GIScience 2016 notes

The bi-annual Geographic Information Science conference is one of the focal point on the field. This year, it was held in Montreal. You can find my talk in a long and separate post. Here are some notes of talks that I took during the meeting.

The conference started with reasons for the location, and a tribute to Roger Tomlinson

Monica Wachwicz, which open the conference with the first keynote, explored her experience in managing a complex set of projects that deal with sensing the environment using geospatial technologies. She summarised her insights:

She aslo described the challenges of recruiting computer scientists, mathematicians, & social scientists to multidisciplinary team.

In the poster session, the work of Daniel Bégin stood out (and he later won an award for it)

Dennis Hlynsky gave the keynote on the second day. As an artist, he is using digital technologies to see the world around him, focusing on the individual and the group. We live in worlds and there are other ‘worlds’ around us – some of too fast, some too slow, some you can’t sense – from gravity to chemistry. He views the creative process as something that is based on guidelines (e.g. Warhol daily procedures), opportunity, being a resident (living in a place and taking it in). Making a playful mess. Critique (is it working, is it not? What my artwork communicate? does it need more or less?). Make sense (we are sense making creatures, telling stories, making sense of things we don’t understand). However, many times he does not know what he is doing and explore things. To witness in a place, what you are witnessing and conveying is important: verbal storytelling, narrative, science and experiments, drawing and painting, photography and film, text, maps, data analysis. We’ve tried to mechanise witness – for example perspective, which force understanding the world from a single point of view. Photography is also a form of mechanisation. There is also the issue of mechanisation in emphasising efficiency which part of industrialisation. The technology changes in photography is important. Until the 1970s, the cost of photographic equipment was limited the opportunity of what is recorded. Since then, the acceleration of sharing photos and evidence change things, with the affordability of cameras in phones. The process of taking photos with phones (which are cameras) make the recording of moments in life much more common, but also the need to switch them off in order to be at the moment. There are many opportunities to do creativity with cameras – for example, providing them to all the students in class, and created a system that allow people to share images, but explicit human intuition to link things, not an automatic tags analysis. There are subsets of the world that communicate with each other, even if they can’t understand it fully.  Interestingly, YouTube is focusing on data driven relationships, while Vimeo is about human led curation. There are different was of organising and understanding the world. The claim that an image is worth a thousand words can be turned on its head – you need to understand the context and meaning of the world, and this is not possible without it.

Helen Couclelis talked about ‘Encyclopedia Gallica of (improbable) event and the why GIScience is not like physics’. The informational standpoint, events, processes, endurance, non-event, do not have a user independent definition. Road networks look different from a perspective of a tourist and biologist, so we need to find a way to create information that support their use. Events are more complex: flood – flood can be an event, process to those in charge of evacuation, occurrences for disaster statistics, noise to everyone else. As metascience – GISCience is a framework for optimising the relation between the interests of information seekers and data in any spatio-temporal domain. She suggest a user-centred GIS with the notion of R-Events to help in search process. The empirical and informational aspects in information systems as distinct epistemic layers.

Genevieve Reid & Renee Sieber compared indigenous ontologies of time. It provide a case for inclusive semantic interoperability, and ensure representation and accessibility for indigenous knowledge. SNAP/SPAN frameworks for ontologies have a very basic notion of time in a Newtonian way and as always progressing and unilinear. In contrast, in TEK, time can be spiral, branch, triangle, cyclical, or double spiral – future that incorporate the future (in Maori culture). Eastern Cree culture see part and present leading to the future. In TEK time is not temporal but social. There are no fix – creation stories include notions of creating a river through a specific story. Time also has an agency. On the basis of these different concepts, she progress to suggest an inclusive model of relationship trough ontological representation. Time is not a simple model but into spatial temporal relations. GIScience can’t ignore the different social constructions of time – excluding indigenous concepts is ontological violence and risk of loss of indigenous knowledge.

