At long last, a team publication from the GEOIDE grant The Participatory Geoweb has been published in ACME journal. This paper, co-authored by myself, Jon Corbett, Chris Gore, Pamela Robinson, Renee Sieber, and Patrick Allen, takes a critical view of the general enthusiasm for Geoweb projects. We challenge the commonly held notions that the Geoweb is ‘easy’, and highlight several implementation challenges derived from a variety of case studies. For those working with the Geoweb, crowdsourcing, and VGI, I would recommend this as a good overview of the challenges of both developing these types of tools and implementing them within a community context. It’s open-access, so please check it out!
I’ve co-authored an exciting new paper with Dr. Renee Sieber from McGill University. It is currently online first with Government Information Quarterly. With this piece we take a look at the dominant models of open data provision by government and start to lay out what the challenges are for delivering open data. We tried to make this both a reflective look at where open data is, and also to push civic open data forwards, examining how open data works as part of open government strategies. I’ve copied the highlights below. A pre-print copy is available.
- We define four main models for how government delivers open data; data over the wall, code exchange, civic issue tracker, and participatory open data.
- We define challenges for the continued delivery of open data, including; conflicting motivations, the shifting role of government, and the fragility of ‘mission accomplished’.
- We propose that open data be framed as more than provision, but rather as way for government to interact with citizens.
This past winter semester I launched a new course at the University of Waterloo called “The Geoweb and Location-Based Services“. This 4th-year course introduced senior undergraduate students to the theoretical concepts and practical techniques of Web 2.0, Volunteered Geographic Information, Open Data, the Geoweb, and location-based services using mobile phones. As part of this course, students worked in groups to complete a major project.
One project that stood out was “Tweet-Mapping American TV Ratings” by the team of Andrea Minano, Sarah Knight, and Michael Goldring. The aim of their project was to analyze the relationship between social media and the popularity of television shows through ratings. To do this, they gathered data from the social media network Twitter. According to Socialguide.com, 32 million individuals in the United States tweeted about television in 2012. Additionally, studies recently conducted by the television ratings company Nielsen, suggest that Twitter is a robust way to derive TV ratings. Here you can see a map of TV shows from Friday March 22nd, with the location of individual tweets shown:
The use of Twitter to rate the popularity of TV shows was tested in this project by gathering tweets from March 18, 2013 to March 24, 2013, and then mapping their spatial distribution. Click here to view an interactive map of these results. These individual tweets were then aggregated to the state level, to give the most popular shows per state. These most-tweeted TV shows per state were then compared to official national TV ratings. For those Walking Dead fans, you will be pleased to note that Twitter is basically taken over on Sunday nights:
This series of maps were made with Leaflet, an open source web-mapping platform. Seven web-maps were created for each day of the week. In each of these, US states were symbolized according to its most-tweeted TV show. Pie charts can be seen by clicking on each state displaying the three most tweeted TV shows per state. Finally, two bar graphs accompany each map: one showing the three most tweeted TV shows at a national scale, and another showing the three highest-rated TV shows at a national scale.
Overall, the results proved the initial hypothesis correct since there was a clear correlation between most-tweeted TV shows and official TV ratings. However, it is important to note that the results continue to offer some limitations. For instance, there are data limitations because only 1% of tweets are geolocated, and the age of people using Twitter ranges primarily from 18 to 29 years. For this reason, TV shows that are popular with older audiences may not be tweeted about but continue to receive high ratings. Future studies may be conducted in this newly-researched subject; yet, it is evident that tweets have a relationship with TV ratings in the United States, and these can be effectively mapped to find any relevant spatial patterns.
We all know that climate change is having a major impact on weather patterns around the globe. One industry that is particularly exposed to these changes is the ski industry. Though large mountain/high elevation ski resorts may remain insulated from the impacts of shorter ski seasons and more erratic weather, those ski resorts at low altitude are particularly vulnerable to a changing climate. As a mid-latitude, lower elevation (comparatively) ski region, the Pyrenees are one area where the impacts of a changing climate are pronounced.
