Benefits and Constraints of Using Geographic Information System (GIS)
1.1 Research Background
cThis is no more evident than in the proliferation of Geographic Information Systems (GIS) across a variety of disciplines, with the common goal of capturing, storing, analysing and visualizing spatial information. GIS in practice, by virtue of its technical complexity and cost, has traditionally been limited to the operations of Governments and commercial organisations (Craig et al., 2002). Despite these barriers non-profit organisations and community groups are increasingly looking to adopt GIS on the premise that it will be able to positively transform their operations through better decision making and influencing public policy through greater analysis and the presentation of professional visualisations (Sieber, 2000b, Sieber, 2000a). Given this burgeoning interest, there has been a concerted effort by GIS and Society (GISoc) research groups to develop and espouse concepts such as Public Participation GIS (PPGIS) which provides a unique approach to make GIS and spatial data available to non-traditional users allowing them to integrate local knowledge and engage in decision making (Sieber, 2006).
1.2 Research Objectives
The focus of this research project is to investigate the benefits and constraints for the application of a Geographic Information System (GIS) within a community based project. Specifically the research considers a reframing of PPGIS to help better guide the processes, resources and characteristics required to implement a community based GIS. The following questions will guide the research and development of the community-based GIS: 1. Do contemporary PPGIS pragmatic approaches address the original ontological debates of GIS and Society? 2. Can psychogeographic principles help better guide the requirements for a community based GIS? 3. What spatial data sets are available and usable for community groups within Melbourne, Victoria? 4. Do available datasets satisfy the requirements of community groups? 5. Can community knowledge be effectively integrated with traditional spatial data sources?
1.3 Research Rationale
As people become more aware of local, regional and global issues through the mainstream media and the Internet they, as a result, expect to be better informed by Governments and organisations and allowed to contribute to decisions that shape their own lives and the society in which they live. If those issues comprise spatial knowledge, then a GIS is a natural option for facilitating discussions and conveying local knowledge (Carver, 2003). Despite this opportunity to empower communities many GIS practices (including PPGIS) and available spatial data often do not adequately represent community needs and concerns (Elwood, 2006). This research thus aims to explore and develop a framework for which current GIS and related technologies can be successfully reconstructed to allow communities to express their own knowledge about place and spatial relations through visualizations and narratives. Specifically, the proposed research has been designed to assist the Blackburn Lake Sanctuary (BLS) Advisory Committee to implement a GIS which will be enable them to store and map the location of various vegetation and salient features within the BLS in Melbourne, Victoria. By integrating publicly available data sets with community knowledge it is hoped that it will further legitimise the activities of the BLS Advisory Committee while not compromising their goal of contributing to local government policy and increasing the effectiveness of their activities.
1.4 Research Methodology Contained here is an outline of the subsequent chapters and research methodology. The research will be organised into three major chapters - literature review; case study; and discussion and conclusions. Chapter Two - Literature review - examines the relevant literature regarding GIS and Society, PPGIS and psychogeography providing an overview of the historical background and ontological framework of these research paradigms. An examination of the principles of psychogeography and the research design of previous PPGIS studies will be completed, providing a comparative study of their different methodologies and methods. These comparisons will assist in developing a theoretical framework for a community-based GIS which will guide the case study to follow. Chapter Three - Case study - introduces the Blackburn Lake Sanctuary case study and attempts to implement the methods established within the theoretical framework introduced in chapter two. An exploratory case study has been employed because it is a valuable method for investigating the nature and effects of implementing technology within a complex milieu (Sieber, 2000b). In order to increase the rigour and validity of the case study observations, open-ended interviews and questionnaires will be conducted. Chapter Four - Discussion and conclusions - reviews the research objectives in relation to the major research findings as well as the limitations of the methods and theories employed. Pattern matching techniques will be employed to compare the observed and verified information with the framework developed through the research project. If the observed and predicted information correspond then the research methodology maybe strongly validated (Sarantakos, 1998). This chapter also addresses the limitations of the research and future research opportunities.
1.5 Conclusion This chapter has established the objectives and rationale for conducting research into developing a community-based GIS. A research methodology has also been proposed to describe how the research statement and associated objectives will be achieved. The next chapter will review the relevant literature - including theoretical models and research methodologies used by previous researchers in the field of PPGIS and psychogeography.
2. Literature Review
2.1 Introduction In the previous chapter, the objectives, rationale and methodology were presented to help guide the research into developing a community-based GIS. The research outlined in this thesis covers a number of interdisciplinary fields - all of which are continually evolving. These fields include public participation GIS (PPGIS), community mapping and psychogeography. This chapter begins by investigating the role of GIS in society including the motivation and foundation for PPGIS and the advantages and disadvantages of PPGIS praxis. The chapter also explores the topic of psychogeography and the reasons why its principles may help characterise and drive the successful development of a community GIS.
2.2 GIS and Society - a brief history "Mountains dark with forests rose above the rooftops, the jagged black summits silhouetted against the evening light. Higher than them all, though, was the tip of the Schneeberg, glowing, translucent, throwing out fire and sparks, towering into the dying brightness of a sky across which the strangest of greyish-pink cloud formations were moving, while visible between them were the winter planets and crescent moon." (Sebald, 2002: 50) Storytelling is an extremely powerful means for conveying an image of the world and in some way or another every story takes place somewhere and relates knowledge of geography and a sense of place (Cartwright, 2004, Erle et al., 2005, Cartwright et al., 2009). One way to represent geographic stories and our understanding of the spatial organisation of the physical environment and its relationship with humans is through a map. An attempt to bring together the science of geography with the art of map making has been the Geographic Information System (GIS) - which is a computer system for capturing, storing, querying, analysing and displaying geographically referenced data (Chang, 2008). What differentiates a GIS from other databases and computer systems is its ability to combine large amounts of spatial data from diverse sources, group the data into layers or categories, analyse the data for patterns or relationships and produce improved visualizations (Sieber, 2000a, Sieber, 2000b). For these reasons GIS technology has become an important tool for use by many levels of Government, Universities and organisations involved in activities ranging from conservation, advertising and marketing, health, crime, land-use planning and social services - or any activity containing a spatial component (Sieber, 2006). However it is only recently that GIS use has expanded to non-traditional users such as non-profit organisations and community groups. This accessibility has been the result of decreased costs in hardware, software and improved user interfaces which means the user no longer has to learn specialised command languages (Craig et al., 2002). The attraction to the utility of GIS, by non-traditional users, is much the same as traditional users in that it can assist in new ways of understanding a problem, but it may also help in influencing public policy through more sophisticated analysis and the presentation of professional looking images (Sieber, 2000b). Despite this perceived ease-of-use and increasing ubiquity, the GIS has been criticised by some circles as being an elitist technology which merely enhances existing power structures (Carver, 2003). This critique is heavily influenced by postmodernist principles, which place an increasing emphasis on the contributions of wider society and recognises that knowledge and values are constructed through a multiplicity of social and cultural forces. These arguments first surfaced within the paradigm of critical cartography which exposed the inherent subjectivity in, and rhetorical content of maps, thus implying that maps are as much a reflection of (or metaphor for) the culture that produces them, as they are an abstraction of the physical environment (MacEachren, 1995). These examinations have also been employed within social and critical GIS debates which challenge the use of GIS in decision making as being objective and neutral. Instead it has been maintained that GIS utility is often confined to 'experts' whom produce privileged knowledge given their unique access to data, technology, resources and position to structure the inquiry and design the output (Duncan and Lach, 2006). This view of GIS as a return to the principles of technocratic positivism may be construed as anti-democratic because decisions reliant on a GIS may exclude diverse forms of spatial data, such as community knowledge, in favour of ordered Government data conceptualised into points, lines and areas (Crampton and Krygier, 2006). Many academics, such as Pickles (1995), believe that the increased popularity of GIS within the geography discipline has meant that the availability and access to geographic data has become more influential than knowledge or experience of a unique environment or subject (Craig et al., 2002). Concerns regarding the hegemonic and subjective role of GIS lead to a number of workshops in the mid-nineties on 'GIS and Society' (GISoc) sponsored by the National Centre for Geographic Information and Analysis (NCGIA) (Craig et al., 2002). GISoc was focused on how the spread of the technology was affecting "the political, economic, legal and institutional structures of society; and how societal processes affect the form taken by the technology itself" (Carver, 2003: 65). GISoc research furthermore questioned whether current GIS practices and available spatial data adequately represented community needs and concerns and whether a new ontological framework was needed to help empower less privileged groups in society (Elwood, 2006). It was questioned whether it would be possible to develop a 'bottom-up' GIS which could successfully incorporate community participation and thus either displace or validate decisions made with 'top-down' GIS approaches, implemented in most Government and commercial GIS projects (Craig et al., 2002). From these reflections the notion of Public Participation GIS (PPGIS) was developed and defined as "a variety of approaches to make GIS and other spatial decision-making tools available and accessible to all those with a stake in official decisions." (Schroeder, 1996) In other words, the intention of PPGIS praxis was to incorporate local community perspectives into decision making, ideally leading to solutions which might otherwise not have been attained using traditional data sources and esoteric problem framing and analysis (Carver, 2003). Even though PPGIS was initially seen as a reconstructed democratic GIS, there is still much polarized debate regarding whether GIS technology is empowering or marginalising. These opinions, however, can often be seen as a reactive and predisposed view of information technology (IT) in which individuals and groups must react to the technology as having a positive or negative social effect (Sieber, 2000a). The effectiveness and social and political implications of GIS use within communities, however is much more complex and is generally contingent on a set of unique local factors such as culture, policies, standards, people and technology (Duncan and Lach, 2006). While PPGIS applications are an extremely positive move to address the original GISoc concerns regarding the social, political and knowledge practices of GIS - contemporary PPGIS have seemingly introduced new contradictions concerning data access, representation and hegemony (Elwood, 2006). For instance, although much literature acknowledges the importance of bridging the gap between technology and community knowledge, many PPGIS applications continue to adopt a technocratic view of GIS and are often lacking extensive public interaction let alone the integration of community data (Sieber, 2006). The next section of this thesis examines more closely the current contradictions in PPGIS applications, paying particular attention to the ambiguities in the use of the terms public and participation. The purpose in critiquing the underlying assumptions of these terms is to further an understanding of the original ontological meaning of PPGIS and how the varying use of the terms has affected the incorporation of community knowledge in PPGIS projects.