Lex Comber et al. talked about “A Moan, a Discursion into the Visualisation of Very Large Spatial Data and Some Rubrics for Identifying Big Questions”, while looking at trends in anti-depressant. Databating meaning manipulation of database. There is an increasing amount of data, and demands (experiences of changing title from GIScience to Data Analytics, create challenges). There is a lack of asking serious questions or knowing what is it for, and ‘letting the data talk for itself’. There is opening of data – for example, GP practice prescription: the practice, the drug, the postcode of the patient. The Postcode can be linked to geography. Demonstrate that it’s possible to producing stupid results by only going data fishing. If we have a plan, on the other hand, we can see urban/rural areas. Need to use: view, refine, and zoom. If you are looking for a needle in a haystack, then making the stack bigger is not making it easier.

Jim Tatcher et al. (delivered by David O’Sullivan) ‘Searching for a common ground (again). Mentioning Golledge et al. 1988 A ground for Common Research. Need to identify common terms and how they are used.  The model of seeing the world as layer cake, is still significant. Harvey Miller mentioned it in 2003 that Euclidean space can be problematic, and it is associated with the quantitative geography. However, in old books that are all sort of representations that look different. From Bunge to Haggett, there are representations that are not Euclidean. The paradigm of GIS caused the adoption of this model. Tobler’s first law actually appeared as ‘throwaway remark’ and travelled through geography in different ways. Need to consider why Euclidean is accepted for granted when in earlier period there were many experimentation.

Has GIScience Lost its Interdisciplinary Mojo?

The GIScience conference is being held every two years since 2000, and it is one of the main conferences in the field of Geographic Information Science (GIScience). It is a special honour to be invited to give a keynote talk, and so I was (naturally) very pleased to get an invitation to deliver such a talk in the conference this year. The title of my talk is ‘Has GIScience Lost its Interdisciplinary Mojo?’ and I’m providing here the synopsis of the talk, with the slides.

My own career is associated with GIScience very strongly. In 1992, as I was studying for my undergraduate studies with a determination to specialise in Geographic Information Systems (GIS) by combining computer science and geography degrees, I was delighted to discover that such studies fall within a new field called GIScience. The paper by Mike Goodchild that announced the birth of the field was a clear signal that this was an area that was not only really interesting, but also one with potential for growth and prospects for an academic career, which was very encouraging. This led to me to a Masters degree which combined environmental policy, computer science, and GIS. During my PhD, I started discovering another emerging area – citizen science, with two main pieces of work – by Alan Irwin and Rick Bonney marking the beginning of the field in 1995 (I came across Irwin’s book while looking into public understanding of science, and learn about Bonney’s work much later). According to OED research, the use of citizen science can be traced to 1989. In short, GIScience and citizen science as a recognised terms for research areas have been around for about the same time – 25 years.

Over this period, I have experienced an inside track view of these two interdisciplinary research fields. I would not claim that I’ve been at the centres of influence of either fields, or that I’ve analysed the history of these areas in details, but I followed them close enough to draw parallels, and also to think – what does it mean to be involved in an interdisciplinary field and what make such a field successful? 

The use of terms in publications is a good indication to the interest in various academic fields. Here are two charts that tell you how GIScience grown until it stalled around 2010, and how citizen science have been quiet for a while but enjoying a very rapid growth now.

First, from Egenhofer et al. 2016 Contributions of GIScience over the Past Twenty Years, showing the total number of publications with the keywords GIS or GIScience, based on a Scopus query for the years 1991 through 2015, executed in July 2015. Notice the peak around 2009-2010.


And here is Google Trends graph for comparing GIScience and Citizen Science, showing that in the past 8 years citizen science has taken off and increased significantly more than GIScience:


I think that it’s fair to say that these two fields as inherently interdisciplinary.

In GIScience, as Traynor a Williams identify already in 1995: “Off-the-shelf geographic information system software is hard to use unless you have sufficient knowledge of geography, cartography, and database management systems; are computer-literate” and to these observations we need to add statistics, algorithms development, and domain knowledge (ecology, hydrology, transport).

Citizen Science also includes merging knowledge from public engagement, education, science outreach, computer science, Human-Computer Interaction, statistics, algorithms and domain knowledge (e.g. ecology, astrophysics, life science, digital humanities, archaeology).

Both fields are more than a methodology – they are contributing to scientific research on different problems in the world, and only a very reductionist view about what they are will see them as ‘a tool’. They are more complex than that – which is why we have specific scholarship about them, periods of training, dedicated courses and books, conferences and all the rest.

