his triad of space, these three ideas can be used to describe how an individual recognises what agency they have in a space and the effects they can have on the space:

1. Gestures are the actions an individual can perform in a space. In video games we can interpret these as the set of actions that a player possesses for interacting with a game's systems. What gestures a player can perform are likely to be restricted depending on the state of the game's system at any particular moment in time.
2. Marks are the long-lasting or permanent effects and changes an individual can make to a space. In video games we can see these as advancement through the game's main structure.
3. Traces are the impermanent, often very fleeting effects and changes that an individual can make to a space. Traces can often feel inconsequential. These are the results of most of the actions that a player will perform while playing a video game.

The number of gestures, marks and traces can vary wildly between different games. Looking at the numbers of each could give us some idea of the nature of that particular game. A large number of gestures indicate a high degree of complexity to a game, while fewer gestures suggest simplicity. In Crusader Kings II (Paradox Interactive, 2012), a game where players play a medieval dynasty from 1066 to 1337, players have an expansive selection of options available to them, from decisions maintaining their lineage, to whether and when to wage war and maintaining relationships with the thousands of characters that exist in the game world at any one time. Compare this to games of the Endless Runner genre where it is possible for games to have as little as one input as in Canabalt (Adam Saltsman, 2009).

The number of marks and traces are slightly more challenging to draw clear conclusions from. Unlike the number of gestures, it is not possible to view them in isolation and draw inferences about the game, rather, they must be viewed together. It is the permanent/impermanent contrast between marks and traces that gives them this important link. Viewed at a higher level than individual games, we see similarities in the number and type of gestures, marks and traces in games of similar genre. Fighting games usually have a large set of gestures for the players to explore and master, but their pool of marks and traces are conversely very small, traces are usually contained to one match, for instance the diminishing of a characters health (always refilled the next time you play them) while marks are restricted to progression systems outside the action of individual fights. We can compare this set of gestures, marks and traces that make up a genre with the set that make up another. Racing games have similar marks and traces to fighting games. Traces are confined to a single race, consisting of things such as a player's lap time or their position, while marks again make up the progression system and structure of the game outside of the individual races. Gestures however change drastically between these two genres, with racing games having significantly fewer actions available to the player than the complex, seemingly endless lists of actions available in fighting games. With this simple application of comparing the gestures, marks and traces of these two genres we can easily see similarities and differences between them that may be more difficult to identify with a more traditional reading.

The Power of Play

Why is it important that we analyse games in this way, separating them into and interrogating their different spaces? We must do this to ensure that players are given adequate room to play and do not have their spirits crushed by an oppressive designer. It is important to remember though that a playful nature to a game should be the main goal of games designers, that developing playful games means giving more control towards the players and allowing them to fill works with their own virtues. This will take place when we create games that ask the player how they want to play and are accepting of any answer. To tie this back to the theory of spaces, we must seek to create ludic spaces that are not comprised of oppressive representations of space. We must not force ideas on players about how games are meant to be played, nor offer too small a space for representational spaces to flourish.

Play has long been recognised as having radical potential, in the ideas of spontaneity and moments, propounded by Lefebvre elsewhere or in the idea of situations (Situationist International, 1957); we see the importance of a sudden realisation of agency or connectivity. It is this realisation that the well-played game can engender. The magic circle we enter when we play games, the rules of society that we disregard when we begin to play, allows us to see the possibilities of the everyday. The idea of carnival has long been recognised as a valuable form of resistance and the time has come to add play to the same toolkit. Carnival is a mixing of the political and the aesthetic, a time when the traditional relationships that govern society are upturned, divisions between reality and unreality, between life and art are blurred. Carnival is a time when all people can play. The Situationists, a collective in mid-20th century France, named for the creation of “situations” an idea very similar to Lefebvre’s moments (Lefebvre was closely associated with the Situationist International and a close friend of its figurehead Guy Debord, but was never a member himself) understood the importance of play in carnival, “Free creativity in the construction of all moments and events of life is the only poetry it [the first conscious critique of everyday life] can acknowledge, the poetry made by all, the beginning of the revolutionary festival. Proletarian revolutions will be festivals or nothing, for festivity is the very keynote of the life they announce. Play is the ultimate principle of this festival, and the only rules it can recognise are to live without dead time and to enjoy without restraints.” (Situationist International, 1966). The Situationists maintained that even individual action helped create a revolutionary and liberating atmosphere for all. “All people have the same will to authentic self-realisation, and…their subjectivity is strengthened by the perception of this subjective will in others. This way of getting out of oneself and radiating out, not so much towards others as towards that part of oneself that is to be found in others, is what gives creative spontaneity the strategic importance of a launching pad.” (Vaneigem, 1967)

Conclusion

I have tried to present here a new way of looking at games and other playful artefacts that allows designers to create games that give players adequate room to play. It may be argued that any type of structure to a game will only result in repression of a person's freedom or ability to play. Indeed in his book Homo Ludens, Johannes Huizinga maintains that “Really to play, a man must play like a child.” (Huizinga, 1955). In the politics of play today this is a naive approach. We as humans almost instinctively understand that there is virtue in more structured types of play and willingly submit to systems of rules as we know that paradoxically they will allow us agency that we do not have outside of the game.

