Geographic information science II

I Introduction

In my first progress report (Roche, 2014) I examined the concept of the ‘smart city’ from the standpoint of spatial enablement. I especially analysed emerging research on smart cities, particularly those addressing the potential role of geographical information sciences in the development and implementation of the concept of smart cities. The main argument was that a smart city is first of all a spatially enabled city: a city where location, place and any other spatial information are available to local governments, citizens and businesses as a means of organizing their activities (Williamson et al., 2010).

‘Place’ is the central concept explored in this second report. Some recent research (Batty et al., 2012; Chourabi and Nam, 2012; Lévy, 2013a) supports the idea that today’s hypermodern ‘digital cities’ rely more on an approach to the city based on a network of places rather than on an approach to the city built on areal spaces. Indeed active mobility, geocommunication practices, and digital networks have deeply transformed the way people organize themselves and interact with each other in urban spaces, and have provided them with new means for expressing their platial sensitivity. Does this mean that there are more places and fewer spaces in spatially enabled cities? Is the intelligence of a city mainly related to its ability to sound out the genesis of urban places? Is ‘platial intelligence’ the real intelligence needed for a city to improve its smartness? These issues also challenge the design of spatial models used to build GIS, making it relevant to explore these questions from the standpoint of GISciences. These questions reference debates about the importance of placed-based GIS (Goodchild, 2011; Goodchild and Li, 2012; Roche and Feick, 2012) and the conceptual shift from a classical ‘layer-cake view of the world’ to the ‘networked cupcakes view of the world’.

The question of place goes back to Greek philosophical thought and especially the complex relationship between the Chora (Platonic) and the Topos (Aristotelian). Chora is essentially relation-sensitive. Place from this perspective depends on people, events, things, and objects, and simultaneously all of these also depend on the place itself. Places (Chora) are created from these complex relationships and change through time. In some ways, this concept of place is fully in line with Tobler’s First Law of Geography: ‘Everything is related to everything else, but near things are more related than distant things’. However, the Topos is the place of Cartesian coordinates (x, y, z), the one of the cartographer. Topos is geolocated in a whole absolute Euclidian space, and a totally objective, abstracted and geolocated point (Berque, 2013).

Based on this dichotomy, and as a general meaning, place (Chora) is more related to the sensed or sensible components of the geographic space. It refers to relationships between people and space, what Lévis and Lussault (2013a) name ‘spatiality’, while space (Topos) is more defined as the Euclidean support of human activities. According to the Oxford Dictionary, ‘place’ refers to ‘a particular position, point, or area in space; a location’. Place is one of the most important founding concepts of modern geography, and the most discussed in the academic world, with place and space feeding deep and strong debates for decades.

The aim of this report is not to dig into those debates. Rather, I introduce the polysemic nature of place in order to better understand why it is so discussed in the geographical information sciences, and to what extent this could be relevant to demonstrate the usefulness of ‘geo-platial intelligence’ in smart city approaches. The first section below focuses on the concept of place from different complementary points of view. The second explores more precisely the platial perspective in geographical information sciences. Finally, I underline the relevance of this concept of place for smart cities, specifically how ‘geo-platial’ intelligence could contribute to improving urban intelligence and then to making cities smarter.

II Place at the forefront of geographic science

Place is usually and essentially a concept associated with the area of geographic sciences (Quesnot and Roche, 2015). From this geographical perspective, the word ‘place’ is commonly used to designate ‘a particular position, point, or area in space; a location’. Even if it seems a very simple word, apparently easily understandable, many researchers have tried to explain it and to give it a deep and useful geographical meaning (Winter and Frekca, 2012). Space generally refers to the surface and near-surface of the Earth, organized by coordinate systems (e.g. latitude/longitude; x, y), as well as to concepts like distance and direction (Goodchild and Li, 2012). Concepts related to a specific space are measurable or computable within that space, the Euclidian space of the GIScientists or of the land-surveyor. Place is defined as a social construction: a place is a ‘named domain that can occur in human discourse (by contrast, references to latitude and longitude in human discourse are of course extremely rare)’ (Goodchild and Li, 2012: 84). From this perspective, place is typically related to specific events ‘taking place’ in a given location at a specific time (Chan Chun et al., 2014). Nevertheless, place can also persist through time, depending on the temporal nature of the event.