A very shallow comparison will note that GIScience was born as an interdisciplinary field of study, and experience consolidation and focus early on with research agendas, core curriculum which was supposed to lead to stability and growth. This did not happen (see Patrick Rickles comments, from an interdisciplinary research perspective, on this). Take any measure that you like: size of conferences, papers. Something didn’t work. Consider the Esri UC, with its 15,000 participants who are working with GIS, yet only a handful of them seem to be happy with the identity of a GIScientists.

In contrast, Citizen Science is already attracting to its conferences audience in the many hundreds – the Citizen Science Association have 4000 (free) members, The European Citizen Science Association 180 (paid) – and that is in the first 2 years since they’ve been established. It doesn’t have an explicit research agenda, and have an emerging journal, but the field also benefits from multiple special issues – there is almost a competition among them.

As a GIScientist this is a complex, and somewhat unhappy picture. What can I offer to explain it? What are the differences between the two fields that led to the changes and what we can learn from them? It is worth exploring these questions if we want the field to flourish

Engaging with Interdisciplinary research

The wider engagement with these fields is also linked to my personal and direct engagement in GIScience research that goes beyond disciplinary boundaries. Over the years, I was also involved in about 20 multidisciplinary, cross-disciplinary, interdisciplinary, and transdisciplinary projects. I also found myself evaluating and funding x-disciplinary projects (where cross, inter, multi or trans  stand for x). The main observations from all these is that many times, projects that started under the interdisciplinary flag (integrating knowledge from multiple areas), ended with mostly multidisciplinary results (each discipline addressing the issue from its own point of view). However, here are nine lessons that I’ve learned, which can also help evaluating the wider fields of GIScience and citizen science.

First, Get them young & hungry – when established professors are joining an interdisciplinary project, usually they have a clear personal research agenda, and the likelihood that they will be open to radically new ideas about their area is low. You can get excellent multidisciplinary projects with experienced researchers, but it is much harder and rarer to have interdisciplinary or transdisciplinary project – there is too much to lose. That mean that early career researchers are the most suitable collaborators who can develop new directions. At the same time, in terms of job potential and publications, it is very risky for PhD students to get into interdisciplinary research as this can reduce their chances of securing an academic job. With appropriate funding (as we done in Bridging the Gaps) and specific support to people at the more secured stage of early career (after securing a lectureship/assistant professor position), we’ve seen interdisciplinary collaboration evolve.

Second, in x-disciplinary projects, you’ll find yourself being undermined, unintentionally which will hurt. Disciplines have different notion of ‘truth’ and how to get to it (in philosophy: epistemology and ontology). What is considered as an appropriate methodology (e.g. fixation with randomised control trials), how many people need to participate, how they are selected and more. When people from another discipline use these concepts to question your practice it can feel as undermining the expertise, and the disciplinary knowledge that you are offering to the project…

logo-ercThird, there are also cases of being undermined, intentionally. Interdisciplinary proposal are evaluated by experts from different fields, and no matter how much they are told to focus their comments on their discipline, they will comment on other aspects. Moreover, proposal evaluators can assess the novelty in their area, not the overall innovation, reducing the likelihood of ‘outstanding’ mark that make it more likely to get funded. For example, in an early version of what was now funded by both EPSRC and ERC, a Research Challenges Board rejected the proposal because it “seemed so high risk to us is that there are many links in the chain… is it clear that even if everything works there would be real value from these sorts of devices? You use the example that the forest people might be able to tell if there were poachers in the area. Yet can that really be shown? Do forest people understanding probabilistic reasoning? If there any evidence that illiterate people can use maps, digital or otherwise?“. It’s important to note that both ERC and the EPSRC programmes were aimed at risky, interdisciplinary projects, but in more standard programmes, it is difficult to get funded.

Fourth, look out for the disciplinary scrounger. They might not be aware that they are disciplinary scrounger, but this is how it happens: Interdisciplinary research open up new tools and methodologies and people who know how to use them for the research team as a whole. While there is a supposed shared goals that will provide benefits to all sides, a savvy researcher will identify that there is an opportunity for using resources to advance their own research in their discipline, and find ways to do that, even if there are no apparent benefits to the side that give the resources. This act is not necessarily malicious – from the researcher perspective, it is exactly a demonstration of interdisciplinary contribution.

