Tame the beast

A recent article in The Economist about data and transparency addresses the possible downsides of open government data. While the article sites some positive examples of efforts governments have made to make data more open, it acknowledges that, “whatever governments do, the presentation of endless facts can fall flat unless there are independent developers who know what to do with them.”

Even people used to dealing with data can be overwhelmed by all the raw stuff. In March Development Gateway is launching AidData, a repository that tracks more than 1 million projects spanning 81 donor agencies and totaling more than $4.1 trillion dollars over 40 years. Although clearly a rich source of information, one researcher who got a sneak preview of the database characterized it as an “unwieldy monster.”

Such a reaction didn’t surprise us. We know that it will take a variety of tools and a multi-disciplinary approach to make the data meaningful to a broad range of people. Toward that end, we are holding the Aid Information Challenge, an informal conference to bring together software developers and development practitioners to come up with applications that will make the data easier to access and more helpful to users. This event, held at the World Bank on March 12, is co-sponsored by AidInfo, ForumOne, and the World Bank. This video explains the purpose of the event. To register for the Aid Information Challenge, and help us tame the beast, click here.

The public launch of AidData will take place at the Aid Transparency and Development Finance conference in Oxford, U.K., March 22 – 25. Read more about AidData here.

Communicating the value of open data to donors

by Elizabeth Corley

At last week’s unconference I led a discussion on the topic of communicating the value of open data and standards to donors. Here are some of the key issues we discussed.

1. What is open development? Our definition was an attempt to make information about the development process open and transparent for all those who care about it and deal with it, including donors, recipients, taxpayers, regulators, practitioners, and academics. One desired end result is to create a feedback loop.

2. Data sharing should be part of the planning process. It must be considered at the inception to allow for easy sharing. There is a value in unstructured data, before it is assembled into a written document. Projects could be designed with shorter, smaller reports that include leading and lagging indicators.

3. It is incumbent on the implementers of a project to share data. One example of best practice here is the AMREF project in Katine, Uganda. In partnership with Barclays Bank and the Guardian, AMREF allows public scrutiny of its ongoing project.

After our session, the subject of communicating the value of open data and standards come up in conversation with Owen Barder of aidinfo. How do you show donors the possibility of how the data can be used (when sometimes we don’t even know)? He pointed to an example of a new technology sited in an article titled "Web Squared" by Tim O’Reilly and John Battelle.

Imagine if you had a device that could give you information about what was around you. Let’s say you walk by a school in Cambodia. You point the device at a building and on the screen appears the names of donors who funded the construction of the school or the provision of text books. Nice credit for the donors.

But imagine the school is in ruins and the device tells you how much money has been poured into it. Who would want to be associated with failure? This topic also came up in our session and it is one of culture. In some sectors, such as technology, failed enterprises means you have learned a lot. In development, particularly in Washington, we avoid public exposure and the risk of public failure. Maybe that is why we are willing to release data to the taxpayers—but only after the report has been vetted, polished, and refined for public consumption.

Camp brings together those with a passion for development and sharing data

by Elizabeth Corley

It was great to meet so many people working on issues of data and collaboration at the open development camp today. The event showed how much people care about international development and their passion for sharing data. Owen Barder kicked off the event by giving context to the challenge at hand and emphasizing the value of open data and open standards.

The suggestions for topics at the camp ranged from technical issues like maping and crowdsourcing to broader questions on communication and collaboration. You can see here what the grid looked like for session one. Over the course of the day we had sixteen sessions. Notes on each session will be posted on opendev.ning.com.

When asked for his impression of the camp, Matt Gyory, our communications intern, zeroed in on what is next. “Really the core of this day was not so much what was discussed, but the next steps that should be taken. Those that discussed data standards should try to get together to map out what they would like to see in particular and run with their ideas from there. Those that discussed the importance of measuring aid effectiveness should attempt to adopt the strategies offered. Those that discussed open data and development should attempt to add the collection of data and distribution of that data in their projects so it becomes a habit rather than an expensive addition.”

The event was co-sponsored byAidInfo, ForumOne, Global Development Commons, and the World Bank. Special thanks goes to Sameer Vasta, the bank’s social media strategist, who not only secured ample room for breakout sessions, but also delivered on the refreshments.