Lefebvre’s methods were not written with such an application as I have devised in mind but I believe it is a worthwhile endeavour. We have seen how well his theories of space map onto games and the framework they can provide in games' analysis and design. More work is needed however. I have not addressed the effects that communities and the discourse around games have on how they are played. These at the very least form many more spatial practices. The ways that people interact with games has never been more dictated by forces outside of the game itself, the construction of the meta-game, the effect of memes and social media, the demeanour and method of popular game influencers such as streamers and other video content producers - all this has an effect on what we play and how we play. These may even go as far as dividing the representations of space in games in two, those conceived by the designer and those conceived by the community around a game. Ian Bogost has written recently of how the playing of games is being delegated and play is being avoided altogether. In his discussion of Untitled Goose Game (House House, 2019) he writes how “It serves up the usual goals and challenges, but it does so in a way that allows the multitudes of people who encounter the game to skip playing it entirely. Instead, they delegate the effort to a smaller group, which delivers parcels of enjoyment by condensing them into bite-size memes.” These new ways of designing and experiencing games all require further thought and research, though that is out of the scope of this paper, which only seeks to introduce new frameworks through which future work can be done. Hopefully the model and mode for thinking about games presented here leads to more playful games where designers are more critical and aware of what they ask of the players and the conceptual spaces that they ask players to play in. Players too stand to gain from this model as an increased critical perspective on the games they play to ensures that they can find the most worthwhile way to spend their time.

References

Saltsman, A. (2009). Canabalt. Microsoft Windows. Self-Published

Bethesda Game Studios. (2015). Fallout 4. Microsoft Windows. Maryland: Bethesda Softworks.

Bethesda Game Studios. (2011). The Elder Scrolls V: Skyrim. Microsoft Windows. Maryland: Bethesda Softworks.

Burch, A. (2013-2017). No Wrong Way to Play. https://nowrongwaytoplay.tumblr.com/ [Accessed 25/11/2019]

Bogost, I. (2006). Unit Operations: An Approach to Video Games Criticism. Cambridge, MA: MIT Press.

Bogost, I. (2019). Don’t Play the Goose Game. https://www.theatlantic.com/technology/archive/2019/10/dont-play-the-goose-game/600472/ [Accessed 25/11/2019]

Cover Copy. Lefebvre, H. (1991). The Production of Space. Malden, MA: Blackwell Publishing

King. (2012). Candy Crush Saga. iOS. Online: St Julian's, Malta

House House. (2019). Untitled Goose Game. Nintendo Switch. Portland: OR: Panic Inc.

Huizinga, J. (1955). Homo Ludens: A Study of the Play Element in Culture. Boston:
Beacon Press

Hideyuro, F and Yonezu, M. (2017), Development and Design of The Legend of Zelda: Breath of the Wild. CEDEC. September 2nd, 2017, Yokohama

Klepk, P. (2015) Car Shuffleboard Is The Goofiest New Way To Play Rocket League. https://kotaku.com/car-shuffleboard-is-the-goofiest-new-way-to-play-rocket-1724846200 [Accessed: 27/11/2019]

Lefebvre, H., Translated by D, Nicholson-Smith (1991). The Production of Space. Malden, MA: Blackwell Publishing

Nintendo Entertainment Planning and Development Division. (2017). The Legend of Zelda: Breath of the Wild. Nintendo Switch. Kyoto: Nintendo Co., Ltd.

Merrifield, A. (2006). Henri Lefebvre: A Critical Introduction. New York, NY: Routledge

Paradox Development Studio. (2012). Crusader Kings 2. macOS. Stockholm: Paradox Interactive.

Psyonix. (2015). Rocket League. Microsoft Windows. San Diego, CA:Self-Published

Scoot, R. (2007). Nexus Mods. https://www.nexusmods.com/ [Accessed 25/11/2019]

Sicart, M. (2013). Beyond Choices: The Design of Ethical Gameplay. Cambridge, MA: MIT Press.

Situationist International. (1966). On The Poverty of Student Life. http://library.nothingness.org/articles/SI/en/display/4 [Accessed 25/11/2019]

Situationist International. (1957). Report on the Construction of Situations and on the International Situationist Tendency’s Conditions of Organisation and Action. http://www.bopsecrets.org/SI/report.htm [Accessed 27/11/2019]

Valve Corporation and Turtle Rock Studios. (2004). Counter Strike: Source. Bellevue, WA: Self-Published

Vaneigem, R. (1994) The Revolution of Everyday Life. London: Rebel Press Left Bank
Books

Illustrations

Buck Nasty. (2012) Figure 3. surf_buck_this https://gamebanana.com/maps/167858 [Accessed 27/11/2019]

Hideyuro, F and Yonezu, M. (2017), Figure 1. World Triangle Illustration. CEDEC. September 2nd, 2017, Yokohama

Hideyuro, F and Yonezu, M. (2017), Figure 2. Triangle Fractal Illustration. CEDEC. September 2nd, 2017, Yokohama