Two weeks ago, we were excited to read that the World Bank took a public position explicitly endorsing citizen mapping tools that keep control of citizen-generated geo-data in the hands of the folks that created it, not the organization providing the mapping tool used to generate the data. This statement was in direct response to months of concerns expressed by a range of organizations and experts, including Global Integrity, over the Bank's deal with Google to provide quick access to Google Map Maker data for humanitarian purposes. That agreement, which has never been made public, seemed at times to endorse the use of Google's Map Maker tool itself, which puts user-submitted geo-data in Google's control. Jon Mitchell at ReadWriteWeb has a nice wrap-up and summary of the issues here.
The blissfully clean language from the Bank is worth repeating:
"…the World Bank only supports citizen-mapping efforts that give users free access to the map data they create. While citizens are free to choose the projects and tools that best meet their goals, our guiding principle is simple: if the public helps to collect or create map data, the public should be able to access, use and re-use that data freely."
At Global Integrity, we're encouraged by this clear, direct policy and applaud the Bank for taking this step, which no doubt annoyed some friends in Mountain View.
For us, a final concern remains, which we hope the Bank addresses in the future: the fact that the underlying agreement between Google and the World Bank remains publicly inaccessible.
As we have blogged previously
, Google has declined to release the agreement despite protestations from colleagues at the Bank that nothing in the agreement is actually "secret" or disparaging. I trust that to be the case, but none of us have any way of knowing what other concessions Google asked for (and received) from the World Bank as part of the agreement. For example, is the Bank obligated to report to Google on how governments and multilaterals use the Map Maker data released under the arrangement? And if so, should we worry about that reporting or not? We have no way of knowing.
In the future, I can see little rationale for why an open data agreement shouldn't, always and without exception, be a public document. It cuts completely counter to an open data ethos to execute a deal to ostensibly liberate vast amounts of previously shrouded information while simultaneously arguing that the underlying agreement is too sensitive to release. If we want to encourage more public-private partnerships around "open" and "big" data, a guiding principle for all parties involved should be that the terms of the partnership need to be as transparent as the information being liberated.