Fifth, in an interdisciplinary research it is critical to develop a common narrative, early. As the project progresses, it will shift and change. Because of the disciplinary differences, it is very easy to diverge and work on different issues, with some relationship to the original proposal. Especially in case where the funder evaluate the project against the proposal (e.g. in Horizon 2020), it’s critical to have a common story. The project can be harmonious and show good progression, but without a common narrative that is shared across the team, there can be troubles when it come to evaluation by external people as the outputs do not all fit neatly to their idea of what the project is about. In another project, Adaptable Suburbs, we deliberately shared reading lists between teams to help understanding each other, which bring us to…

Sixth, highstreetconsider the in-built misunderstanding. Terminology is an obvious one. For Anthropology, scale, from small to large is individual, household, community – and for cartography city is small scale, while house is large scale. However, these are easy – it can take time, and long discussions to discover that you’re looking at the same thing but seeing something completely different. As Kate Jones suggested when she worked on the Successful Suburban Town Centres project. In the image above urban designers see the streets, but not the people, while human geographers who look at census data will tend to see the people, but not the urban structure that they inhibit. There are many other examples of subtle, complex and frustrating misunderstanding that happen in such projects.

Seven, there will be challenges with publications – those that are written. Publications are critical academic outputs, and important for the individuals, teams, and the project as a whole. Yet, they are never easy – different disciplines have very different practices. In some, the first position in the author list is the most important, in another, the last. Some value single author monograph (Anthropology), other conference paper with multiple authors (Computer Science). This creates tensions and a need for delicate discussions and agreement. Moreover, and linked to Six – writing joint publications is an opportunity to expose interdisciplinary misunderstanding, but that make the writing process longer.

Eight, it is important to realise that many times interdisciplinary publications will never be written  – because academic careers, promotion criteria, visibility, and recognition depends on disciplinary practices, within projects disciplinary papers and outputs are written first. The interdisciplinary outputs left to a later stage – and then the project end and they never get written. They are actually dependent on voluntary investment of multiple contributors, which make it very difficult to get them done!

Finally, nine, is the importance of coffee and lunch breaks (and going out together). Team members in interdisciplinary projects are usually coming from different departments, and it is challenging to organise a shared space. However, by putting people together – computer scientists sitting next to a geographer, designer, anthropologists – it is possible to achieve the level of trust, relationship and the development of new ideas that are needed in such projects. In ExCiteS, we have a designated ‘social officer’ for the group.

On the basis of these experiences, I’d argue that Interdisciplinarity is always hard, risky, require compromises, accommodations, listening, and making mistakes. The excitement from the outputs and outcomes does not always justify the price. Frequently, there is no follow-on project – it’s been too exhausting. The analysis that Patrick Rickles done across the literature can provide you with further information on challenges and solutions.

From projects to research fields

Considering the project level challenges, viewing interdisciplinary areas of studies emerging is especially interesting. You can notice how concepts are being argued and agreed on. You can see what is inside and what is outside, and where the boundary is drawn. You can see how methodologies, jargon, acceptable behaviour, and modes of operations get accepted or rejected – and from the inside, you can nudge the field and sometimes see the impact of your actions. Here are some observations about GIScience and citizen science evolution.

First, citizen science seem to be growing organically, without a deliberate attempt to set a research agenda, define core curriculum, or start with nationally focused research centres, in contrast to GIScience, who had all of these. There is an emergent research agenda: data quality, motivations & incentives, interaction design, management of volunteers, and more. These are created according to views of different people who join the research area, opening opportunities for new collaborations. It is noted that GIScience, in practice, allowed for many other areas to emerge – for example crowdsourcing, which was not in the last version of the research priorities that are listed on UCGIS website, and also seemed to stop doing these exercises.