2.3 What does the Public and Participation in PPGIS really mean? Any process or technology which enhances a communities access to information and provides the chance to participate in decision making should be seen as a step in the right direction; however the original ontological framing of GISoc has become misconstrued by some practices of PPGIS (Carver, 2003). The original concept and application of PPGIS has been reshaped and become more disparate over the years as a range of disciplines (such as urban planning and conservation), have applied different approaches and technologies to achieve a unique set of priorities and goals (Sieber, 2006). Furthermore the growing enthusiasm of Governments and commercial organisations for participatory planning has lead to a number of diverse initiatives which vary in terms of the inclusiveness of community knowledge and empowerment potential (Elwood, 2006). However, public participation is a complicated concept that can have multiple meanings which lead to numerous interpretations and societal expectations (Schlossberg and Shuford, 2005). An examination of the nature of public and participation practices in GIS applications is thus critical in developing a greater understanding of the ambiguities in the PPGIS process and how these may have diverged from the original vision of GISoc. It is extremely important to identify whom the 'public' is when engaging a PPGIS project because it will ultimately determine who is included within the project and what types of outcomes and goals may be achievable (Schlossberg and Shuford, 2005). The Collins English Dictionary (1982) defines 'public' (adj) as a means "of relating to, or concerning the people as a whole" reflecting the intended meaning within PPGIS; and many applications do continue to be developed for a general public (Sieber, 2006). There is however a number of PPGIS projects who take a more ambiguous view of 'public' and often use it interchangeably with definitions which more loosely resemble a stakeholder. In other words many projects deem their public to be those who are affected by, bring knowledge or information to, and possess the power to influence a decision or program (Sieber, 2006). The public and their interests are often, however, very different from stakeholders and thus would heavily influence the problem framing and objectives of a GIS project (Wood, 2005). Furthermore a public can be demarcated by a range of factors such as geographical, economic, social or political; and the composition of a public may change over time (Schlossberg and Shuford, 2005). Determining what constitutes the public has become especially complicated as technology has become more pervasive. For instance a web-based GIS may potentially be accessible by a wider portion of society however it raises questions around digital divides and geographic scale. Thus, is anyone who is able to access the application still deemed part of the public even though they maybe geographically distant to the issue and decision making? (Sieber, 2006) In general people local to an issue should be interested enough to get involved in a debate given their geographic proximity. It has been demonstrated however that as scale increases not only do people at regional, national and global levels become interested and involved in an issue but also a higher percentage of people at the local do as well because it has amplified into a wider discourse (Carver, 2003). Consequently Aitken (2002) suggests that instead of perceiving issues or decision making as being scale dependent and developing PPGIS projects for stakeholders which have their scale fixed, GIS projects should, alternatively, be directing their attention towards developing a GIS which would enable community issues and knowledge to jump scale from local to larger public discourses or vice versa (Aitken, 2002, Sieber, 2006). This is an important aspect because there is often a concern that local activities are dismissed as being part of community politics and are denied significant advancements by State and Federal Governments and thus the opportunity to emerge and engage individuals at all scales (Aitken, 2002). From this perspective a community-based GIS, where community is defined as "a group of individuals who are bound together by a common characteristic or a common intent and who enjoy a relatively high degree of mutual social interaction" (Jones et al., 2004: 105) offers the prospect of transcending the rigid scale conceptualised upon community politics and local activism enabling them to contest structures of power and dominance at the very scales they exist (Aitken, 2002, Gaile and Willmott, 2005). Harris and Weiner (1998) acknowledged in their research on the power relations associated with GIS use - that participatory GIS practices have the potential to simultaneously empower and marginalise groups (Sieber, 2006). As a result it is imperative to understand the nature of the participatory process and who benefits and why (Craig et al., 2002). One such way to help conceptualise the levels of public participation is through a ladder metaphor. First conceived by Arnstein (1969), the basic premise of the 'participation ladder' is that each rung of the ladder represents a different level of participation - the bottom rung represents zero opportunity to participate while each rung above represents increased level of participation in the decision making and thus greater public empowerment (Carver, 2003). Wiedemann and Femers (1993) later produced an adaptation of the ladder which conceived of public participation as not only providing access to information but also suggesting that informing the public of decisions is another form of participation (Tulloch and Shapiro, 2003). This concept is significantly flawed as it firstly misrepresents the commonly understood meaning of the word participation in PPGIS which The Collins English Dictionary (1982) defines as "to take part, be or become involved, or share." Secondly the ladder metaphors do not acknowledge the potential for participation to change over a period of time (Schlossberg and Shuford, 2005). Thirdly, the participation models fail to include oppositional groups whom do not cooperate with public decision making but participate in the formation public policy through other influential methods such as protests (Sieber, 2006). The incorporation of the word participation in many GIS projects implies a method of consensus building which presupposes a level of top-down decision-making as well as a degree of homogenization between participants. Certain individuals however may be better able to participate or contribute to decision making than others. Consequently, disproportional levels of participation may effectively disempower individuals and adversely affect the desired outcomes of a community (Sieber, 2006). Consequently some scholars have insisted on applying 'participatory' for autonomous grassroots activities and employing 'participation' to describe those projects which are more top-down in their approach (Elwood, 2006). Again while this is a neat way to demarcate GIS projects which employ various degrees of top-down and bottom-up methods - these definitions fail to acknowledge that both methodologies are crucial to any successful GIS project and community decision making. In fact it is fervently maintained that in order to enable citizens to better identify and comprehend how the role of GIS and technical discourses are bound up in decision making and how decision making can be informed by GIS knowledge, communities must have access to spatial information developed by Governments and commercial organisations as well as contributing their own spatial knowledge (Brown, 1998). Within this section it has been demonstrated that the attitudes and arguments that frame many PPGIS projects have succeeded in producing an illusion of influence and contribution by communities to decision making when actual control still resides with the traditional powers, such as Government. Instead of attempting to build an impossible consensus amongst a public with disparate tastes, values and experiences, a community-based GIS should concentrate on developing a community's ability to construct their own facts with the aid of available third party resources, from which their personal geographic stories may emerge and translate to various members of society (Wood, 2005). Another way forward could be to draw upon principles of Situational psychogeography which also attempts to combine subjective and objective modes of study by positing that one's self cannot be divorced from the urban environment and that one's psyche and knowledge of the city must transcend the individual if it is to be of any use in the collective rethinking of the city (Sadler, 1998, Wood, 2005). In the following section an examination of the origins of psychogeography will be conducted - clarifying how the principles behind this practice may help establish a framework for practice of GIS and Society and specifically the incorporation of local knowledge in GIS.
2.4 What exactly is Psychogeography? During the 1950s a number of highly politicised groups emerged in opposition to the ideals of modernism; these groups promoted programs that would reform the practice of art and life by directly intervening in the human environment and bringing about a social revolution (Sadler, 1998). One such group were the Lettrist International who conceived of the notion of Unitary Urbanism, which would later be the developed into the praxis of Psychogeography. Unitary Urbanism was envisaged as "the theory of the combined use of arts and techniques for the integral construction of a milieu in dynamic relation with experiments in behaviour." (Knabb, 2006: 52) In other words, Unitary Urbanism was considered a social project whose vision was the unification of space and architecture with the social and individual body (Sadler, 1998). In 1957 the Lettrist International and the International Movement for an Imaginist Bauhaus (IMIB) merged to form a new artistic-activist movement known as the Situationist International (SI). The SI was similarly critical of modernist principles which anteceded the rational mind at the expense of the imagination. These criticisms are most clearly evident in the SI's opposition to modern architecture and urban planning which they argued shaped people into rigid patterns of behaviour (Sadler, 1998). Furthermore they believed that increasing urbanism and capitalism had reduced life to mere production and consumption behaviour that ensured that "everything that was directly lived has moved away into a representation." (Debord, 1964) In other words experienced space had been reduced into mere representations of spaces and in turn re-envisaged as capitalist spaces (McDonough, 2002). SI believed that members of society were increasingly experiencing life as spectators devoid of dialogue and without a sense of being involved or interacting with one another. Once this 'spectacle' of modernity and urbanism, represented through images, products and activities, and authorised by the state, had been unveiled, society would be able to rediscover the authenticity of city life underneath (Debord, 1964). By resisting the hegemony of the state the SI sought to radically transform urban spaces through different practices including the subversion of cartography. Specifically by directing the spectator's senses towards the contradictions in the abstractions and mediations of the state, the aim was to draw the spectator "into activity by provoking his capacities to revolutionize his own life" (Debord, 1957: 25). Taking from the original methodology of Unitary Urbanism, psychogeography was proposed as a method of urban investigation which studies "the precise laws and specific effects of the geographical environment, consciously organised or not, on the emotions and behaviour of individuals." (Debord, 1955) In other words, psychogeography was intended as a methodology to help make people aware of the ways in which the urban environment and everyday life is conditioned and controlled and encouraged the exposing of these concerns (Plant, 1992). Psychogeography in practice utilised a technique conceived as the Theory of the DÃ©rive, in which individuals dÃ©rive (literally: 'drifting') through an environment letting themselves be drawn by the attractions of the terrain while still seeking to unmask the contradictions in the abstracted space (Plant, 1992). The dÃ©rive was an attempt to reappropriate the meaning of the city by removing the myths in the state's representations by having people walk and experience the landscape first hand, thus constructing through narratives a more concrete collective space (Mcdonough, 1994). While the dÃ©rive offered a new way of surveying urban space, a new way of representing these spaces had yet to be found. The SI were not disillusioned with the idea of mapping practices, in fact they regarded mapping as an important component to aid in the changing and organisation of urban spaces (Pinder, 1996). They believed however, that the structures and imperatives utilised in mapping exposed the desires of those wishing to impose order upon the city. The SI ambition was thus to illustrate the strange logic and apparent disorder of cities by producing maps which demonstrated those intimacies of the city typically absent from a traditional street or topographic map (Sadler, 1998). Consequently the SI developed a concept called DÃ©tournement, which loosely translates as a diversion or rerouting of pre-existing aesthetic elements (Knabb, 1995). An example of this is where existing maps and aerial photographs were juxtaposed or rearranged to produce a new spatial meaning; an alternative experiential or existential truth (Ungar, 2005). Thus the SI were able to reconstruct the cartography of a city by reconciling conventional geographies, sociologies, and cartographies together with experienced spaces, producing a map which is "terrestrial, fragmented, subjective, temporal, and cultural" (Sadler, 1998: 82). While Debord announced the disbandment of the SI in 1972, the traditions underpinning psychogeography continue to influence many works of literature, films, urban design and geographic practices (Ford, 2005). Wood (2005) draws attention to one contemporary psychogeography project - Jake Barton's 'City of Memory' - which combines psychogeographic principles with a GIS to build a collective urban memory through the participation of a number of people. In an interview with Wood, Jake Barton described his project as utilising 'top-down and 'bottom-up' resources to create an 'emergent' and 'curated' experience. Precisely by extending these terms to form the foundation of any GIS and Society project, Wood hypothesised that what would emerge was a GIS designed by a third-party or community-based intermediary (top-down); the public would formulate a specific framework that fits their unique goals (bottom-up); the bottom-up and top-down activities and goals are not independent of each other, but rather co-exist (curated); the outcome of the project has not been foreseen or influenced towards a specific outcome by any party, but rather emerges organically from the facts obtained and analysed (emergent). Thus resulting in a map and information which "has not been exactly made by the public but which without it has no content at all and deflates into a frame around nothing" (Wood, 2005: 13). Following on from this preliminary research by Wood a wider investigation of these terms will be conducted, laying the framework for a more appropriate community-based GIS as originally envisaged in GISoc debates.