For GIS purposes, places are typically poorly defined in ‘space’ and time as well as in semantic description (Agarwal, 2005), characterized by fuzzy or even indeterminate boundaries (Goodchild and Li, 2012). Indeed, Levy (2013 b) defines place as a space where the notion of distance is irrelevant. Quite similarly, for Lussault (2013), place is the smallest complex spatial unit of society that exists. It has social significance, in terms of practices and representations, if it is registered as an identifiable object, has identificatory or collective functioning, and carries joint values from which individuals can potentially (though not always) recognize themselves. Thus, a place has to be identifiable to exist. It has a spatial identity that allows differentiating itself from other places; place needs to be named in order to exist (Goodchild and Li, 2012, Vasardani et al., 2013).

Places have properties, but there may be substantial differences in individual perceptions of those properties, and their importance in defining places. Yet Tuan (1977) assimilates the place to an object loaded with meanings, an empowered stabilized space within which each individual is anchored. But the spatial identity of an individual cannot be reduced to its place of anchorage. Our spatial identity is constantly evolving through our daily practice of places. This notion of spatial identity is closely linked to spatial value (Lussault, 2013). Within the field of geography, the value of a place is defined by the relationship maintained between an individual and the place in question (Stock, 2004). These relationships are both physical (i.e. concrete practice) and philosophical (i.e. everything that relates to the imagination and representations). Lussault (2013: 1065) defines the spatial value as the ‘totality of socially valuable qualities of a space’.

Moles and Rohmer (1978) also focused on attachment as an intrinsic component of places. As Schwach (1993) explains, the degree of appropriation and the emotional dimension are both factors that were taken into consideration in the definition of Moles and Rohmer’s proxemics. Bailly’s work on representations (1990, 1992, 1993, 1985) also contributed to the enrichment of the definitions of place, regarding place as an identifiable object on the surface of the Earth, but also as a living space with its own identity which is ‘loaded with several meanings (symbolic of certain orientations), its limits (cultural, historical, symbolic…), its images and finally by the symbolic and functional properties assigned’ (Bailly, 1990: 377). Clerc (2004) proposes another way of understanding place. He argues that the definition of place depends on whether one adopts the perspective of spatial analysis – place is an objective point defined by its coordinates, and in interaction with other places – or the perspectives of human geography that takes into account the subject and its experience of places (Quesnot and Roche, 2015).

Goodchild (2011) follows a very close dichotomy with regard to GISciences, framing two main approaches to defining the concept of place. Firstly, place as a concept refers to multiple spatial concepts (location, distance, spatial dependence, spatial heterogeneity …) that have been extensively studied and classified. Geographers have tried to show how they are acquired and developed through human cognition. From this perspective, spatial thinking is seen as a specific and original form of intelligence based on those concepts (Gardner, 1999, quoted by Goodchild, 2011). Place is a more powerful concept than space with respect to understanding of societal dynamics, even if its complexity and inherent vagueness (locational, boundary, description …) make this formalization very difficult, especially in digital environments (Jones et al., 2008). Secondly, place is a context. In social science place is used as a means to define context, the geographic area within which humans live their lives and, more precisely, the relationships between people and their environment. Place is defined as a socio-cultural phenomenon whose meaning depends on the subject and experience (Richter et al., 2012). Fundamentally, place from this context perspective is more about sense, quality, and relationships between people and spaces.

