OpenStreetMap has a lot to share for Open Government and the Global Goals
The data is excellent, but more importantly, the 11 years of experience in OpenStreetMap holds great wisdom for any open data project.
Open data sets collaboratively created in OpenStreetMap are frequently the best available anywhere, and when there are gaps in coverage, data quality can be improved with amazing speed. And the kind of data collected in OpenStreetMap is directly relevant to Global Goals. To address good health, we need to know where health services are located and what is offered; clean water requires understanding the location and quality of sources; and onward.
Maybe you’re wondering how this actually works, and whether it could be applied to other types of data? OpenStreetMap is clearly “alive” in ways that other data sets usually are not. How can this approach be beneficial to data sets for Open Government and the Global Goals?
These are the key design elements to OpenStreetMap’s success worth emulating. Building on the Open Data Charter, but moving collaboration front and center, here is OpenStreetMap wisdom in the form of …
Collaborative Data Principles
An audacious, inspiring goal
Mapping the entire world, voluntarily running around with GPS units and tracing satellite imagery, sounds impossible … and exciting. Openly mapping the whole world is easy to understand, a vision that will draw support for a long time to come.
Anyone can contribute
OpenStreetMap is open to absolutely everyone. You can just jump in and start contributing to the goal.
Some OGP commitments, though opening government, still keep citizens and government isolated. There is a high barrier to entry to participate.
Trust the local community above all
OpenStreetMap defers to the judgement of the local community on how to map the place they live. Residents everywhere are experts on their own lives, on their place, and no one will make a better map than locals.
Partners flexibly commit to the network
Just like with individuals, institutional interest in OpenStreetMap comes and goes. Funding might end or priorities change. OpenStreetMap is not dependent on any one government, civil society organization, company, university, or international organization. Association with OpenStreetMap is voluntary, as long as you support the aim and are a good citizen. But if you have something else to do, totally fine.
Standards and schemas evolve fluidly
OpenStreetMap represents properties of all the features in the world in “tags”. There is no standards body deciding how to classify roads globally, in the Mexico, Mali, and Myanmar. OpenStreetMap discusses (and discusses again), but there is no barrier to just doing it. Consensus comes over time and gets embodied in how tools in our software ecosystem interpret tags.
This methodology in OpenStreetMap is one of the hardest for traditional GIS people and statisticians to accept. It’s messy, and a big change from traditional data management practices … and it works. And while this is a bit of a technical point, this design factor in OpenStreetMap is actually one of the most critical to its success.
In some ways, this approach is already being adopted by things like Humanitarian Exchange Language.
Experts and non-experts learn from each other
A common misconception, OpenStreetMap is not only amateur mappers. Professional geographers and cartographers have been in the mix since the start, and many of us have learned tons as we go. And the naivete of OpenStreetMap was essential; if you “knew better”, it would sound impossible.
If everyone is going to contribute, then learning is key. The capacity to work with data and digital tools, and the confidence that comes from doing it yourself, is a goal in its own right.
Sources and methodologies are fully transparent
OpenStreetMap is very careful about where it gets its data from, to be sure that the data is truly open. The community draws on personal knowledge and surveying, or from clearly openly shareable sources. This requires thorough documentation of sources.
Beyond an audit trail, the many different methodologies developed for generating and integrating data in OpenStreetMap are fully transparent, so that anyone can replicate.
A single global, universal, integrated database
There is something very powerful about working in the same database, whether you are an economist in the World Bank, or a slum dweller in Nairobi. And it ensures that data created everywhere is equally accessible, not silo’d by border or organization, but present without boundaries … just like the challenges we face in the world today.
Next week, the Open Data Charter will continue its launch and expansion at the Open Government Partnership Summit. It’s ambitious, wide ranging, and generates tons of political capital for very very good work. The focus is on open government data, and that is so necessary.
The Open Data Charter certainly includes sections that support the kind of collaboration in OpenStreetMap, like
21 e. Allow users to provide feedback, and continue to make revisions to ensure data quality is improved as necessary;
21 h. Be transparent about our own data collection, standards, and publishing processes by documenting these processes online.
27 c. … data users have sufficient information to understand the source, strengths, weaknesses, and analytical limitations of the data;
37 c. Create or support programs and initiatives that foster the development or co-creation of datasets, visualizations, applications, and other tools based on open data;
And these are great, but that is not far enough. If we hope to use data to drive extraordinary change in our planet, everyone needs to get involved. That’s why the Collaborative Data Principles are worth considering.
Recent mapping in Tanzania illustrates how citizens and governments collaborate more effectively when they work together in OpenStreetMap. Dar es Salaam is one of the fastest growing cities in Africa, with all the challenges of such rapid urbanization, while regularly facing the risk of severe flooding. Starting in 2011, GroundTruth Intiative facilitated complete mapping of Tandale ward, working directly with community members, local government, slum dwellers associations, and students.
Humanitarian OpenStreetMap Team, building on this mapping, and working very closely with local government and universities, supported by the World Bank, has over the past year helped citizens completely map the entire infrastructure of Dar es Salaam, critical data for risk analysis and improving resiliency.
Meanwhile, John Snow International, working with the national government and supported by USAID, has collected GPS tracks from medical supply delivery vehicles across the country and created a comprehensive national road database. This has been voluntarily imported into OpenStreetMap, and is now accessible to the widest possible user base.
None of this Tanzanian data would exist relying solely on government open data, nor only on citizen action. Government and citizens worked together, openly, collaboratively, and transparently.
OpenStreetMap is a model of massive and open collaboration that demonstrably works. I truly believe Open Government and the Global Goals will only be met with massive and open collaboration, especially for the data driving and measuring the goals. The world can learn from OpenStreetMap.
These are only some initial thoughts on what OpenStreetMap has to share for Open Government and the Global Goals. I would love to hear your ideas.