2.5 Top-down Top-down integration of GIS is usually undertaken by an outside individual or agency who provides the GIS model, data, analysis and representation (Talen, 2000). Often the major distinction between a top-down and bottom-up approach, in participatory projects, is determined by where the decision making lies and by the level of commitment required by the public. With a top-down approach a Government or organisation would typically provide the data and representations which would be used in deliberation with the public, who are required to make a short-term commitment. In contrast, a bottom-up approach would require the public to have ongoing access to GIS data and the resources to capture data, conduct analysis and produce representations (Talen, 2000). Governments and commercial planners will often implement a participatory GIS with top-down goals in order to better understand a neighbourhood dynamic, improve public sector management and enhance social service provision. This process theoretically serves the public by introducing policies and services based on a community's perception of the data, analysis and representations framed by Governments and planners (Sieber, 2006). Top-down GIS models can also help circumvent deterrents such as cost, complexity and access to data which often impeded non-profit and community groups from implementing a GIS. The cost of hardware and GIS software have decreased dramatically over the years and there are now many open source GIS solutions available for free use; however it has been shown that any cost and resources required in the implementation, operation and maintenance of equipment, no matter the amount, will be a significant barrier for adoption, especially for underprivileged groups (Brodnig and Mayer-SchÃ¶nberger, 2000, Leitner et al., 2002). Furthermore many individuals may lack knowledge about the availability and means of obtaining a GIS and spatial data (Elwood, 2007). Many of the GIS packages available are user-friendly for many operations, however the more functionality a group requires for their GIS, the greater the need for experts to help operate the software or at a minimum provide training; this constitutes another investment of time and money by the community group (Brodnig and Mayer-SchÃ¶nberger, 2000, Leitner et al., 2002). It has been noted that a large amount of spatial data available for use within GIS projects has been collected and manipulated by Governments or commercial institutions. Consequently some researchers also consider Government and commercially generated data which is made available to local residents for use within community GIS projects to be subsumed into the wider definition of top-down PPGIS (Sieber, 2006, Talen, 1999). In theory access to secondary data represents opportunities for community groups to acquire accurate and current spatial data in an appropriate format, increasing the efficiency in implementing a GIS and performing analysis (Sieber, 2000a). Secondary datasets however, are not always publicly available due to technical, organisational, political, legal and economic factors (Sieber, 2007). The availability of data is a concern for any GIS project and access may prove problematic depending on legislative rules on public data in different states and countries (Sieber, 2007). Even if data can be easily accessed, other concerns can arise regarding the appropriateness of the data and costs involved (Sieber, 2006). An attempt to provide improved data access to users has been the implementation of Spatial Data Infrastructures (SDIs), which are generally setup and maintained by Governments and private institutions. Australia's SDI is known as the Australian Spatial Data Infrastructure [ASDI] and is managed by The Spatial Information Council of Australia and New Zealand [ANZLIC] (formerly known as the Australia New Zealand Land Information Council). The goal of a SDI is to enable users at all levels, both within and between countries, to locate, access and visualise spatial information within a centralised location. SDIs are typically implemented on the internet because it allows users to interact with the data transparently from a multitude of locations. SDIs are also cost effective and efficient because the spatial data is centrally managed, meaning there is no collection effort required and data standards are established allowing for greater portability and interoperability (Hall and Leahy, 2008). Appropriate access to data sources - while important - can be futile if the data available is not appropriate for the intended application. This is often the case because Government and commercial strategies and priorities are often very different from those of community groups, leading to data which does not fully represent the perceptions and attributes which are meaningful to a community (Elwood, 2007). Furthermore for data to be appropriate, users must determine whether the available data is sufficiently complete and accurate; not all groups will demand high positional accuracy, but many will demand high degrees of attributional accuracy. Data access and appropriateness is also often prejudiced by its format and organisation. For instance, data may be classified or aggregated and thus potentially inappropriate (Sieber, 2007). Just as commonly, spatial data may not be at a scale or resolution which is meaningful or useable for local level applications characteristic of community group projects (Elwood, 2007). Local governments constitute the most likely source of secondary data, however access to this data is heavily influenced by the local political climate. This means that if a non-profit or community group has alienated itself from the government or data suppliers this may preclude them from having access to the data (Sieber, 2007). Many Governments may also restrict the distribution of data because of concerns of privacy, security and liability because of inaccuracies (Sieber, 2007). There are continuing debates regarding whether the data that has been produced by Governments should be able to be accessed by the public domain or whether the data can be sold to buyers as a form of commodity. Again, costs for accessing public databases may preclude community groups from implementing a GIS, however it has been demonstrated in much research that often communities are able to obtain secondary data through informal channels and close interpersonal linkages (Sieber, 2007). In summary, access to secondary data from Governments and private organisations made available through SDI and other channels, can constitute an effective method for populating a GIS and undertaking analysis which can effectively enhance a community's activities or perceptions of their neighbourhood or region. However not all secondary data is appropriate and many communities will need to obtain and collate their own spatial data to operate their GIS effectively. That said, as Debord illuminated in his account of dÃ©tournement, "any elements no matter where they are taken from, can serve in making new combinations"; meaning contemporary GIS secondary data can be manipulated or transformed by community groups to befit their needs Next, by adding their own perceptions to the data, communities are able to reappropriate the GIS so that it successfully represents their perceptions (Elwood, 2007). Thus secondary data can be extremely useful in adding context to a situation, such as obtaining road lines or parcel data, which communities can use to plot dangerous intersections or problem areas. From this point we now turn and see how community knowledge can be integrated through a bottom-up approach.
2.6 Bottom-up Much research has focused on the technical issues of data collection, accuracy, interoperability and access however the production and dissemination of data represents more than technical issues but also the social and cultural relations which exist within contemporary society. From this perspective GIS projects which use secondary spatial datasets exclusively can be problematic because the collection, categorisation and distribution of data conducted by Governments and private institutions are based on their own agendas and decisions; that is what is included and what is not, meaning that only a single geography can be produced (Pickles, 1995). Sceptics of top-down GIS also attribute its failure to approaches which inform communities rather than incorporate communities in GIS practice and decision making, thus marginalising local knowledge (Talen, 2000). Furthermore this method excludes communities from ever having ownership of a GIS, which many commentators consider to be essential to successful adoption and empowerment (Sieber, 2000a). Subsequently much research has focused on alternatives to top-down GIS approaches in which local knowledge and narratives are encapsulated and fundamental to the operations of a GIS (Talen, 1999). These bottom-up GIS approaches allow communities to characterise their environment and communicate community perspectives, rather than (re)convey Government generated 'objective' data (Talen, 2000). Local knowledge, which according to Mailhot (1993) is "the sum of the data and ideas acquired by a human group on its environment as a result of the group's use and occupation of a region over many generations" is fundamental to any project which studies the local environment because it can be such a rich source of information. Furthermore, this situated knowledge may help determine the meanings and values which local residents attribute to particular features, thus lending insight into how an area is conceptualised (Talen, 1999). GIS can play a central role in supporting the communication of local spatial knowledge because it provides an effective means of analysing and representing spatial complexity and can help discover relationships which may validate community needs and activities (Talen, 2000). Bottom-up GIS is not a method for replacing local meaning; rather it helps strengthen perceptions and depth of communication about community issues and preferences (Talen, 2000). By providing communities with the means to evaluate their area and articulate their perceptions spatially they are equipped with a much more complex vocabulary for which to discuss pubic issues (Talen, 2000). Local knowledge, though, often exists in the form of stories and images and as such is qualitative data. GIS use would then seem problematic because a GIS is largely understood as a tool for the storage, analysis and representation of quantitative data. GIS does however have the capacity to integrate qualitative data and practices because it can handle other types of information such as photographs, videos and narratives (Kwan and Knigge, 2006). According to Kwan (2002: 274) the "representational possibilities of GIS can be used for enacting creative discursive tactics that disrupt the dualist understanding of geographical methods, where visual images (albeit generated and composed with digital technology), words and numbers are used together to compose contextualized cartographic narratives in geographical discourse." Thus the primary task in implementing a bottom-up GIS is to create ways in which community perceptions can be collected and translated into data which the GIS can store and represent (Talen, 2000). The SI practice of dÃ©rive can be useful for the collection of local data because it is a purposeful method for scouting and retrieving information about an area (Coverley, 2006). In a dÃ©rive, one or more persons can delimit a spatial area and let themselves be drawn by the attractions of the landscape enabling them to analyse the disregarded characteristics of an area (Debord, 1956). Using the act of walking and experiencing a space first hand, communities are able to reappropriate spaces and unmask the abstract spaces represented in many GIS datasets. Then, as Debord (1956) hypothesised "with the aid of old maps, aerial photographs and experimental dÃ©rives, one can draw up hitherto lacking maps of influences." The observations and narratives of a community can be captured and integrated into a GIS through methods such as questionnaires, focus groups and mental maps. A mental map can be defined as the unique perceptual maps which exist in everyone's mind, reflecting our spatial perceptions and knowledge of local, regional and global geographies. For instance we often develop detailed maps of our local neighbourhood which we use to navigate and communicate routes of interest such as getting from home to work. Mental maps are not generally considered on a conscious level but are continually refined over time and we intuitively use these maps to make decisions everyday and through our lives (Maantay and Ziegler, 2006). As suggested by Lynch (2002) in his text 'the image of the city' there is a public image of any given city space which has been produced through many individual images. These images are unique and are necessary for any individual to function effectively within their environment, but for most people these images are not appropriately captured by maps produced by the State. While maps may contain some significant landmarks, they will often omit 'local landmarks' - those only visible in restricted localities and from certain approaches. Examples of local landmarks include trees, seats, store fronts and other urban details. These local landmarks are invaluable references, especially to local communities, as these referents are frequently used clues of identity and often become increasingly relied upon as the area becomes more and more familiar (Lynch, 2002). Additionally, networks of citizens or communities as everyday observers of a landscape have situated knowledge which can constitute a source of data that in many cases is much more detailed and current than any current satellite imagery is able to offer and has the potential to be a significant source of knowledge which would assist scientists and geographers to further their understanding of many social and environmental factors. Local observation and data may also reveal and clarify local life and activities which go unnoticed by the national mapping agencies or the media (Goodchild, 2007). Mental maps can be extremely valuable when used in the development of a bottom-up GIS because they can help determine important areas of the environment and a consensus as to what communities' priorities are and what a community considers important to preserve, improve or invest in. By compiling and drawing a composite community mental map it may help elucidate where resources should be directed and as well as developing a visual portrayal of a communities vision (Maantay and Ziegler, 2006).
2.7 Curated As the previous two sections demonstrate, there are a number of competing goals between top-down and bottom-up GIS approaches. In this section an attempt to relate the curator (top-down) and community (bottom-up) to form a collective representation of the landscape will be expounded. In the text 'Community participation and Geographic Information Systems' Leitner et al (2002) summarises six models for making GIS available to community organisations. These models include: Community-based (in-house) GIS; University-community partnerships; GIS facilities in Universities and public libraries; map rooms; Internet map severs; and neighbourhood GIS centres. Each of these models differs from one another along numerous dimensions including location, communication structures, and interaction, however one commonality that each of these models does have is the necessity for curatorial practices. Curators have traditionally been responsible for the acquisition, authorship and representation of objects to audiences within specialised spaces such as museums, libraries, art galleries and other cultural institutions. While this definition is still true in many respects, the curator's role has also shifted and expanded in response to environments where technology is more pervasive and greater interactions with audiences are required. As such Russo and Watkins (2004) suggest that curatorial practices have been reconsidered around the following key areas: 1. A curator must be able to facilitate community engagement in the production and display of interactive experiences 2. A curator must be able to act as agents of technology transfer in collaboration with new media technologists and most importantly 3. A curator must be able to employ a structured methodology for enabling the first two practices to be delivered. Using this line of thinking the six GIS models can be explicated as methods in which curators establish contexts which can capture accurately the needs of user groups and allow them to provide their own contexts. In this sense each actor involved in the implementation of these models has in some way assumed the role of a curator through interpretation, selection and filtering of data and by creating chains of meaning through data association and juxtaposition (Paul, 2006, Scholz, 2006). Curatorial practices are also discernable in more abstract models such as the Internet map server which can be understood both literally and metaphorically as a form of 'software curation' - "a practice that is partially automated, dynamic, collaborative, and redistributed in terms of power relations and curatorial control." (Krysa, 2006: 10) Even though the construction of situations was necessarily collective in its preparation and development the role of a curator was also acknowledged as being fundamental to the success of psychogeography and dÃ©rives by the Situationist International. The curator would be "responsible for coordinating the basic elements necessary for the construction of the dÃ©cor and for working out certain interventions in the events." (Knabb, 1995: 43) It was also noted however, that the relationship between the curator and public was only a temporarily subordination rather than a permanent specialization (Knabb, 1995). The role of the curator is also somewhat analogous to a 'GIS champion' which much existing PPGIS literature acknowledges as being key to any successful GIS implementation. The GIS champion plays a crucial role in the utility of GIS because they generally are best suited to successfully drive the implementation of the GIS and help align GIS practices with the internal goals of the community (Sieber, 2000a). The GIS champion will also help promote the technology through the group and continue to gather appropriate data and allow the GIS practices to mature and be adopted. The organising assumptions behind a curated GIS lies in its ability to provide the means by which users can explore an environment, which integrates information from a multitude of sources, so that the user can view other peoples data, experiment and imagine possible solutions and formulate their own views (Carver, 2003). In order, however, to model a community-based GIS on this framework we have to acknowledge that the goal is not to introduce a GIS into a social structure or even for the social structure to utilise GIS as merely a tool but rather ensure that the GIS has been appropriated by the social structure (Wood, 2005). This is because creation of local data and providing access to State data does not directly guarantee or endow a community with the power to emancipate themselves from the decision making bureaucracies. By transforming the GIS practice and by inventing their own processes a community may be able to construct an otherwise unattainable knowledge (Sieber, 2000a). In essence this means that the fundamental role of the curator in community-based GIS is to produce engaging community outcomes, while also protecting the interests of all participants, through the seamless integration of quantitative and qualitative data and methods (Klaebe et al., 2007). Situational psychogeography acknowledged this ability to combine subjective and objective modes of study by positing that ones self cannot be divorced from the urban environment and that local knowledge must transcend the individual if it is to be of any use in the collective rethinking of the city (Sadler, 1998). As such the Situationists developed the notion of dÃ©tournement as a method of representing psychogeographies acknowledging that "the mutual interference of two worlds of feeling, or the bringing together of two independent expressions, supersedes the original elements and produces a synthetic organisation of greater efficacy." (Debord and Wolman, 2006: 15) So by employing mixed data types - quantitative and qualitative - a GIS will be strengthened, even in instances when the datasets are contradictory, because the community will then be able to gather a better understanding of the inherent inconsistencies and partiality in both data methods. In some instances both sets of data may in fact be accurate yet were produced from different perspectives (Knigge and Cope, 2006). Additionally, because mental maps are subjective and reflect a person's background, experiences and knowledge, everyone's mental maps will differ significantly meaning community data will be diverse and contain its own inaccuracies. A person's mental map will often not reflect geographic reality but rather our perception of the geography; for instance distances of length are something which may become distorted in our mental maps. This is not because people have a lack of skill in cognitively measuring distances, but distances are often dependent on our perception of the distance such as aesthetically pleasing, dark, safe, uphill, rough terrain, a number of factors (Maantay and Ziegler, 2006). By integrating and organising collections of mental maps and narratives into recognisable data as well as co-opting state data, curators will be able to help contextualise the cartographic knowledge of both the community and the State through the mediation of their relationship, hopefully increasing the authenticity of a community's activities (Knigge and Cope, 2006). In this sense, the Curator can act as a hub which a communities GIS can be built and related and eventually users will be able to explore their own ideas and reach their own conclusions or a consensus may emerge which can lead to effective decision making (Russo et al., 2008).