III The platial perspective in GISciences

Place is definitively related to qualitative description and the naming of a specific location rather than to the geometric characterization of space. Digitally representing coordinates (x, y or longitude, latitude) or geometric objects (polygon, line …) in a Euclidian space is easier than representing the sense of a geographical context. From the very beginning of geographical information technologies (early 1960s) place has never really been taken into account. GIS in particular has always provided Euclidian representations of space – the space of the surveyor. Doueihi (2011) argues that the 21st-century digital world requires all knowledge (sounds, images, texts …) to be formalized ‘in a code of just two symbols, 0 and 1’ (in Goodchild, 2011), while Goodchild and Li (2012) argue that ‘The geographic information technologies that have emerged in the past few decades are almost exclusively spatial’ (p. 84).

Thus, geographic information technologies have weaknesses with respect to describing and analysing the sociospatial dynamics that characterize contemporary human societies, as well as supporting spatial decision-making and territorial (especially urban) governance. The recent emergence of geodesign as a new field is a quite good example of these weaknesses. While GIS have partially emerged from landscape architecture and urban design, those fields have rapidly chosen to abandon GIS in favour of tools that are better able to provide and support open, smooth, sometimes ambiguous spatial representation of sensitive dimensions of places (Li et al., 2012; Roche, 2008; Wilson, 2014).

Human discourse and thinking tends to focus on places and their sensitive characteristics, with less attention to the coordinates, distances, and directions of the geospatial approach (Goodchild, 011). The importance of place seems to have grown with the development of VGI / geosocial / geoweb 2.0 fields (Elwood et al., 2013, Khan et al., 2013). Citizens are more and more engaged in the production of place-based information. Personal places are being self-staged by using people’s words and perception. In this context, citizens relate much more to the notion of place (place names, points of interest, vague characteristics, schematic and non-planimetric, events …) than to the notion of space (coordinate) (Goodchild and Li, 2012). This is one reason why a platial perspective is increasingly prominent in GISciences. Much recent research has sought to build conceptual foundation for GISciences – ontologies for this still quite new discipline – with many foundational concepts proposed and discussed (Kuhn, 2012). Place is among them and its relevance is evident in two new application domains that have strong ties to GIScience: geodesign (Nedkov et al., 2014) and smart cities (Walters, 2011).

As articulated within GIScience, the platial perspective is based on a conception of space focused on human discourses, a sensible dimension of human-space relationships, social value, place name and location with less planimetric control. A place is typically defined by a named event (or action, or dynamics, or sense) that takes place (or is associated) to a specific location at a specific time (punctual, recurrent, period). Roche (2012) formalizes this conception of place with this function: P = f (N, E, L), where P is the place, N the Name, E the Event, and L the Location. Event refers to a large spectrum of meanings, the space within which humans carry out habitual aspects of their lives, such as shopping, work, recreation, and sleeping. Such spaces are largely unique to the individual, and likely also to vary through time as habits change, as spaces are learned, or as people migrate (Goodchild, 2013).

Discussions of core spatial information concepts are currently in progress in an open platform hosted by Werner Kuhn’s team at UCSB. ¹ Those conceptual proposals are essential to build the theoretical foundation that is needed for designing geoplatial-oriented methods, technics and tools. Indeed, there is a growing need of place-based technologies, and the idea of place-based GIS is currently gaining ground (Elwood et al., 2013, Roche and Feick, 2012). The digital age requires the notion of place to be well defined and formalized in order to embed it in computing systems. Kuhn and his team do not propose a specific conceptualization of place as a core concept but, rather, propose location, objects and events (by association) as concepts especially in line with the platial perspective.