Second, there is an interesting difference in inclusiveness. Although there are different variants of citizen science, across events, conferences and projects, there is an attempt to be inclusive to the different variants (e.g. volunteer computing or ecological observations) though tensions remain and need maintenance. In GIScience, there have been inclusive activities, of workshops that brought together people from Human-Computer Interaction in the late 1980s, or the excellent series of meetings about GIS and Environmental Modelling. There is clear separation, for example in spatial analysis, where different methods are now appearing in ecology, but they are not shared back with the general GIScience. It is worth considering how to make such events and consider active inclusiveness, where researchers from different areas will find their place and reasons to participate.

It might be that citizen science is also more inclusive because of the interaction with people outside academia (participants) and the need to focus on things that matter to them, whereas GIScience has largely been for/by scientists. However, citizen science gets backlash for “not doing REAL science”, but it’s still grown. Maybe, in the process of GIScience trying to validate itself, it’s cut itself off from other research areas (even though GIS use continues to grow)?

Third, there is a sharp difference in the relationship with practitioners – GIScience decided to focus on fundamental questions and laws, while citizen science is a deliberate integration between researchers (the science of citizen science) and practitioners who are running volunteering programmes. The interaction between practice and science is bringing research questions to light and provide a motivation for addressing them with interdisciplinary teams. It might be that separation between science and systems in GIScience need to be blurred a bit to open up new opportunities.
bookcoverFinally, GIScience benefited from having a disciplinary name, and attention by a growing group of researchers who are committed to the field – job titles, positions, journals and conference do matter in terms of visibility and recognition. Citizen science, on the other hand, is only now starting to have a proper home and networks. There are ongoing discussions about what it is, and not everyone in the field is using the term ‘citizen science’ or happy with it. The actual conference that led to the creation of the Citizen Science Association was titled ‘Public Participation in Scientific Research'(!). The coherence and focus on understanding how important key phrases are, more than dislike of their potential meaning is valuable for the coherence of a field and stating that you have knowledge that can be shared with others.

New areas for Interdisciplinary research

To complete this discussion, I point to the opportunities that citizen science open for interdisciplinary collaborations with GIScience – It provides examples for longevity of VGI data sources, that can be used to address different research questions. There are new questions about scales of operations and use of data from the hyper local to the global. Citizen science offer challenging datasets (complexity, ontology, heterogeneity), and also a way to address critical issues (climate change, biodiversity loss). There are also usability challenges and societal aspects.

In final account, GIScience got plenty of interdisciplinary activity in it. There are actually plenty of examples for it. In terms of ‘mojo’ as being attractive for researchers from other area to join in, there are plenty of opportunities – especially if the practice of using GIS within different research and practice problems is included in the framework of GIScience.

This post benefited from discussions and comments from Patrick Rickles, who is our local expert in GIS use in an interdisciplinary settings. You should check his work.

Esri User Conference – Science Symposium


Esri Science Symposium

As part of the Esri User Conference, Dawn Wright, Esri Chief Scientist, organised a Science Symposium that gave an opportunity for those with interest in scientific use of Esri GIS to come together, discuss and meet.

Dawn Wright opened and mentioned that the science symposium is aimed to bring people people from different areas: hydrology, ecology or social sciences – together. The Esri science programme is evolving – and there is official science communication approach. There are different ways to support science including a sabbatical programme. Esri will launch a specific challenge for applications of data sets for students with focus on land, ocean and population. Esri will provide access to all the data that is available and the students are expected to carry out compelling analysis and communicate it. It is an activity in parallel to the global year of understanding. There are also sessions in the AGU meeting that are support by Esri staff.DSCN1690