2.8 Emergent In dialogue with Wood (2005) Jake Barton describes his 'City of Memory' mapping project as utilizing resources and citizen participation to generate an emergent effect or an outcome which has not been foreseen. The use of the adjective emergent within this context and the context of community mapping is extremely appropriate given that it implies an impartial and democratic quality - often absent from many PPGIS projects - and also according to The Collins English Dictionary (1982) meaning something that is "coming into being noticed" or an entity "recently independent" . The desire for an emergent outcome while not unique is often understated as a key objective of GIS and participatory mapping projects. Rather the majority of PPGIS literature posits that the goal for community groups starts and ends with empowerment. While there is no universal meaning of empowerment in PPGIS applications Carver (2003) defines it as "the process by which stakeholders identify and shape their lives and the society in which they live through access to knowledge, political processes and financial, social and natural resources." Participation in decision making and GIS use, however, does not always result in empowering a community. This is because the motives of various individuals, including traditional power holders, may undermine the process through unequal power relations or perverse tactics such as limiting access to information (Carver, 2003). Furthermore it has also been noted that groups contain numerous outcomes and goals and these goals will vary depending on a communities particular culture and personal ideologies (Sieber, 2006). Craig and Elwood (1998) noted that the goals of community based groups implementing GIS can often be classified into four distinct categories: administrative (e.g. locate members or activities), organisational (e.g. recruit members or obtain grants), tactical (e.g. search for suitable location) and strategic (e.g. evaluate success of activities). It should also be noted that there are many other groups whose goals are less tangible, sometimes competing and contradictory and cannot necessarily be easily measured (Sieber, 2006). It would also be too simplistic and problematic to acknowledge that each individual whom contributes to a community-based GIS has a homogenised set of values and beliefs. Positioning emergence as the objective of any community GIS project will ensure that the loftier goals of a community are not misconstrued in favour of technocratic driven outcomes. By utilising top-down and bottom-up approaches more actors are involved in the process and as such various narratives can be incorporated increasing the likelihood that an emergent outcome results (Wood, 2005). Emergent outcomes are desirable because they postulate that no one group influence the process of decision-making more than another. Instead emergent outcomes presuppose that a number of narratives and perspectives are captured enabling an outcome to emerge, not necessarily of consensus, but one which truthfully represents the spatial knowledge of the group. Debord (1957) highlights this important characteristic within the SI original conception of 'constructing situations' through the methods of unitary urbanism and psychogeography: "we have neither guaranteed recipes nor definitive results. We only propose an experimental research to be collectively led in a few directions that we are presently defining and toward others that have yet to be defined" (Knabb, 1995: 25) These democratic outcomes are reliant and characterised by the mixed methods approach described in the previous chapter. By compelling a community to explore and query diverse datasets from multiple angles and to deliberate on any prominent themes, consistencies or discrepancies, it enables not only connections between space and community identity to be revealed, but also multiple versions of reality to emerge (Knigge and Cope, 2006). This methodology also implies an ongoing process and continual renegotiation and reflection. This approach in psychogeographic projects is paramount because research on the physical environment "entails bold hypotheses that must constantly be corrected in light of experience, by critique and self-critique." (Knabb, 1995: 7) By thinking about a community-based GIS as long term negotiated creations of 'process cartography' we can conceptualise the process of mapping as an ongoing "dialogue amongst residents" and "where the focus is on the dynamics by which terrain features are created and made influential." (Rundstrom, 1991, Offen, 2003) In this sense the process of mapping at the local level can allow the identity of a place to emerge. Another central aim of the SI was to develop a " concrete construction of temporary settings of life and their transformation into a higher passionate nature." (Ford, 2005: 50) Similarly, an emergent outcome presupposes that the activities of a community should not be limited to a particular group or fixed to a particular scale but should in fact open the possibility of coming into being or noticed by society at various scales. In other words the outcomes associated with a community-based GIS should allow the community to mobilise (if required) and jump-scale from the local to global or regional to local - whatever the case may be (Aitken, 2002). Jumping of scales has been often addressed in relation to politics of scale and as a result a distinction was developed between spaces of dependence and spaces of engagement. Spaces of dependence refer to the social relations which people depend for the realisation of essential interests; spaces of engagement refers to the spaces for which people might utilise (or develop) networks of associations in order to further the social practices of the community (Cox, 1998, Haarstad and Floysand, 2007). Within spaces of engagement individuals or communities are able to re-articulate their narratives and knowledge, essentially jumping scale, and transcending the abstract boundaries enforced on particular groups (Haarstad and Floysand, 2007). It should be noted again that jumping scale does not presuppose a greater level empowerment, but maybe a potential effect of developing social relations and communicating knowledge and claims across scales (Haarstad and Floysand, 2007, Aitken, 2002). By capturing knowledge of a community and co-opted quantitative data of the state within a GIS, communities will be able to communicate spatial stories which elucidate and perhaps "promote a renegotiation of positions of power in politics." (Haarstad and Floysand, 2007: 305). In other words community groups having redefined their GIS may able to use their perceived greater legitimacy to impart an ideological framework - different from the dominant thought - to the public inspiring them to adopt the agenda of the community (Sieber, 2000a).
2.9 Conclusion This chapter has explored the early development of PPGIS and provided an overview of the inherent ambiguities in contemporary PPGIS praxis. Additionally, a contextual account of psychogeography was presented, relating how some of its core principles maybe applied to the implementation of a community-based GIS. Specifically four key terms were identified, by way of an innovative mapping project called 'City of Memory', and explored in some detail to help better frame the examination of psychogeography and GIS. With these terms in mind the next chapter will apply the theoretical framework to the case study involving the BLS, hoping to verify or disprove many of the hypotheses contained within this chapter.
3. Case Study
3.1 Introduction "He wondered what the map would look like of all the steps he had taken in his life and what word it would spell." (Auster, 1992: 129) This chapter presents a case study of GIS implementation within a local community and applies the methodologies and theories expounded within Chapter Two. It is anticipated that the theory explored and developed within the previous chapter will provide a robust framework to not only guide the implementation of a GIS for the BLS Advisory Committee but also ensure any findings can be related to historic and future investigations.
3.2 Context Blackburn Lake Sanctuary (BLS) is located within Blackburn a suburb of Melbourne (Victoria, Australia) almost 20kms east of the CBD. The BLS site is approximately 75 acres (30.4ha) of remanent and regenerated bushland and as such is a very important environmental asset to the Whitehorse City Council and its local residents. The significance of the Sanctuary however, extends beyond the local community and to the wider Melbourne metropolitan area given that it is one of the few remaining areas of natural urban bushland (Dempsey, 2009b, Dempsey and Rees, 2009). The Sanctuary contains a great level of biodiversity, including a large variety of indigenous and native vegetation as well as supporting an environmental niche - the Valley Heathy Forest - an Ecological Vegetation Class (EVC) identified by the Victorian Government. BLS is also home to a wide range of animals, insects, fish, reptiles and is one of the most important bird refuges in metropolitan Melbourne (Dempsey, 2009a). It is also an important historical site having had many owners and uses since the lake was formed in 1889 by the damming of Gardiner's Creek. One such owner was the Adult Deaf Society of Victoria whom in 1909 used the area as a place for people to live and work in, growing flowers and vegetables for selling at the Victoria Market (Dempsey and Rees, 2009). The Sanctuary is currently owned by the City of Whitehorse, however it has appointed an Advisory Committee consisting of interested local resident volunteers to assist in the monitoring, enhancement and maintenance of the Sanctuary through working bees and other environmental activities (Council, 2009). Key activities of the Advisory Committee also include promoting the environmental and ecological values of the Sanctuary by providing relevant community and school education programs, preparing presentations and displays at the Visitor Centre and conducting monthly community activities. Through their daily experiences and in-depth understanding of the Sanctuary's contemporary and historic environment, the Advisory Committee are also able to liaise between the community and the council on issues relevant to the BLS (QUESTIONAIREE). Much of the information regarding the BLS is confined to the experiences and knowledge of a number of local residents and Council members, however as time has come to pass much of this historical and local knowledge has vanished as individuals have moved or passed on. Furthermore from the Advisory Committee's experience increased Council staff turnover has contributed too many of the council records and knowledge regarding the Sanctuary disappearing or becoming untraceable (QUESTIONAIRE). Realising that much of this information is invaluable to the Sanctuary, community and Council the Advisory Committee approached RMIT University, by way of some common contacts, to provide advice on ways in which information could be collected, archived and used by the Advisory Committee to aid their activities. Specifically the Advisory Committee was hoping to develop a central depository containing high-quality information on plants, including their locations, which could subsequently be used to provide vital advice to Council and parks people, coordinate planning planting programs, weed management, seed collections and direct working bees. By having information on significant plant species located within BLS, it is hoped that the Advisory Committee could also assist in determining policies and planning applications as well as presenting cases to VCAT (Victorian Civil and Administrative Tribunal) (QUESTIONAIRE). Given the spatial component (e.g. plant locations) involved in the activities of the Advisory Committee, a GIS was recommended as a tool which the Advisory Committee could incorporate into their activities and use to capture, store and represent their local spatial knowledge. Through a community-University collaboration it was envisaged that a GIS could be implemented using the theoretical framework developed in the previous chapters. The case study provides a unique opportunity to compare how a community-based GIS can be implemented within an urban area using the principles of participatory GIS, community mapping and psychogeography. Results from this case study may also help future projects recognise the challenges in implementing a community-based GIS as well as the contributions of a GIS to a community group's goals and missions.
3.3 Research methods This case study was conducted using action research as the principle methodology of investigation. An agile action-research based approach was chosen because it enables "the application of fact finding to practical problem solving in a social situation with a view to improving the quality of action within it, involving the collaboration and co-operation of researchers, practitioners and laymen." (Burns, 1990: 252) This type of action-research can also be characterised as situational (it diagnoses a community issues and attempts to solve it), collaborative (because it requires the resources of researchers and the community), participatory (since researchers and the community take part in the implementation of the findings) and self evaluative (in that the research requires ongoing evaluation and modifications should be made to adjust to new scenarios and practices) (Sarantakos, 1998) The work towards a BLS community-based GIS was initiated with a requirements analysis based on unstructured interviews and a distributed questionnaire that went to multiple members of the Advisory Committee. A requirements analysis was deemed crucial because the process enabled all parties to determine the needs of the GIS, including which functions they wished to perform, who may benefit from the GIS, the data that will be needed, the accuracy, precision and completeness of the data required, and the appropriate software and hardware needed to support the system (Harmon and Anderson, 2003). The analysis was then followed with the formulation of requirements, which were continuously validated in numerous meetings, with the Advisory Committee, throughout the life of the project. The next section will discuss the results of the requirements analysis and some of the findings during the development of the GIS utilising the structure of the discussion within the literature review: top-down, bottom-up, curated and emergent.