I have had the pleasure of working with Marc Pons, a PhD student at Universitat Politècnica de Catalunya (BarcelonaTech), in Spain. Marc has been working on developing an agent-based model (ABM) to explore the climate change impacts on winter ski tourism in Andorra, a key skiing destination in the Pyrenees. Marc’s work has looked at how artificial snowmaking can serve as an adaptation response to extend a marginal ski season, or to ensure adequate snow coverage for peak ski times, such as holidays. The benefit to using an ABM for this type of work is that one can quickly develop and test alternate scenarios. For example, in his work, Marc tests the impact of artificial snowmaking on several different Andorran ski resorts – each with a unique geography and elevation that impacts how effective snowmaking is. Also, Marc has taken into account several different scenarios of climate warming, allowing him to present best/worst case scenarios. I see this rapid scenario development as one of the strengths of ABM, particularly in how it can be used in climate adaptation research. Future work can focus on the skier response to changing snow conditions, helping to determine which resorts, operating in a competitive marketplace, can expect to draw more skiers. There are clear business implications of this research, especially when considering how closely tied a local economy is to a major attraction such as a ski resort. This research will be published very soon in the journal Climate Research, but you can take a look at a pre-print version here.
I’d like to take a moment to highlight some of the recent work that our team at McGill has been involved with as part of the Geoweb for Community Development in Rural Quebec project. One of our partners, the Corporation de développement de la Rivière Noire (CDRN) has become very involved with developing Geoweb sites. The first, Géoweb Junior, was developed during summer 2011 by Andréane and Pierre, two undergraduate research assistants at McGill University. Géoweb Junior was a test case for a more detailed Geoweb tool on forest management that is currently being rolled out by CDRN with McGill support. Géoweb Junior provided students (approx. age 10) at a summer day camp organized by CDRN with info sheets to go and gather information on the local environment.
Students collected all manner of observations, from environmental problems, to species of wildlife. Many of them even drew pictures. This information was plotted on a paper map attached to the observation sheet. Students then input this information onto a Google Maps-based Geoweb site and observations were categorized.
In this way, students were able to participate in the data gathering aspect of environmental science, demonstrating how their local information could be georeferenced, and shared publicly via the Geoweb. You can take a look at their observations on the CDRN web page devoted to the project. This type of project, though fairly simple, demonstrates the ability of the Geoweb as a tool for a variety of community-based organization tasks. These could include data collection, participation, discussions, information and data sharing, and many others.
Technical details: This Geoweb implementation uses a Google Maps API tool (V2). This is hosted on a separate server and integrated into the WordPress CMS (rivierenoire.org) using an iFrame tag. Points that are added via double clicking on the map are saved in the database and can be edited or deleted by the user with an appropriate password. We plan to make this system available as an open-source template for anyone to use.
A chapter from my dissertation has been recently published in Environment and Planning B: Planning and Design. This paper, titled “Negotiating constraints to the adoption of agent-based modeling in tourism planning” presents material from a series of interviews that I conducted with tourism planners in Nova Scotia. These planners gave their opinions on the potential for Agent-based modeling as a planning support tool within tourism planning practice and identified several areas of adoption constraint.
This research is actually (and unintentionally) quite timely – there seems to be much made about ‘Geodesign‘ these days and the possibility to connect geospatial analysis tools and approaches within a policy context. Here is a recent meeting from The Centre for Research in Social Simulation at the University of Surrey that discusses the interface between ABM and policy. Everyone seems to be trying to figure out how geotech tools can help decision-makers make better decisions – a noble pursuit, for sure! However, as I (drawing from many others – thanks Helen Couclelis) point out in the paper, there is a fundamental disconnect between the modeler and the planner or policy developer. Modelers (and scientists) thrive on the ability to be wrong about things – a luxury that the policy developer can’t afford (to put it mildly).
When I think about Geodesign, I am excited that GIS and geospatial tools can make an impact within decision-making. But I also hope that this general level of enthusiasm for tools and approaches is accompanied by a similar investment in research that looks to identify and negotiate the adoption constraints associated with technology implementation. As I’m finding out with my current Geoweb research, there are a unique set of constraints created by an organization (community, government, corporation) that can serve as a massive impediment to using any technology. A balanced view going forward should be a research priority!
As part of the project “Geoweb and Community Development in Quebec“, two teams of McGill School of the Environment students spent the fall term 2010 working with a community-based watershed monitoring agency CDRN (Corporation de développement de la rivière Noire) to explore the potential for the Geoweb to serve as a conduit for citizen participation in watershed management. These student groups developed two tools, conducted a series of workshops with community members, and produced reports and instructional materials. McGill Public Affairs produced a short film about the group activities that gives an excellent overview of the project and the potential for the Geoweb in a community development context.