Margaret Leinen (president, American Geophysical Union) who is working on marine and oceanography gave the main talk ‘what will be necessary to understand and protect the planet…and us?‘. Her talk was aimed at the audience in the conference – people who’s life focus is on data. What is necessary to understand the planet is data and information – it’s the first step of understanding. There are many issues of protecting and understanding the planet – we need to understand planetary impacts on us. The first example is the way we changed our understanding of climate change on the ocean. When we look at the change in sea surface temperature in the 1990 we can see changes up to 2 degrees F. The data was mostly collected in traditional means – measurements along the paths of ships. Through studies from ship records over the years, we have created a view of ocean heating – with different results between groups and researchers with lots of hand crafted compilation of records. In the last decade things have changed: ARGO floats are going up and down through ocean, and make all the data is available – there are 3839 operational floats, reporting every week. This is a completely new way or seeing the data, with huge scale international collaboration. Now we can see the annual cycle and determined the slope in the change in heat content. We have a 10 years time series for the depth of 0-2000m. We have a much more detailed information of the changes. There is an approach to make these devices that will understand the full planetary budget on heat through the whole depth of the ocean. The EarthScope Facilities also provide a demonstration of detailed sensing data – understanding the Earth and it’s movements. Many seismometers that are used for over a decade – the US array provided a massive increase in the resolution of seismic measurements. In 2011, the network identified the Japanese Honshu earthquake. The measurement provided a new class of earthquake modelling that can be used in engineering and science. GPS also provides new abilities to understand deformation o earth. Permanent GPS receivers – many of them – can provide the resolution and accuracy to notice subtle movement, by using very sophisticated statistical filtering. HPWREN – High Performance Wireless Research and Education Network – provide a way to transfer information from sensors who are very remote, then then linked through line of sight communication, and the network provide a reliable and resilient public safety network. The network support many sensing options. There are fire cameras that are linked to it, that alert to provide real time information to the fire department. WiFire is a programme that aim to deliberately work on this issues. GIS data is used to assess surface fuel. In summary: Earth science is going through huge transformation through collaboration of large groups of researchers who are using dense sensing networks. We can now monitor different processes – from short to long term. We gain new insights, and it is rapidly transform into local, regional, national and global responses.

After her talk, a set of responses was organised from a panel, including: Mike Goodchild, John Wilson, Marco Paniho , Ming Tsou, and Cyrus Shahabi.

John: discussion about GIScience – the examples that we’ve seen point to future challenges. We can train people in the spatial sciences, and insist that they’ll learn another area, or change the earth sciences, so people learn about spatial issues, or somewhere in between, with people becoming aware of each other language. Spatial scientists have little capacity to learn a new areas – and same is true for earth scientists. The only viable path is to work together – it’s about working in interdisciplinary teams and enabling people to work with them. Data acquisition is moving fast and it is a challenge to train graduates in this area. Only recently we start thinking about solutions. Academics are experts in dealing with problems in the world, and instead we need to suggest solutions and test them.

Marco: the principle and ideas are problems that are familiar in GIScience although the specific domain of the problem was not familiar. Issues of resolution and scale are familiar in GIScience. We have a long way to go in terms of details of describing a phenomena. We need to see how systematic are we now in acquiring data? We need details of the maps of the heating of the ocean, and understanding what is going on. What is the role of remote sensing in helping us in monitoring global phenomena? We need to think about down-scaling – get from aggregate data to more detailed understanding something locally. What is the role of citizens in providing highly local information on phenomena?

Ming: we need to remembers about ‘how to lie with maps?’ – we need to be very careful about visualisations and cartographic visualisation. Each map is using projections, cartographic representation, and we need to think if it is the appropriate way to ask if that is the appropriate way to present the information? How can we deliver meaningful animation. Cartography is changing fast, but today we need to look at 2000-5000 scale, but we are using now levels and not scale. The networks and models of wildfire are raising questions about which model is appropriate, how many variables we need and which sources of information, as well as the speed of the modelling. Need to think which model is appropriately used.

Cyrius: there are more and more sensors in different context, and with machine learning we have an increased ability to monitor cities. In case of existing models – we have cases of using more data analysis in computer science.

Margaret: we have new ability to move from data, model, analysis and keep the cycle going. In the past, there was gulf between modelling or observations, we don’t see a divide any more and see people going between the modelling and the data.

Discussion points: We need to consider what is the messages that we want to communicate in our maps – we need to embrace other disciplines in improving communication. We need to implement solutions – how much uncertainty you are willing to accept. Every single map or set of data is open and other people can look and change it – this is a profound change.

The earth system is an interrelated system – but we tend to look at specific variables, but data is coming in different resolutions, and details that make it difficult to integrate. Spatial statistics is the way to carry out such integration, the question is how do we achieve that.

It’s not enough to have data as open but the issue is how to allow people to use it – issues of metadata, making it able to talk with other data sets. Esri provide a mechanism to share and address the data.