3.4 Top-down Results from the meetings and questionnaire revealed that the Advisory Committee had numerous records regarding vegetation, bird surveys, fauna records, fungi records, water monitoring and historic meeting minutes, however none of this data was in a format easily recognisable by a GIS. The Advisory Committee also possessed aerial photography procured from the Department of Sustainability and Environment (DSE) in 2006. This aerial photograph was in both hardcopy and digital form and thus could be used within the GIS. The digital image, however, was of very low resolution and was extremely limited in its usefulness (e.g. digitizing) however was still useful in providing context to locational information. Due to the lack of GIS data for which to add context to the data collected by the Advisory Committee additional data had to be captured or sourced from a 3rd party. The datasets which needed to be procured included: * Roads * Waterways * Parcel * Paths * Contours Given that the BLS covers only a small area it was determined that one underlying factor influencing any data obtained would be its necessity to have been digitised at a scale of 1k or less. As indicated within the literature review a community's ability to gain access to spatial data resources to fulfil their needs is crucial to the success of their GIS project. Spatial Data Infrastructures have become the dominant framework within many countries - including Australia - for encouraging access to standardised spatial datasets. As such an investigation into the practicality of accessing and leveraging spatial data from Australia's Spatial Data Infrastructure (ASDI) was considered to be extremely pertinent. According to the ANLIC website the ASDI "comprises the people, policies and technologies necessary to enable the use of spatially referenced data through all levels of government, the private sector, non-profit organisations and academia." (2009: para 1) Thus the key components of the ASDI include: the Australia Spatial Data Directory (ASDD); Standards; and Spatial metadata. The ASDD (http://www.ga.gov.au/asdd/) is the gateway which enables individuals to discover appropriate spatial datasets by simply searching the associated metadata using various search terms. Government and commercial organisations are responsible for the maintenance of their individual nodes within the ASDD. As of the August 2009 there were the following 21 ASDD nodes: * ACT Geographic Data Directory * Australian Hydrographic Service - Product Metadata * Australian Hydrographic Service - Publication Metadata * Australian Hydrographic Service - Source Metadata * BRS and Australian Natural Resources Data Library (ANRDL) * Bureau of Meteorology * CSIRO Marine and Atmospheric Research * DEWHA Discover Information Geographically (DIG) * Geoscience Australia * IndexGeo Pty Ltd - Eco Companion catalogue * Murray-Darling Basin Authority * NSW Natural Resources Data Directory * NT Spatial Data Directory * Other Commonwealth Agencies (hosted by BRS) * PSMA Australia Limited * Qld Spatial Data (QSIC) * Queensland Department of Natural Resources and Water Data * RAN Directorate of Oceanography and Meteorology * South Australian Spatial Information Directory * Tasmanian Spatial Data Directory * Victorian Spatial Data Directory Upon searching the ASDD for the required spatial datasets it was quickly realised that the custodians of the required data were the Victorian Spatial Data Directory (VSDD). Navigating to the VSDD website revealed that the VSDD provides Vicmap data via their Land Channel website (http://www.land.vic.gov.au). While the data was conveniently and immediately available for download it was found that the data incurred a fee which varied according to the dataset and the extent of the data required; procuring property data for the Blackburn locality incurred a cost of AU$153.89 (inclusive of GST). While this does not seem too expensive or excessive, tallying up the costs for each of the required datasets and factor in that the geographic extent of the project crosses multiple localities, the increasing cost of obtaining these datasets became a discouragement to the relatively resource-poor community group. As such it was decided that the data should be obtained via informal channels such as local government contacts and RMIT University's arrangement with the State Government of Victoria. In this instance the datasets obtained from RMIT via an arrangement with the Department of Sustainability and Environment (DSE) were: * Land parcels * Roads * Creeks and Lakes * 2005 EVCs (Native Vegetation - Modelled 2005 Ecological Vegetation Classes) * Contours (10m) After all the data was obtained and uploaded into the GIS it was 'cut' so that only spatial data related to the BLS and surrounding suburbs was kept. The data was then examined and critiqued by the Advisory Committee whom noticed a number of ambiguities, including the generalization of the lake's shape and the lack of data regarding the wetlands towards the east of the lake. Additionally it was noted that the contour data was not accurate in the sense that it didn't take into account the road running along the western boundary of the Sanctuary. Consequently a number of changes had to be made to the data - using the knowledge of the Advisory Committee and the aerial photograph - in order to increase the accuracy of the obtained data and ensure it aligned with the knowledge and needs of the Advisory Committee. As a result, even before adding data directly collected by the Advisory Committee, it could be argued that the GIS had been transformed into a unique system which did not resemble any of the online interactive GIS offered by Government. In fact in some instances the data had become more accurate and precise given the local knowledge of the Advisory Committee. One of the key datasets identified that could not be obtained with any accuracy from any sources were the sanctuary's pathways. As such it was decided that this dataset needed to be captured using GPS or digitised from high resolution aerial photography. While this was not surprising it does indicate somewhat that the current SDI offered in Australia does not adequately fulfil the needs of community groups - especially groups whom are resource poor. This drawback of the SDI also showcases a lack of public involvement and contribution of spatial data by diverse groups unattached to Governments and private institutions. This could be due to a lack of functionality regarding the ability for diverse groups to add data to the SDI.
3.5 Bottom-up Much of the vegetation data collected by the associated levels of Government in Australia have been classified into various categories known as Ecological Vegetation Classes (EVCs). An EVC consists of a number of vegetation communities which can be characterised by recognisable environmental attributes and ecological processes (Industries, 2008). There are approximately 300 EVCs associated with the state of Victoria however the vegetation which the Advisory Committee considers important and significant are not necessarily contained within any of the formulated EVCs and thus is inappropriate within the scope of the Advisory Committees objectives. While in the future it may be useful to included specific vegetation which conform and correlate to specific vegetation significance or importance registers endorsed by the Whitehorse City Council or State Government, the Advisory Committee did not desire to collect data representative only of these vegetation registers or EVCs but instead sought a fairly detailed vegetation list which could be adapted in the future. They were also hopeful that in the future they would be able contribute to significance lists based on their specific knowledge of the vegetation in the region. Collecting information of all of the vegetation within the BLS was deemed unrealistic and as such the Advisory Committee formulated a detailed list of the specific species that they wanted to capture within a GIS and the types of associated attribute information. It was also decided that a GPS should be utilised to capture the location of the vegetation because fairly accurate positional information was a necessity. This decision was made because one of the fundamental problems the Advisory Committee were having was precisely representing the locations of the vegetation on maps which had been observed in the field. This was because each individual often had cognitively different views of where they were in relation to a topographic map, creating conflicting accounts of where particular diminutive vegetation species were located and making re-finding them extremely difficult. Consequently a hand held GPS (TDS Nomad) with ArcPad software was employed as well as a TSC1 Asset surveyor Software Pathfinder Pro XRS when more precise locations were required. Once enough local vegetation locations were collected the data was uploaded into ArcGIS and edited so that the table schema (appendix ***) reflected the types of associated attribute information required by the Advisory Committee. Next, the collected the data was juxtaposed with the obtained Government data and presented to the Advisory Committee for comment. One of the findings of that demonstration was that when zooming in on specific areas of the Sanctuary it became very difficult for the Advisory Committee to orient themselves with their own internal cognitive map. For instance, the orientation of the north arrow within the GIS did not correlate with their day-to-day reference to north. For many North was orientated directly along a road which ran along the western boundary of the Sanctuary however the direction of this road actually corresponded to North-West in the GIS representation. Secondly it was found that abstract boundaries, paths, bridges, lake and streams were not significant enough landmarks to help orient individuals. Subsequently much more data was needed to be collected which correlated with the Advisory Committee's own mental map. This was achieved by inviting members of the Advisory Committee to sketch their mental maps of the BLS including their perceptions of significant landmarks. From these sketched mental maps many landmarks not originally considered were found to be important to the members of the Advisory Committee; these features included playgrounds, seats, gates, observation points and the bmx tracks. Defining the accuracy and precision of the data was also important because it could be added to the metadata - to give more information and comfort to any future users. In fact once the datasets were constructed it was important to sufficiently document and characterise the data using Australian/New Zealand metadata standards (AS/NZS ISO 19115:2005). Subsequently enabling future users whom are unfamiliar with the data to develop a better understanding of how and when the data was collected as well as the meaning of the data's attributes, thus enabling them to apply the data appropriately and efficiently.
3.6 Curated The primary role of the Curator (or GIS expert) was to guide the Advisory Committee through the implementation process, educating and broadening the GIS discourse and helping translate the views of the group into meaningful GIS data. As stated previously, the role of the curator has somewhat shifted from being suppliers of information to providers of usable tools, technology and knowledge which allow people to explore their own ideas and formulate their own conclusions (Russo et al., 2008). Thus the choosing of the most appropriate tools to represent the data was considered paramount to the success of the project. Some of the guidelines and objectives which were established during the requirements analysis phase to ensure the most appropriate GIS were chosen included exported data format, price and functionality. The key factors which influenced the decision regarding the format of the data and the type of GIS to be implemented was the Advisory Committee's desire to have a GIS which was compatible and interoperable with the GIS operated by the Whitehorse City Council (ArcGIS). The reason for this was to increase the legitimacy of the committee's activities and also increase the possibility of directly influencing decision making by the Council by contributing data directly to their GIS. In other words the GIS implemented had to be able to export data to either ArcGIS Shapefiles or other open-standard interchange formats such as GML (Geographic Markup Language). With any project, an organization, government or community must decide whether the completed projects benefits will outweigh the time and cost of its implementation. Each group or individual will have different sized budgets and goals set for the project. Often the cost of proprietary software will excluded some groups from ever experiencing its benefits. However over the past decade there has been a major push by various groups (such as software enthusiasts, universities, governments and organisations) to develop software which is 'open source' and can be used by individuals and groups for much less money or for free. Open source software by its basic definition means that users have access to the software's source code. For community organisations, not-for-profit organisations or even the lay user open source software may be best option (Steiniger and Bocher, 2009, Hall and Leahy, 2008, Neteler and Mitasova, 2004). In the case of the Advisory Committee the GIS had to be fairly inexpensive and such only three viable options were considered: 1) obtain community conservation grant from ESRI in order to obtain ArcGIS; 2) obtain funding or an available GIS license via the Whitehorse City Council; and 3) implement an open source GIS freely