3/4/2017

Crowdmapping as a new data source for journalists

 

Crowdsourced data, especially for mapping, is a boon for data driven journalism. In 2015, Nepal’s earthquake was mapped in an astounding 48 hours. The number of volunteers increased to over 2,400 mappers, most of them international, a number that increased exponentially from the initial range of seven to 100 mapping volunteers present before the earthquake occurred.

nepal-earthquake-map2.jpg

Image: Crisis map of the Nepal Earthquake. Source: Kathmandu Living Labs.

A significant use of crowdsourced data for mapping, or crowdmapping, is to inform crisis responses like the Nepal earthquake by providing a medium for citizens to communicate with one another and with those seeking to help victims. The benefits to affected peoples are immediate information sharing and visualization of dire and urgent events. These apps have the ability to fill information gaps and even provide aid for disaster victims. Volunteers from across the globe also can contribute to crowdsource entire maps of post-disaster road infrastructures and refugee sites. As a platform and medium, crisis mapping has become so popular that it is increasingly replacing traditional mapping methods for humanitarian emergencies. This is also a huge benefit to journalists as they demonstrate connectivity between open source software, humanitarian crises, and crowdsourcing. According to the Tow Center’s Guide to Crowdsourcing, “Crowdsourcing allows newsrooms to build audience entry points at every stage of the journalistic process—from story assigning, to pre-data collection, to data mining, to sharing specialized expertise, to collecting personal experiences and continuing post-story conversations”.

The wow factor

Crowdsourced apps, like OpenStreetMap, Harassmap, Anti-Eviction Mapping Project, FixMyStreet, allow for the rapid collection of reports and data from the ground. Individuals can easily contribute and share information in ways that were previously impossible to do. The apps provide real-time information, and cheap or even free communication, create situational awareness, produce additional data unmatched by previous mapping processes, and increase connection between online and offline networks. The apps have changed this transfer of information, for the good of those affected by crises and the journalists reporting on them. In particular, in situations where government gatekeepers have limited the flow of information, crowdsourced data has enabled journalists to report more freely and in-depth on emergencies.

harasmap.PNG

Image: Harassmap.

The reality in 5 points

But let’s get real. Crowdsourced apps have a highly nuanced and complex process with many problems. Here’s five points.

1. Some crises are sexier than others

Like it or not, some disasters are more attractive than others. A sexy disaster brings volunteers, a team, resources, media, an entire international community watching. There is traction. If the disaster is not sexy, then no one will develop the app and few will contribute to its development.

There’s something urgent about a crisis that can draw in 1000s of volunteers to spend long hours digitizing broken infrastructure and geocoding text messages. After all, “you’re saving lives”. For longer, more chronic crises, whether they’re about lack of affordable housing and evictions or long standing problems of sexual harassment--yes, both of these have been crowdmapped--the immediacy is lacking. It might be easier for a remote mapper to say, “It’s not my problem” or, more insidiously, “It’s their fault”. There’s also enduring issues of digital inequity. It will always be easier for certain people to contribute their concerns.

Due to the unevenness of contributions, the implications for journalists is that this may follow their worst impulses, to “chase ambulances”. They may have to work harder should they wish to track a crisis lacking a large following. The same argument applies if they are asking the public to contribute to a story for them.

2. These apps are far from being zero-cost

The majority of the time community development disasters are not sexy. These cases are recurring events and instances of community building, public policies, community assets, and NGO work within the community. They are often chronic, and it is hard to capture and crowdsource data for them. Yet, these apps promise the same positive results. Practitioners can leverage time and cost efficiencies, while  individuals can tap into a larger pool of talent, activate more participants, and create greater ownership of the process.

These crowdsourced apps are often touted as being easy to use with effortless app development. But this definition of ‘easy’ varies greatly, depending on context, community, and group. In practice, their implementation is experienced very differently and deemed as ‘difficult’ by many - and this leads to sampling bias issues.

Moreover, the people developing these apps require significant skills to complete them. There are many steps involved: verification, documentation, integration with other systems, SMS debugging, and taxonomy development are all critical components that require considerable technical expertise.

Additionally, the actual cost of deployment is not what it seems. Free and Open Source Software is often not really free. Telecommunications hardware like a GSM modem and an SMS plan are all additional costs that need to be taken into consideration when setting up new mobile apps. The administration of the apps also requires human resources, commitment and new job creation for sustainable outcomes.

3. Participant engagement is opaque

There is a discrepancy between who participates and contributes to crowdsourced data, and whether they are local or remote. Originally, these apps were used for crowdmapping disasters, where users were local but almost exclusively the app developers were remote, albeit some with local connections. In community crises, local context and local capacity in the mapping process means everything. Community development practitioners know that this is a labor-intensive and high-contact face-to-face activity. Therefore, developing a crowdmapping app depends on the intensive building of a local technical base. This is problematic when there are no local technical experts to be upskilled. It then becomes critical to have a sustained physical presence in the communities, which in turn builds trust: a key factor in aligning local and remote participant goals with those of the community in order to garner a sense of reciprocity.

4. The problem with “disruption” as a transformative tech

Furthermore, disintermediation effects are ignored in the disruptive tech process. The communication medium is crucial yet often gets left behind when these apps roll out. Knowing the answer of how participants will communicate can make or break the success of an initiative. As we have noted, in many cases, participating community members do not possess requisite skills for the specifics of the applications. Ironically, when attempting to diffuse the skills through manuals and tutorials, we -- the researchers -- are often the ones who gain new skills in deploying telecoms hardware and software with Crowdmap and their administration.

Indeed, any new technology poses challenges in adoption. Interest in a technology project cannot be based on the technology community’s uses. A community organizer may own a smartphone, and not use the crowdsourced app. An activist may have a blog and Twitter account, and not have a Facebook account. On the contrary, they feel strongly against it.

There is still have a long way to go. These hurdles get in the way of content and the natural process of communicating messages from participants. And so finally, when the data is on the map, the content is fragmented, which makes it difficult for a journalist to write a cohesive story with the information available.

5. The technical literacy of journalists

Reporting has now been taken to a new level with the availability of digital tools. When using crowdmapping, technical literacy, in addition to data literacy, is often overlooked. Journalists need to correctly understand the way that these platforms function, their protocols for releasing information and the internal teams required to clean, geo-locate, anonymize, and validate the data. If they lack this understanding can lead to data privacy concerns, which may arise when apps encounter technical issues. For instance, a major security bug that can take a significant amount of time to fix, with sometimes irreversible consequences. On another level, maps may act as a witness to crimes committed by the state, such as harassment, detention or torture. Therefore, journalists must be careful not to reveal sources via data contributions.

Although the vastness of information sharing across these platforms allows for flexibility in the way journalists report stories, validation and fact checking in real-time is often problematic. Being on the ground and being online requires strong efforts of collaboration and a critical approach to any crowdsourced data that is being shared. Moreover, technical literacy needs to be interwoven with existing standards of journalistic ethics and integrity.

This article is based on the authors' research article “Confronting the hype: The use of crisis mapping for community development”. Read the full article here.

Comments