There is uncomfortable relationships between science and policy – the better the models, there is more complex the issue of discussing them with the public. How to translate decimal points to adjectives for policy making. This creates an issue to communicate with the public and policy makers. There is a need to educate scientists to be able to communicate with the wider public.

Another issue of interdisciplinarity – encouraged as graduate, but not when it come to the profession. There are different paths. Once you land a job, it is up to how you behave and perform.

Considering the pathways of integration, the challenge between the modellers and the observationalists. We can think about identifying a path.

Machine learning might need to re-evaluate how we learn and know something. There is also need to think about which statistics we want to use.

Margaret: what is different now – a growing sense of lack of being able to characterised the things that are going on. The understanding about our ignorance: in the past we had simple linear expectations of understanding. We finding that we don’t understand, and the biosphere and what is does to the world. There are so many viruses in the sea air, and we don’t know what it does to the world. The big revolution is the insights into the complexity of the earth system. How not to simplify beyond the point that we will loose important insight!

UCGIS webinar [Geographic information | Citizen] Science: reciprocal opportunities

At the request of Diana Sinton, the Executive Director of the University Consortium for Geographic Information Science (UCGIS), I gave the seminar talk about the linkage between Geographic Information Science and Citizen Science. A detailed description of the talk and the slides are available here.

The webinar announcement is at The webinar was recorded, so for UCGIS members it should be available later on.

OpenStreetMap in GIScience – Experiences, Research, and Applications

OSM in GIScience

A new book has just been published about OpenStreetMap and Geographic Information Science. The book, which was edited by Jamal Jokar Arsanjani, Alexander Zipf, Peter Mooney, Marco Helbich  is “OpenStreetMap in GISciene : Experiences, Research, and applications” contains 16 chapters on different aspects of OpenStreetMap in GIScience including 1) Data Management and Quality, 2) Social Context, 3) Network Modeling and Routing, and 4) Land Management and Urban Form.

I’ve contributed a preface to the book, titled “OpenStreetMap Studies and Volunteered Geographical Information” which, unlike the rest of the book, is accessible freely.

Geographic Information Science and Citizen Science

Thanks to invitations from UNIGIS and Edinburgh Earth Observatory / AGI Scotland, I had an opportunity to reflect on how Geographic Information Science (GIScience) can contribute to citizen science, and what citizen science can contribute to GIScience.

Despite the fact that it’s 8 years since the term Volunteers Geographic Information (VGI) was coined, I didn’t assume that all the audience is aware of how it came about or the range of sources of VGI. I also didn’t assume knowledge of citizen science, which is far less familiar term for a GIScience audience. Therefore, before going into a discussion about the relationship between the two areas, I opened with a short introduction to both, starting with VGI, and then moving to citizen science. After introduction to the two areas, I’m suggesting the relationships between them – there are types of citizen science that are overlapping VGI – biological recording and environmental observations, as well as community (or civic) science, while other types, such as volunteer thinking includes many projects that are non-geographical (think EyeWire or Galaxy Zoo).

However, I don’t just list a catalogue of VGI and citizen science activities. Personally, I found trends a useful way to make sense of what happen. I’ve learned that from the writing of Thomas Friedman, who used it in several of his books to help the reader understand where the changes that he covers came from. Trends are, of course, speculative, as it is very difficult to demonstrate causality or to be certain about the contribution of each trends to the end result. With these caveats in mind, there are several technological and societal trends that I used in the talk to explain how VGI (and the VGI element of citizen science) came from.

Of all these trends, I keep coming back to one technical and one societal that I see as critical. The removal of selective availability of GPS in May 2000 is my top technical change, as the cascading effect from it led to the deluge of good enough location data which is behind VGI and citizen science. On the societal side, it is the Flynn effect as a signifier of the educational shift in the past 50 years that explains how the ability to participate in scientific projects have increased.

In terms of the reciprocal contributions between the fields, I suggest the following:

GIScience can support citizen science by considering data quality assurance methods that are emerging in VGI, there are also plenty of Spatial Analysis methods that take into account heterogeneity and therefore useful for citizen science data. The areas of geovisualisation and human-computer interaction studies in GIS can assist in developing more effective and useful applications for citizen scientists and people who use their data. There is also plenty to do in considering semantics, ontologies, interoperability and standards. Finally, since critical GIScientists have been looking for a long time into the societal aspects of geographical technologies such as privacy, trust, inclusiveness, and empowerment, they have plenty to contribute to citizen science activities in how to do them in more participatory ways.