available on the internet. After reviewing their budget, the Advisory Committee came to the conclusion that it would be still too costly to obtain any proprietary software, even through the ESRI community conservation grant. This was because the grants offered by ESRI still incurred a payment of $275 for ArcView Single use ($418 for each additional extension) and $1199 for maintenance. Also obtaining a license through the Whitehorse City Council did not eventuate because of a range of factors including cost and inability to accommodate the Advisory Committee under their current licensing agreement with ESRI. Hence it was decided that the implementation of an open source GIS was the only cost effective option. However it was still unclear whether the available open source GIS's would be able to adequately perform all which the Advisory Committee asked. As such an investigation into a number of open source GIS products needed to be performed ensuring they had the functionality required, high usability, could export the data to the required formats and had a vibrant user community for which to rely on for any updates or queries. In terms of GIS, open source software has emerged as a viable option over the past few years. There has been a major push by many groups to develop open source GIS. As of 30 January 2009 the FreeGIS Organisation (freeGIS.org) listed 348 GIS related free or open source projects. Given the plethora of 'free' GIS options how would a group narrow the list down to one suitable system? There are a number of ways to delineate and classify GIS software to help narrow the choice these - some examples include: * Based on the development language (i.e. C, C++, java etc) * Whether the project was founded and now marinated by a commercial or non-commercial group. * What operating system the GIS has been developed * What purpose the GIS server was developed (i.e. viewing geodata, editing data or analysing data) * Whether the user wants a desktop GIS or a web map server. (Steiniger and Bocher, 2009) Using the information supplied by FreeGIS Organisation (FreeGIS.org) and the research conducted by Steiniger and Bocher (2009) as an informed base it was decided that Quantum GIS, SAGA and GVSIG would be evaluated and compared to each other using the functionality of ESRI's ArcGIS as a guide. Quantum GIS (QGIS - http://qgis.org/) is an open source GIS which was first published in 2002 as a more user-friendly alternative in comparison to other more advanced open source GIS's. Quantum GIS is capable of being executed on Linux, Unix, Mac OSX, and Windows and supports a number of vector, raster, and database formats and functionalities. In addition Quantum GIS contains a Graphical User Interface (GUI) to the majority of GRASS GIS tools - thus increasing its powerfulness (Steiniger and Bocher, 2009, Schilberg, 2009). SAGA is a GIS which was born out of the Department of Geography, University of GÃ¶ttingen (Germany) in 2001 - however its continued development has recently become the responsibility of the University of Hamburg (Germany). SAGA was originally conceptualised as a tool to aid the scientific analysis of physical geography. As such the first few releases of SAGA supported a range of raster analysis including relief analysis, soil mapping and visualization. Over the years the functionality of SAGA has amassed to include a range of data including digital terrain models, vector data and tables and has a growing collection of modules in areas of geo-statistics, image processing and polygon overlays (Steiniger and Bocher, 2009, Schilberg, 2007). gvSIG stands for Generalitat Valenciana, Sistema d'InformaciÃ³ GeogrÃ fica and is a GIS which was originally founded by the Regional Council for Infrastructures and Transportation (CIT) of Valencia (Spain). The impetus for designing an open source GIS was due to the Valencian Government wanting to switch all their systems to Linux based computer infrastructure and as such needed a GIS to replace the functionality of ESRI's ArcGIS (specifically ArcView). The gvSIG project is one of the most highly funded and its primary development is lead by IVER (Spanish based engineering and information technology company) and several universities. Given its objective gvSIG can handle a range of vector and raster formats and offers its program in a range of languages (e.g. Spanish, English, French etc) and as such has a large international following (Steiniger and Bocher, 2009, Schilberg, 2008). Based on the questionnaire and meetings it was decided that the following GIS functionality would be required in the short term and potentially looking towards the future: * Vector/Raster Data read and write * Vector/Raster Data creation and editing * Data creation and editing * Coordinate Transformations/Projections * Analysis tools o Buffering o Interpolation o Measuring distances o Creation of 3D views o Temporal analysis o Layer overlay tools Results of the comparison between the two sets of software can be seen in the table below:
As can be seen from the table above Quantum GIS (when coupled with GRASS GIS), SAGA and gvSIG are each fairly powerful alternatives to ArcGIS, ably performing many of the same operations as ArcGIS and also in some instances have more capabilities, especially when ArcGIS does not have any licensed extensions. This is most notable in the case of Quantum GIS and gvSIG and their ability to perform many interpolation algorithms and also when reading and writing to different file formats. Based on the requirements by the BLS Advisory Committee, each of the three different GIS provided relatively the same functionality and all were able to export to a file format readable by ArcGIS. In terms of having a vibrant user community to provide updates and provide support it was found that each GIS contained available documentation including start-up and installation instructions, user guides and developer manuals. Each GIS also contained a forum which could be easily navigated to from their respective websites. That said, much of documentation regarding gvSIG had not been translated into English - including courses and tutorials and many of the instructions for additional extensions. In that regards gvSIG, despite its large international community and funding, which will ensure its longevity, had to be excluded as an option, because a lack of English written documentation could be huge impediment in the future for a mainly English speaking group from Australia. The level of usability was then considered to determine whether Quantum GIS or SAGA was the preferred GIS. Determining the level of usability is not a simple task and requires much deliberation between all stakeholders. Essentially one must determine how well users can use the functionality of the system (Nielsen, 1993). There are many usability heuristics which can aid in the design and evaluation of user interfaces such as - user control, users ability to recognise objects and actions and the inclusion of help menus (Nielsen, 2005). Based on these before mentioned criteria Quantum GIS has a much more usable user interface than SAGA. This is because it contains a large toolbar with easily identifiable icons compared to SAGA which heavily relies on a written menu tree - which may take some experience in order to navigate efficiently. Furthermore Quantum GIS adds each layer to the one viewing window - overlaying each subsequent added layer. SAGA on the other hand requires the user to select the viewing window for each layer to be displayed in - which is much more demanding and difficult for a lay user. Lastly help files do not exist for SAGA whilst for Quantum GIS when selecting 'help content' from the help menu you are immediately directed to a page with links to online user guides, manuals and forums. Based on this analysis it would seem that Quantum GIS has the most utility and is the most usable open source GIS available and thus the most appropriate for use by the BLS Advisory Committee.
ArcView 9.3 (no extensions)
QGIS (1.0.2 with GRASS plugin)
|SHP (Shapefiles) GML (Geographic Markup Language) DXF (AutoCAD - Drawing interchange format) ||SHP GML DXF ||SHP ||SHP GML DXF |
|SHP GML DXF KML (Keyhole markup Language) SVG (Scalable vector graphics) ||SHP GML DXF MIF (MapInfo interchange format) TAB (MapInfo native format) PostGIS KML (Keyhole markup Language) ||SHP ||SHP GML DXF MIF (MapInfo interchange format) TAB (MapInfo native format) PostGIS KML (Keyhole markup Language) |
|JPEG TIFF ECW ArcInfo ASCI GRID ASC (ASCI raster) ||JPEG TIFF ECW ArcInfo ASCI GRID ASC (ASCI raster) ||TIFF ||JPEG TIFF ECW |
|JPEG TIFF ECW ArcInfo ASCI GRID ASC (ASCI raster) PNG GIF ||JPEG TIFF ECW ArcInfo ASCI GRID ASC (ASCI raster) PNG GIF ||TIFF ||JPEG TIFF ECW ArcInfo ASCI GRID ASC (ASCI raster) |
Data creation and editing
|Yes ||Yes ||Yes ||Yes |
|Yes ||Yes ||Yes ||Yes |
|Yes ||Yes ||Yes ||Yes |
Layer overlay tools
|Yes - Intersect - Join - Union - Clip ||Yes - Intersect - Join - Union - Clip - Difference ||Yes - Intersect - Join - Union - Difference ||Yes - Intersect - Join - Union - Clip - Difference |
|Yes ||Yes ||Yes ||Yes |
|No ||Yes - IDW - Bilinear spline - Regularlised spline with tension ||Yes - IDW - Triangulation - Nearest nieghbour - Ordinary kriging - Universal kriging - Bilinear spline - Thin plate spline ||- IDW - Ordinary kriging - Universal kriging |
|Yes (tracking analyst) ||No ||No ||No |
Creation of 3D views
|No ||Yes ||Yes ||Yes |
3.7 Emergent In general the emergent element of the project is something which will come into notice over time, once more data is collated and the Advisory Committee becomes more confident in using the GIS. Despite this a partial account of the execution of the events undertaken to allow the emergent qualities of the project to prosper can be related as well as the issues faced. In Jake Barton's 'City of Memory' project he notes that by improving access to the map through an online website, public kiosks and providing access at museums the likelihood that the public will participate increases and thus the emergent outcome of the project will be amplified. Theoretically access and editing of BLS spatial data could have been achieved through the use of an online GIS or web 2.0 technologies which would have allowed not only members of the community to contribute to spatial datasets from their own homes but also the wider community. When this idea was originally proposed it was discovered that some of the data which was going to be collected by the group was considered 'sensitive' (e.g. significant native plants) and thus the BLS Advisory Committee would only want to share that information with specific parties such as the Whitehorse City Council. An online GIS was considered a privacy issue because in the past the Advisory Committee had experienced problems of people removing native vegetation from the BLS for commercial or private use and were thus unwilling to disclose exact locations of significant vegetation. This situation exemplifies the importance of the curator's role in ensuring the best interests of the community are met while also paying particular attention to privacy issues. These concerns around privacy did somewhat affect the emergent quality of the project because not all members of the public or community could actively contribute to the formation of the GIS and the spatial data. However, there are other processes in place to circumvent this concern; for instance, any person who spends time in the Sanctuary is able to join the Advisory Committee or contribute to their activities. Furthermore users of the Sanctuary are able to comment on different species of plants which they have identified by filling in a form at the Visitors centre - which members of the Advisory Committee regularly follow-up on. In relation to jumping scales it is difficult to say whether a fully functional BLS GIS will help generate greater interest and funding to the activities of the Advisory Committee. It was witnessed however, that when the results of the constructed GIS were presented to members of the community and some employees of the local council, much lively communication and a constructive debate occurred regarding the location of various vegetation as well as the potential future use of the GIS. This presentation demonstrated - in a small way - that a GIS built with both a top-down and bottom-up approach is able to assist in articulating an equivalent dialogue which can be construed by many individuals which may ultimately benefit the working relationship between the community and council, hopefully translating into effective parks management thus benefiting the wider community.
3.8 Conclusion This chapter has offered the method and results of the BLS GIS project: a case study in the implementation of a community-based GIS using notions of curation to coalescing top-down and bottom-up approaches allowing natural outcomes to emerge and subsequently enable the Advisory Committee to contribute to decision-making. The next chapter will discuss the results of the case study in relation to the literature review conducted in the previous chapter.
4. Discussion "We are interested only in participation at the highest level; and in setting autonomous people loose in the world" (Knabb, 1995: 140) Case studies can be seen as a very sophisticated form of storytelling; in this sense they often reveal insights and themes concerning specific situations. The trick then is to be able to extrapolate any findings so that they can be sufficiently related to a wide societal context. Hence this chapter will revisit the goals and objectives outlined within Chapter One, summarising the findings of the research, including limitations and the potential for future research directions.
