ISCN Online Advisory Programme #13: Co-Creating Cities with Citizens - Crowdsourcing Data and Maps

Maps have always enabled us to find new paths, discover new patterns and gain a deeper understanding of the space in which we find ourselves. Their digital forms also allow us to draw, mark and construct together in an unprecedented way. In a volatile, accelerated and confusing world of pandemics, crises and complexities, we are increasingly dependent on citizens "mapping" together in order to create important data bases and patterns for smart cities that are oriented towards the common good.

At the core of this OAP, two solution pitches for map-based crowd-sourcing tools were presented. Angela Oduor Lungati, Executive Director from the Ushahidi, gave insights into Ushahidi, an open source software application which utilises user-generated reports to collect and map data. In the second pitch, Polly Hudson, Project Lead Colouring Cities, Alan Touring Institute & Tabea Danke, Project Manager Colouring Dresden, TU Dresden / Leibniz-Institut für ökologische Raumentwicklung (IöR) presented their open research platform, that enables citizens to collect, verify and visualise data on urban building stocks.

Angela Oduor Lungati took us on a journey back to Kenia of 2007, when tensions during elections ushered the founding of Ushahidi (which means "testimony" in English) as a platform solution that gave citizens the opportunity to spatially flag and record outbreaks of violence. Since then, Ushahidi offers technology to share citizen-generated data to develop solutions that strengthen their communities. Empowerment of people and inclusion in decision-making processes are the main drivers of this participatory open-source project adapted in 160 countries.

Polly Hudson and Tabea Danke moved the focus from this more holistic approach of participation to a specific area of geospatial data gathering for a space where citizens spend most of their time: buildings! The open research approach from Colouring Cities are platforms run by universities worldwide and enable citizens to collect, verify and visualize data on national building stocks. Colouring Cities aims at facilitating knowledge exchange across countries and collaboratively working on complex urban problem solving.

What are the main drivers of both solutions? What kind of mechanisms exist to verify the input of information? And how can people be convinced to share data? Those and other questions were part of our panel discussion following the two solution talks. 

Incidentally, a difference in the solutions' characteristic was highlighted: While the Coloring Cities project is sourcing a theoretically finite set of data points, Ushahidi is mostly collecting inherently dynamic and infinite data structures, e.g. on resilience networks. What implications follow from that regarding building a momentum for crowdsourcing?
Polly described how their concept facilitates institutional anchorage in diverse research disciplines. Experts for energy management might be interested in the age of buildings, something people with expertise in the arts can often readily approximate. While the building stock of a city, too, is ever changing, it happens at a much slower pace and thus invites a perspective on the city as a system.
For Ushahidi on the other hand Angela pointed out that their driver is inclusion and data equity. Momentum is thus built not from a final result or a comprehension of a system to be reached, but rather stemming from the responsibility to improve representation in an additive manner and establish new baseline patterns that were hitherto obscured.

Another discussion revolved around the possibilities and limits to verify crowdsourced data. Angela highlighted how the possibility for anonymous inputs is often a necessary condition for people to provide input voluntarily in the first place. They need to feel safe and comfortable, especially regarding sensitive issues. While this indeed may sometimes "open up the pandora box [of] bias and misinformation", several alleys exist to check such trade-offs. One can be technological (using ever increasing AI and automation capabilities and uncertainty quantification mechanisms), but for Ushahidi the instrument is more the workflows they developed around their mappings which include e.g. a group of trustees in the community or dedicated data managers who can verify or categorize data inputs. Additionally, a mix of data capturing methods can facilitate cross-verification. And, moreover, suitability-for-purpose is an important qualifier: "What's suitable for a primary school project is maybe not suitable for a policy brief", and the other way around. Lastly, Angela added that even the supposed lacks or fault lines in crowdsourced data sets may be illuminating pieces of information ushering new insights. Rather than being crowdsourcing's weak point, data validity continues to be in an exciting process of improvement, akin to the situation for classical ways of data sourcing.

Many more interesting aspects where shared, including how Coloring Cities manages to run on relatively small resources or more tangible examples of how their solutions created hitherto non-existant concrete information, e.g. the anticipation of citizens on which buildings will be demolished. See for yourself in our 25min-recording :