This formalization is essential to making place-based information sharable and usable in a rigorous, scientific way of reasoning. It requires not only definition of the concept and its main component, but also common standard definition and ‘agreed coding scheme to translate knowledge of the real world’s spaces and places into a binary alphabet’ (Goodchild and Li, 2012, 87). But it also requires relevant place-based digital data that can be used in developing these concepts and methods. The advent of VGI and the big geosocial data deluge have meant the emergence of new sources of place-based data (Miller and Goodchild, 2014, Hemsley and Eckert, 2014a, 2014b, Quesnot and Roche, 2014). While Elwood et al. (2013) argue that the palatial nature of these data present an important opportunity for developing platial GIS, realizing this platial perspective in GISciences faces two main issues. First, we must develop new tools, new models and standards to support citizen engagement in the production and diffusion of rich and useful place-based information. Secondly, new platial analysis models are required to extract a sense of place from big geosocial and VGI data (Li et al., 2013). Understanding the sense of urban places in this manner is a key challenge for smart cities.

IV Places in the smart city

As developed in my first report, there is little consensus around the ideas of smart cities, how they are constituted, and how they should be studied (Roche, 2014). Terms used to define this fuzzy concept refer to the idea of improving urban functions and services provided to the population, not only through innovation but also through the combination of networks (especially wireless), sensors/actuators and the active commitment of citizens. I argued that the intelligence of a city should be measured by its ability to produce favourable conditions for citizens, organizations, private companies, and other urban operators to be actively involved in sociospatial innovation. These operators should be able to develop and mobilize digital spatial skills to manage their spatiality. In other words, I argued that a smart city is first of all a spatially enabled city.

Spatial enablement means that: ‘location, place and any other spatial information are available to governments, citizens and businesses as a means of organizing their activities and information’ (Williamson et al., 2010: 2). In a city context, this requires collection, update, analysis, representation, and communication of time and location-based information. But spatial enablement is also about ‘efficiently managing all information and processes spatially’ (Steudler and Rajabifard, 2012: 5). For urban actors, spatial enablement is very close to Jacques Lévy’s (2013a) notion of ‘spatial capital’, built on Bourdieu’s (1980) social capital. Spatial capital is the resources accumulated by stakeholders that allow them to take advantage of spatial dimensions of society (Lévy, 2013a). To develop smarter cities, it is critical to develop and mobilize citizens’ spatial capital and, thus, their social engagement.

Yet contemporary urban spaces are comprised more of networks of places than of continuous and homogenous areal spaces (Castells, 2010; Latour, 2009). In a digital network society (Doueihi, 2011), places are connected through information flux (Mitchell, 1996). Their structure and dynamics change rapidly through time via social transformations and revolutions, as shown by the Occupy movement or the Arab Spring (Lussault, 2012) or by new forms of active and equipped mobilities (Brun, 2013). Modern networked urban places are hybrid – made of both digital information and physical matter (Zook and Graham, 2007; Roche and Feick, 2012). Thus, urban places become more and more complex. Their informational density grows. Understanding and representing them challenges GISciences on many levels: conceptual foundation, spatial analysis tools, spatial thinking models, and so on. In a global location age when ‘Where am I?’ is being replaced by, ‘Where am I in relation to everything else?’, every personal/individual place is self-staged, and carries a locally rooted version of the global whole it forms (Roche, 2013). ² Flichy (2004) describes this as a connected individualism or network of connected egocentric places.

This growing digital socialization produces new sorts of urban places. In particular in this era of social media, the arrival of the first generation of location-based social networks (LBSNs) has revolutionized the world of locative media. In addition to sharing their opinions about the places they visit, users now can be aware of resources and contacts directly available nearby. Moreover, LBSNs introduced the concept of social location sharing (SLS) or ‘check-ins’, and the ability to tell everybody ‘I am/was there!’ (Quesnot and Roche, 2014, 2015).