4.1 Response to research question The primary goal of this research was to respond to the research question: Can Geographic Information Systems be successfully implemented within community based projects? The simple response to this question is 'yes' - given that the BLS Advisory Committee now has a fully operational GIS and that the project has been able to capture the narratives and knowledge of a community to produce a layer of information which does not exist in any GIS managed by Government or any other institution. Evaluating the degree of successfulness of the implementation however is much more complicated, since the project outcomes are varied and are still emerging. For instance determining the movement of species around the sanctuary and investigating its cause and effect may take many years of data collection and GIS operational maturity. By exploring the outcomes of the key objectives outlined within Chapter One it is anticipated that some themes will emerge which will help provide a holistic assessment of the research. The specific key objects of this research project outlined in Chapter One were: 1. Do contemporary PPGIS pragmatic approaches address the original ontological debates of GIS and Society? 2. Can psychogeographic principles help better guide the requirements for a community based GIS? 3. What spatial datasets are available and usable for community groups within Melbourne, Victoria? 4. Do available datasets satisfy the requirements of community groups? 5. Can community knowledge be effectively integrated with traditional spatial data sources? The original aspirations of GIS and Society researchers posited PPGIS as a pragmatic approach which would help contextualise GIS within social and political milieus thus potentially strengthening the engagement of the public with GIS technology and allowing the public to partake or question the agendas of Government or commercial institutions (Sieber, 2006). While this is still true, in many respects what has been found is that the hypothesised blueprint for PPGIS has been reshaped as many diverse disciplines such as planning and conservation have participated in producing multiple approaches with varying technologies and numerous goals (Sieber, 2006). The biggest discrepancy between contemporary PPGIS projects and their design and approaches tends to be related to the degree of involvement by the 'community'. The primary rationale behind entitling the next generation of GIS as - Public Participation GIS - was to deliberately ground future applications in bottom-up approaches so that a far more transparent and empowering GIS was produced. This seemingly definite method has become a lot more ambiguous over time and has essentially become problematic through two distinct practices. The first fails to acknowledge the full importance of a bottom-up approach and instead dictates and informs communities rather than directly involves a community within GIS development and data collection. The second practice is similarly problematic in that it declares to be exclusively employing a bottom-up approach however fails to acknowledge the significant impact and necessity for top-down practices within community GIS applications. What this research has exposed is that bottom-up and top-down approaches cannot be divorced of each other but rather are co-dependent and both crucial to the successful implementation of community-based GIS. Evidence supporting this supposition has been revealed through both the literature review and case study - discerning that it was somewhat implausible, given the technical complexity of GIS and its data intensiveness, to expect communities with limited time and resources to focus all their attention on developing new spatial datasets. Especially when many standardised datasets could be leveraged from other sources albeit often being Government and Universities. Sourcing spatial datasets, however, definitely should not preclude communities from collecting their own spatial data but rather it should be essential to collect some seminal spatial data so as to construct a much more meaningful and accurate community GIS. Given this dichotomy in approaches, this research proposed that psychogeographic principles could help facilitate in the development of a theoretical framework to help guide the implementation of a community-based GIS utilising both top-down and bottom-up approaches. Expressly it was hypothesised that psychogeography could assist in negotiating the partiality of the data and knowledge of the participants by highlighting the importance of each individuals role and knowledge of the geographic environment and recognising that this knowledge is ineffectual unless it can transcend the individual - with the help of a facilitator - to become of greater use to the community (Wood, 2005). The BLS provided a specific case study to explore many of the hypotheses proposed and developed during the literature review. The project could have easily been derailed at any given time, however the enthusiasm and motivation of the members of the Advisory committee ensured that from the outset both time and resources were shared to help guide the project and ensure its completion. The initial requirements analysis proved to be an essential and effective initial phase however also contained a number of failings. Originally a questionnaire was sent via email to individuals to complete in their own time and send back when complete. This was somewhat unsuccessful because the questionnaire proved too difficult to comprehend, especially by people who were less familiar with GIS terminology and who were unfamiliar with the context of some specific questions. Consequently a face-to-face gathering was arranged so that the members of the Advisory Committee could clarify any misunderstandings and openly discuss any misgivings. This approach also enabled a valuable qualitative discussion ensuring that each stakeholder could successfully advise and contribute to the goals of the project. The success of this meeting stimulated the organisation of regular contact and face-to-face meetings which further helped build trust between each stakeholder, thus enabling each individual to feel comfortable enough to actively contribute to the shaping of the GIS. In addition to these meetings a formal presentation was given to members of the community and some employees of the local council outlining the project objectives, methods and benefits to the community. The presentation proved to be a positive initiative, in that it not only improved the group's knowledge of GIS, but also inspired a constructive debate in which each person expressed their views and concerns of the project, allowing the group to build a consensus as to how the project should proceed and what outcomes ought to be achieved. Hence effective verbal communication was fundamental to developing acceptance of the project and generating supportive relationships within the community. During these discussions many of the data requirements were established and the decision was made to leverage as much data as possible from external sources such as Australia's SDI. While it was discovered that the SDI framework provides a rich source of data which can be obtained by navigating to the website and searching for the data, the costs required, no matter how minimal, to procure the datasets provided a major barrier for data acquisition. Investigations also discovered that the SDI lacked localised datasets - for instance there was no accurate data presenting the many BLS walking paths. Additionally the over generalisation of features such as the waterways meant that these features had to be re-captured and re-represented. While this ensured a more accurate GIS was produced, it also meant that much time and effort was spent co-opting data sources instead of dedicating the majority of resources to capturing specific community knowledge such as vegetation locations. Finally, it was noticed that there are currently no feedback loops into the SDI which means that hypothetically the BLS Advisory Committee could not upload their captured or improved datasets so that other groups requiring similar local datasets would not have to exert the same effort recapturing local data or modifying Government datasets. From the commencement of the project it became extremely clear that the members of the Advisory committee had high expectations of what the geospatial technology would be able to deliver in terms of high positional accuracy. In that sense there was a quite a bit of education required to clarify the various grades of GPS technology available for use and the kinds of precision the technology could achieve. Furthermore the type of terrain and existence of dense tree cover and other error producing features within the BLS also exacerbated the possibility of high positional accuracy using GPS - which also needed to be related to the Advisory Committee. Despite this, the Advisory Committee's motivation to learn and actively contribute to the project was exemplary and can be traced to their interest in applying GIS and GPS technology to their activities for a long period of time. The method employed of roaming the Sanctuary in small groups noting the locations of vegetation and specific attributes was seen as very much analogous to the Situationist dÃ©rive in which they drifted through the space conscious of an objective and destinations yet were drawn by the attractions of the terrain and not limited by what it had to offer. For instance on many occasions the Advisory Committee came across species which they did not expect to and in the end were important to the overall vegetation survey. In this sense it was found that the use of GPS was not capturing and producing data which had been mediated by the agendas of the state but rather had been used to express the stories and interpret the knowledge of the advisory committee. In this way GPS was able to "combine the objective and omniscient discourse of cartography with the subjective, grounded experience of the user" (Kwan, 2007: 26). What become obvious during the requirements analysis (and throughout the project) was that the requirements by the Advisory Committee was for a fairly simple, static GIS which could act as a geographical repository of information, rather than a multimedia or internet based GIS which would possibly have enabled far greater community access and spatial knowledge inclusion. The goals and fears around privacy were obviously a major factor in this decision - but just as interesting was the requirement for an easy to use and low-maintenance GIS rather than a highly sophisticated analytical and visualization tool. Hence the curatorial approach to community-based GIS must consider the specific needs of the public rather than assuming all users want or expect rich multimedia maps and representations. In contrary these sorts of approaches may actually hinder a novice GIS community from taking full hold of a GIS if more creative approaches affect the cost, usability and appropriateness for a diverse community group. It was also discovered that the curator plays are critical role in the development and limitations of spatial metadata. For instance, it was deemed imperative that the collected spatial data have associated metadata in terms of producing informal narrative descriptions of the data that could be easily understood by current and future users. By checking the interpretation of data collected with the perceptions of the Advisory Committee the quality and rigour of the GIS was greatly improved. By having the Advisory Committee diagram their mental maps it further facilitated inclusiveness as well as augmenting everyone's own learning, self reflection and action as well as transforming the GIS into a tool more recognisable by the local community (Kindon et al., 2007).
4.2 Limitations of the research One of the key limitations of the research was the application of one case study to investigate the developed research paradigm. The use of a case study did however ensure that many presumptions made in the literature review were debunked because of the pragmatic nature of the activities of the community group. In this regard, it would be interesting to see whether the same conclusions would be reached had additionally case studies been performed with other community groups with different motivations. By having only a select number of members of the Advisory Committee contribute to the requirements analysis phase there is no doubt that the requirements garnered only reflected the motivations of select few with a similar agenda. On reflection it would have been interesting to include members of the council and some individuals from the wide community in the initial analysis phase to establish whether any other requirements were determined pertinent to the development of the GIS. A significant hindrance to data collection was the necessity to record the location of species during specific time periods. In general much of data collection was limited to late spring - when the vegetation was flowering - ensuring that vegetation could be more easily discovered and identified. This constraint meant that much of the data collection had to be completed toward the beginning of the project before the majority of the literature review was completed and while the requirements phase was still progressing. If the data collection had been allowed to occur slightly later along the project timeline methods such as data dictionaries could have been established allowing for a much more efficient data collection to be performed and less data editing to be done post collection. Given that the BLS contains a heavy concentration of remnant vegetation, including large species which provide a thick forest canopy, the use of GPS to record the location of some features were highly likely to have increased positional errors. Given that this particular project did not look at GPS error correction in depth, there is a chance that some positions are inaccurate. Saying that positional and accuracy information has been recorded into the metadata to signify the data capture process. Also for the purposes of this project, high positional accuracy was not really required; rather positions had to be accurate enough to offer the user, when in the field, the ability to sight the species when in its general proximity. There are many Open Source GIS software available for use, in fact as stated in the case study as of as of 30 January 2009 the FreeGIS Organisation (FreeGIS.org) listed 348 GIS related free or open source projects available for use. Due to the size, nature and time constraints of the project, all of these GIS's could not be realistically evaluated individually to determine the most appropriate solution. Instead, the research of Steiniger and Bocher (2009) was leveraged to help guide this research on open source software. If more time and resources had been available it would have been much more beneficial to assess numerous open source GIS solutions to determine the one which best suited the requirements of the Advisory Committee.
4.3 Recommendations for future research In response to the ever growing development and use of geospatial technologies there is much research to be done to further improve the approaches and methods for democratising mapping and ensuring that communities are able to participate and contribute to spatial decision making. More research in the field of PPGIS and community mapping needs to continue to determine whether the implementation of traditional desktop GIS is the most appropriate technology for communities given the intensiveness of the initial set-up. For instance, there is much scope to investigate whether theories and applications of GIScience are actually relevant for use by community groups. Specifically it should be examined whether much of the GIS analytical capabilities are required by novice users to perform their activities as well as whether these capabilities can be easily used by a lay user. Rather there should be a greater focus on researching whether a GIS can be developed which more easily incorporates lay questions such as 'what features are near my home?' without having to translate complex and often convoluted GIS terminology. In other words, can a GIS or other computing technology be developed which are able to translate between computer 'imprecise' instructions and human 'precise' instructions? As stated within the limitations section above, this research did not perform a comprehensive review of open-source GIS software. During the literature review, however, it was discovered that there is certainly a lack of research into the viability for utilising open-source GIS software in the context of community-based projects or novice users. Specifically future research should investigate the ease-of-use of open-source GIS and whether the solutions can be easily installed, updated, maintained and help easily sourced. It was also noted that there are some companies that offer training programs in open source software however it seems as if none of these training programs are offered by Australian based companies. An interesting case study could be to determine the commercial viability of a start-up Australian organisation offering maintenance, development and support of open source GIS software for a range of cliental. Within this particular project the Advisory Committee only required and sourced fairly rudimentary datasets such as roads, parcels and hydrology. However it would be interesting to further research the effectiveness of the SDI infrastructure if other uncommon datasets were required - and whether these datasets met the objective of the community. Almost certainly no single approach such as a SDI are able to completely fulfil the needs of both the data developers, custodians and users, however it would be interesting to research whether the addition of feedback loops and a stronger involvement by the public would help minimise some of the limitations of the SDI. Overall much research still needs to occur to determine innovative methods to better support diverse users to find, acquire and modify data to suite there unique needs as well as contribute to spatial datasets. In addition to the SDI feedback loops there also needs to be more theoretical and pragmatic research done into the potential role Australian human sensory systems could potentially play in environment and urban monitoring. In other words, what frameworks can or need to be fostered to enable local communities, without formal credentials, to perform valid and reliable research in accordance with Government expectations or within the canon of science. Examining the limitations, validity and benefits of assembling masses of volunteered information from diverse citizens could also prove beneficial. Given that the BLS is approximately 75 acres (30.4ha) in size and contains much biodiversity and wildlife, it can be considered an extremely important and significant environment asset of Melbourne. As such there is much scope for ongoing research to be completed using the BLS as a case study. Using this initial research and GIS development, researches could continue to upgrade the functionality of the GIS or perform new studies regarding the wildlife or vegetation adding the collected data to the already established GIS. It is hoped that from this initial case study a burgeoning relationship between the community, RMIT and Government can continue well into the future.