On the other hand, citizen science can contribute to GIScience, and especially VGI research, in several ways. First, citizen science can demonstrate longevity of VGI data sources with some projects going back hundreds of years. It provides challenging datasets in terms of their complexity, ontology, heterogeneity and size. It can bring questions about Scale and how to deal with large, medium and local activities, while merging them to a coherent dataset. It also provide opportunities for GIScientists to contribute to critical societal issues such as climate change adaptation or biodiversity loss. It provides some of the most interesting usability challenges such as tools for non-literate users, and finally, plenty of opportunities for interdisciplinary collaborations.

The slides from the talk are available below.

OpenStreetMap studies (and why VGI not equal OSM)

As far as I can tell, Nelson et al. (2006) ‘Towards development of a high quality public domain global roads database‘ and Taylor & Caquard (2006) Cybercartography: Maps and Mapping in the Information Era are the first peer-reviewed papers that mention OpenStreetMap. Since then, OpenStreetMap has received plenty of academic attention. More ‘conservative’ search engines such as ScienceDirect or Scopus find 286 and 236 peer reviewed papers (respectively) that mention the project. The ACM digital library finds 461 papers in the areas that are relevant to computing and electronics, while Microsoft Academic Research finds only 112. Google Scholar lists over 9000 (!). Even with the most conservative version from Microsoft, we can see an impact on fields ranging from social science to engineering and physics. So lots to be proud of as a major contribution to knowledge beyond producing maps.

Michael Goodchild, in his 2007 paper that started the research into Volunteered Geographic Information (VGI), mentioned OpenStreetMap (OSM), and since then there is a lot of conflation of OSM and VGI. In some recent papers you can find statements such as ‘OpenstreetMap is considered as one of the most successful and popular VGI projects‘ or ‘the most prominent VGI project OpenStreetMap so, at some level, the boundary between the two is being blurred. I’m part of the problem – for example, with the title of my 2010 paper ‘How good is volunteered geographical information? A comparative study of OpenStreetMap and Ordnance Survey datasetsHowever, the more I think about it, the more uncomfortable I am with this equivalence. I feel that the recent line from Neis & Zielstra (2014) is more accurate: ‘One of the most utilized, analyzed and cited VGI-platforms, with an increasing popularity over the past few years, is OpenStreetMap (OSM)‘. I’ll explain why.

Let’s look at the whole area of OpenStreetMap studies. Over the past decade, several types of research paper have emerged.

First, there is a whole set of research projects that use OSM data because it’s easy to use and free to access (in computer vision or even string theory). These studies are not part of ‘OSM studies’ or VGI, as, for them, this is just data to be used.

Edward Betts. CC-By-SA 2.0 via Wikimedia Commons

Second, there are studies about OSM data: quality, evolution of objects and other aspects from researchers such as Peter Mooney, Pascal Neis, Alex Zipf  and many others.

Third, there are studies that also look at the interactions between the contribution and the data – for example, in trying to infer trustworthiness.

Fourth, there are studies that look at the wider societal aspects of OpenStreetMap, with people like Martin Dodge, Chris Perkins, and Jo Gerlach contributing in interesting discussions.

Finally, there are studies of the social practices in OpenStreetMap as a project, with the work of Yu-Wei Lin, Nama Budhathoki, Manuela Schmidt and others.

[Unfortunately, due to academic practices and publication outlets, many of these papers are locked behind paywalls, but thatis another issue… ]

In short, there is a significant body of knowledge regarding the nature of the project, the implications of what it produces, and ways to understand the information that emerges from it. Clearly, we now know that OSM produces good data and are ware of the patterns of contribution. What is also clear is that many of these patterns are specific to OSM. Because of the importance of OSM to so many application areas (including illustrative maps in string theory!) these insights are very important. Some of these insights are expected to also be present in other VGI projects (hence my suggestions for assertions about VGI) but this needs to be done carefully, only when there is evidence from other projects that this is the case. In short, we should avoid conflating VGI and OSM.