Thus, the concept of place is particularly central to developing smarter cities, even as the growing complexity of urban places makes it difficult to sense and understand them. A place is the co-occurrence in between a geographical location, an event or what could be named a platial practice or activity, and a specific time, or as Berque (2013) defines a place (‘lieu’ in French), ‘Là où quelque chose se trouve ou/et se passe’ (Where things are located and/or happening). Before the emergence of Web 2.0 and near ubiquitous mobile internet and smartphones, access to the daily platial practices of citizens (visits, appreciations, popularity, etc.) was technically impossible. Sense of place, one of its fundamental characteristics, is now extractable through the analysis of the digital (spatial) activity generated by social media users. Indeed, geosocial and SLS footprints are data on urban platial practices that were previously inaccessible and unmeasurable (Li et al., 2013, Quesnot and Roche, 2015).

‘Urban intelligence’ is the ability to read these connected complex urban places. Intelligence, from the Latin intellegentia – inter (‘in between’) and legere (‘to choose, to collect’) or ligare (‘to link’) – refers to the cognitive skills needed to understand facts and things and to discover their relationships: the ability to link things together. So the concept of place is important to the smart city precisely because improving urban intelligence requires developing urban stakeholders’ ability for decrypting the genesis of places and their platial practices (Roche, 2012). Smart cities are then not only about data or representation, but also about sending and decoding senses of places and understanding complex relationships (correlation, cause-effects) between physically and digitally connected places.

A smart city is a city where urban stakeholders have developed this ability to identify what is happening in the city and to react properly in a relevant time frame, as part of their spatial capital which today relies both on spatial enablement (spatial skills and spatial thinking capability) and digital literacy. To achieve this, geo-platial intelligence (not in a military or security sense but from the Latin etymology mentioned above) is a key way of understanding urban places and making cities smarter. But it also requires new ways of shaping urban environments. Urban planning and design needs to be more place focused. Saskia Sassen argues that urbanism must be crowdsourced and hacked; it must be rethought from a platial perspective by putting people and place in the heart of urban planning. ³ GISciences can support this goal.

Two recently re-mobilized theories and practices are fully in line with this idea of place-based urbanism. Both are place-making oriented and require spatial thinking about urban issues (need for scalability, multi-perspective, relative location, Tobler’s First Law of Geography, spatial variance, spatial (auto)-correlation, and so on). The first is urban acupuncture, an environmental urban theory that combines urbanism and traditional Chinese medicine theory, considering the city as a living organism and trying to revitalize the whole city by acting on specific target points (Lerner, 2014). Urban acupuncture has some similarities with the concept of tactical urbanism, which focuses on local resources rather than on intensive capitalist urban programs, and promotes participatory urban interventions. The purpose of such small actions is to improve the wellness and wellbeing of communities and to catalyse neighbourhood revitalization. Urban acupuncture focuses on small-scale sensible dimensions (platial) of the city and on actions that positively use and direct energy from and towards urban places (Iveson, 2013). Urban acupuncture and tactical urbanism are place-based approaches to urban transformation that require GIScience concepts, methods, and tools oriented toward platial intelligence.

V Conclusion

A smart city is a place-based city. Smart is about the ability to sense and understand the genesis of connected urban places. To achieve this a smart city needs to be a learning city, a city able to learn about its own behaviour through sensor networks and integrated information systems, and a city that provides urban environments in which citizens can acquire digital and spatial knowledge and skills and develop strategies for managing their platial practices.

The concept of place must be questioned in the context of the modern smart city. There is an urgent need to rethink cities as dynamic networks of connected urban places rather than uniform and continuous sets of areal spaces. Urban design needs new place-based tools and methods. A formalized digital geo-platial intelligence emerging from GISciences could support new approaches like urban acupuncture or tactical urbanism, and engage citizens in crowdsourced and hacked urbanism.

My third report will explore the question of interfaces as a key for enabling platial skills and platial reasoning abilities, and thus allow the development of new forms of digital spatialities for smart cities. While the digital technology is becoming consubstantial to urban materiality, map interfaces, in particular, are the privileged tools for (geographic) indexing knowledge and expertise, as well as a primary means for accessing cities’ informational components and the digital dimensions of urban places.