4.4 Conclusions Despite their many limitations maps are still the preeminent method for organising and visualizing spatial information. For this reason GIS use can be a constructive method for facilitating communication and subsuming the wider community within the spatial decision-making process (Carver, 2003, Gonzalez et al., 2008, Wilson, 2005). Given the extent of PPGIS research and projects being undertaken globally it is safe to say that many projects are addressing the original ontological underpinnings of GIS and Society by developing GIS's that make spatial decision-making data and tools available to non-traditional users. That said, many of the participation techniques used within PPGIS applications do vary according to the expertise required and the time and effort allowed by a community. Given that much research has highlighted the limitations and misuse of PPGIS and GIS in particular, this research project has demonstrated that employing a facilitator who can assist in mapping local perceptions and knowledge can help develop a GIS which contains multiple realities, thus ensuring there is a shared understanding of place. While this study has not provided a foolproof approach to eliminating the ambiguities in GIS technology and the participatory process, it is hoped that the research has illustrated how these ambiguities may emerge and how, by considering both top-down and bottom-up approaches, many of the ambiguities can be reduced or navigated more easily. This research has demonstrated that GIS is not only a system or map which is measured by truth or error but rather can enable communities to re-engage with spaces and experiencing spaces from different perspectives, opening up new ways of seeing places rather than limiting them to one single geographic truth (Pinder, 1996). By providing a framework for a community-based GIS, this thesis has attempted to articulate that rather than fearing or eliminating technical discourse or community narrative from the decision making process, through practices of psychogeography and GIS technology we can make technical discourse accessible to communities and community storytelling and knowledge emendable to reason. Bibliography (2009) What is the ASDI? IN ANZLIC (Ed.) Australian Spatial Data Infrastructure. Canberra, ACT, ANZLIC. AITKEN, S. C. (2002) Public particpation, technological discources and scale of GIS. IN CRAIG, W. J., HARRIS, T. M. & WEINER, D. (Eds.) Community Particpation and Geographic Information Systems. London ; New York, Taylor & Francis. ARNSTEIN, S. R. (1969) A ladder of citizen particpation. American Institute of Planners Journal, 35
, 216 - 224. AUSTER, P. (1992) City of glass. The New York trilogy. London, UK, Faber and Faber. BRODNIG, G. & MAYER-SCHÃ–NBERGER, V. (2000) Bridging the Gap: The Role of Spatial Information Technologies in the Integration of Traditional Environmental Knowledge and Western Science. The Electronic Journal on Information Systems in Developing Countries, 1. BROWN, R. H. (1998) Toward a democratic science : scientific narration and civic communication, New Haven, Yale University Press. BURNS, R. B. (1990) Introduction to social research in education, Melbourne, Australia, Longman Cheshire. CARTWRIGHT, W. E. (2004) Exploring and discovering geographical facts using geographic storytelling. Image, Text and Sound 2004: The Yet Unseen: Rendering Stories. Melbourne, Australia, RMIT Publishing. CARTWRIGHT, W. E., GARTNER, G. & LEHN, A. (2009) Cartography and art, Berlin, Springer-Verlag. CARVER, S. (2003) The future of participatory approaches using geographic information: developing a research agenda for the 21st Century. Urban and Regional Information Systems Associations (URISA) Journal, 15
, 61-71. CHANG, K.-T. (2008) Introduction to geographic information systems, Boston, McGraw-Hill. COUNCIL, W. C. (2009) Blackburn Lake Sanctuary Advisory Committee. IN SOLUTION, A. (Ed.). COVERLEY, M. (2006) Psychogeography, Harpenden, Pocket Essentials. COX, K. R. (1998) Spaces of dependence, spaces of engagement and politics of scale, or: looking for local politics. Political Geography, 17
, 1 - 23. CRAIG, W. J. & ELWOOD, S. (1998) How and why community groups use maps and geographic information. Cartography and Geographic Information Systems, 25
, 10. CRAIG, W. J., HARRIS, T. M. & WEINER, D. (2002) Community participation and geographic information systems, London ; New York, Taylor & Francis. CRAMPTON, J. W. & KRYGIER, J. (2006) An introduction to critical cartography. ACME: An International E-Journal for Critical Geographies 4
, 11 - 33. DEBORD, G. (1955) Introduction to a Critique of Urban Geography. Les LÃ¨vres Nues #6. DEBORD, G. (1956) Theory of the DÃ©rive. Les LÃ¨vres Nues #9. DEBORD, G. (1957) Report on the Construction of Situations and on the International Situationist Tendency's Conditions of Organization and Action. Situationist International Anthology. Bureau of Public Secrets. DEBORD, G. (1964) The Society of the Spectacle (La SociÃ©tÃ© du spectacle). Paris, Editions Buchet-Chastel DEBORD, G. & WOLMAN, G. J. (2006) A user's guide to dÃ©tournement, Berkeley, CA, Bureau of Public Secrets. DEMPSEY, P. (2009a) Animals of Blackburn Lake Sanctuary. IN COMMITTEE, B. L. S. A. (Ed.). DEMPSEY, P. (2009b) Welcome to the Blackburn Lake Sanctuary. IN COMMITTEE, B. L. S. A. (Ed.). DEMPSEY, P. & REES, R. (2009) History. IN COMMITTEE, B. L. S. A. (Ed.). DUNCAN, S. L. & LACH, D. H. (2006) Privileged knowledge and social change: effects on different participants of using geographical information systems technology in natural resource. Environmental management, 38
, 267-285. ELWOOD, S. (2006) Critical Issues in Particpatory GIS: Deconstructions, Reconstructions, and New Research Directions. Transactions in GIS, 10
, 693 - 708. ELWOOD, S. (2007) Grassroots groups as stakeholders in spatial data infrastructures: challenges and opportunities for local data development and sharing. International Journal of Geographical Information Science, 22
, 71 - 90. ERLE, S., GIBSON, R. & WALSH, J. (2005) Mapping hacks : tips & tools for electronic cartography, Sebastopol, CA, O'Reilly Media. FORD, S. (2005) The Situationist International : a user's guide, London, Black Dog. GAILE, G. L. & WILLMOTT, C. J. (2005) Geography in America at the dawn of the 21st century, Oxford ; New York, Oxford University Press. GONZALEZ, A., GILMER, A., FOLEY, R., SWEENEY, J. & FRY, J. (2008) Technology-aided participative methods in environmental assessment: An international perspective. Computers, Environment and Urban Systems, 32
, 303 - 316. GOODCHILD, M. F. (2007) Citizens as sensors: the world of volunteered geography. GeoJournal, 69
, 211 - 221. HAARSTAD, H. & FLOYSAND, A. (2007) Globalization and the power of rescaled narratives: A case of opposition to mining in Tambogrande, Peru. Political Geography, 26
, 289 - 308. HALL, G. B. & LEAHY, M. G. (2008) Open source approaches in spatial data handling, Berlin, Springer. HARMON, J. E. & ANDERSON, S. J. (2003) The design and implementation of geographic information systems, Hoboken, NJ, John Wiley & Sons, Inc. HARRIS, T. M. & WEINER, D. (1998) Empowerment, marginalization and ''community-integrated" GIS. Cartography and Geographic Information Systems, 25
, 10. INDUSTRIES, D. O. P. (2008) Ecological Vegetation Classes (EVCs). IN VICTORIA, S. G. O. (Ed.) Victorian Resources Online. JONES, M., JONES, R. & WOODS, M. (2004) An introduction to political geography : space, place and politics, London ; New York, Routledge. KINDON, S. L., PAIN, R. & KESBY, M. (2007) Participatory action research approaches and methods : connecting people, participation and place, London ; New York, Routledge. KLAEBE, H., FOTH, M., BURGESS, J. & BILANDZIC, M. (2007) Digital Storytelling and History Lines: Community Engagement in a Master-Planned Development. International Conference on Virtual Systems and Multimedia. Brisbane, Australia. KNABB, K. (1995) Situationist International anthology, Berkeley, CA (P.O. Box 1044, Berkeley, CA94701), Bureau of Public Secrets. KNABB, K. (2006) Situationist International anthology, Berkeley, CA, Bureau of Public Secrets. KNIGGE, L. & COPE, M. (2006) Grounded visualization: integrating the analysis of qualitative and quantitative data through grounded theory and visualization. Environment and planing A, 38
, 2021 - 2037. KRYSA, J. (2006) Curating immateriality : the work of the curator in the age of network systems, Brooklyn, NY, Autonomedia. KWAN, M. (2002) Is GIS for Women? Reflections on the critical discourse in the 1990s. Gender, Place and Culture, 9
, 271 - 279. KWAN, M. (2007) Affecting geospatial technologies: Towards a feminist politics of emotion. The Professional Geographer, 59
, 22 - 34. KWAN, M. & KNIGGE, L. (2006) Doing qualitative research using GIS: an oxymoronic endeavor? Environment and planing A, 38
, 1999 - 2002. LEITNER, H., MCMASTER, R. B., ELWOOD, S., MCMASTER, S. & SHEPPARD, E. (2002) Models for making GIS available to community organizations: dimensions of difference and appropriateness. IN CRAIG, W. J., HARRIS, T. M. & WEINER, D. (Eds.) Community Particpation and Geographic Information Systems. London ; New York, Taylor & Francis. LYNCH, K. (2002) The image of the city, Cambridge [Mass.], M.I.T. Press. MAANTAY, J. & ZIEGLER, J. (2006) GIS for the urban environment, Redlands, Calif., ESRI Press : Independent Publishers Group (IPG) [distributor]. MACEACHREN, A. M. (1995) How maps work : representation, visualization, and design, New York, Guilford Press. MAILHOT, J. (1993) Traditional ecological knowledge: the diversity of knowledge systems and their study. IN OFFICE, G. W. P. R. S. (Ed.) Background Paper No.4. Montreal. MCDONOUGH, T. (1994) Situationist Space October 67,58 - 77. MCDONOUGH, T. (2002) Guy Debord and the situationist international : texts and documents, Cambridge, Mass., MIT Press. NETELER, M. & MITASOVA, H. (2004) Open source GIS : a GRASS GIS approach, Boston, Mass., Kluwer Academic Publishers. NIELSEN, J. (1993) Usability engineering, Boston, Academic Press. NIELSEN, J. (2005) Ten Usability Heuristics. OFFEN, K. H. (2003) Narrating place and identity, or mapping Miskitu land claims in northeastern Nicaragua. Human Organization, 62,382 - 392. PAUL, C. (2006) Flexible contexts, democratic filtering and computer-aided curating: models for online curational practice Curating immateriality: the work of the curator in the age of network systems. Brooklyn, NY, Autonomedia. PICKLES, J. (1995) Ground truth : the social implications of geographic information systems, New York, Guilford Press. PINDER, T. (1996) Subverting cartographiy: the situationists and maps of the city. Environment and planing A, 28,405 - 427. PLANT, S. (1992) The most radical gesture : the Situationist International in a postmodern age, London ; New York, NY, Routledge. RUNDSTROM, R. A. (1991) Mapping, postmodernism, indigenous people and the changing direction of North American cartography Cartographica: The International Journal for Geographic Information and Geovisualization, 28,1 - 12. RUSSO, A. & WATKINS, J. (2004) Creative New Media Design: Achieving Representative Curatorial Practice Using a Cultural Interactive Experience Design Method. Interaction: Systems, Practice and Theory. Sydney, Australia. RUSSO, A., WATKINS, J., KELLY, L. & CHAN, S. (2008) Participatory Communication with Social Media. Curator, 51,21 - 32. SADLER, S. (1998) The situationist city, Cambridge Mass., MIT Press. SARANTAKOS, S. (1998) Social research, South Melbourne, Macmillan Education Australia. SCHILBERG, B. (2007) FreeGIS-summary on SAGA. IN E.V., F. (Ed.) FreeGIS database. Intevation GmbH. SCHILBERG, B. (2008) FreeGIS-summary on gvSIG. IN E.V., F. (Ed.) FreeGIS database. Intevation GmbH. SCHILBERG, B. (2009) FreeGIS-summary on Quantum GIS IN E.V., F. (Ed.) FreeGIS database. Intevation GmbH. SCHLOSSBERG, M. & SHUFORD, E. (2005) Delineating "Public" and "Particpation" in PPGIS. Urban and Regional Information Systems Associations (URISA) Journal, 16,15 - 26. SCHOLZ, T. (2006) The participatory challenge. Curating immateriality: the work of the curator in the age of network systems. Brooklyn, NY, Autonomedia. SCHROEDER, P. C. (1996) Criteria for the design of GIS/2. Specialists' meeting for the NCGIA Initiative 19: GIS and society. SEBALD, W. G. (2002) Vertigo, Frankfurt, Vintage. SIEBER, R. E. (2000a) Conforming (to) the opposition: the social construction of geographical information systems in social movements. Geographical Information Science, 14
, 775-793. SIEBER, R. E. (2000b) GIS Implementation in the Grassroots. Urban and Regional Information Systems Associations (URISA) Journal, 12
, 15-29. SIEBER, R. E. (2006) Public Particpation Geographic Information Systems: A Literature Review and Framework. Annals of the Association of American Geographers, 96
, 491 - 507. SIEBER, R. E. (2007) Spatial data access by the grassroots. Cartography and Geographic Information Science, 34
, 47 - 62. STEINIGER, S. & BOCHER, E. (2009) An Overview on Current Free and Open Source Desktop GIS Developments. International Journal of Geographical Information Science, in press. TALEN, E. (1999) Constructing neigbourhoods from the bottom up: the case for resident-generated GIS. Environment and Planning B: Planning and Design, 26
, 533-554. TALEN, E. (2000) Bottom-up GIS: A new tool for individual and group expression in participatory planning. Journal of the American Planning Association, 66
, 279 - 294. TULLOCH, D. L. & SHAPIRO, T. (2003) The intersection of Data Access and Public Participation: Impacting GIS Users' Success? Urban and Regional Information Systems Associations (URISA) Journal, 15
, 55 - 60. UNGAR, S. (2005) Psychogeography. IN CAVES, R. (Ed.) Encyclopedia of the City. London, Routledge. WIEDERMANN, I. & FEMERS, S. (1993) Public Participation in Waste Management Decision Making: Analysis and Managment of Conflicts. Journal of Hazardous Materials, 33
, 355 - 368. WILSON, M. (2005) Implications for a Public Participation Geographic Information Science: Analyzing Trends in Research and Practice. Department of Geography. Seattle, Washington USA, University of Washington. WOOD, D. (2005) Public? Particpation? Geographic? Information? Systems? URISA Conference on PPGIS. Cleveland State University.