Open Belgium
It seems we still live in the infancy of open government data, which is all about availability and quality. In this post we submit 9 suggestions to usher in the mature age: open data as a prized participation tool.
1. Ease of use “Dataset” is too often a synonymous for off-putting and tedious. This isn’t inevitable, as the following ideas show. Simple and eye-catching data visualisation can boost a platform, making for instance use of infographics as appropriate. Structuring data with an emphasis on key content and links to supporting information rather than a one-block presentation makes users’ life easier. A personal dashboard where e.g. favourites can be pinned prompts more recurring visits and more analytical use of the data. Search by keyword is indispensable to find your way around quickly and to drill down into the data. Aggregating relevant data within and across organisations instead of focusing on your own production has the potential to make your venue a by-default go-to when it comes to reliable government information for the area. For example, a city council could curate its own data but also integrate data curated by central government and open source data from not-for-profits and businesses.
Addressing the semantic links between various pieces of content can also be a huge plus. “Local Decision as Linked Open Data” by the Flemish Agency for Domestic Governance seeks to add metadata to council decisions so they can be processed automatically into structured systems linking them to other relevant acts (e.g. other council proceedings or policy memos from the administration). The Unlock software detects semantic relations between heterogeneous pieces of information relating to urban planning and presents them in thematic groups, without the need for manual tagging. Good news is both projects will be presented at Open Belgium 2017.
2. Civic data education There is much to be taught about government data: how to find, understand, interpret, critique it, how to ask the questions that it can answer, how to use it as an evidence, how to find its applications in real life. The often-lacking civic education in schools should be improved to support more engagement later in life. And while we’re at it, it should also be extended with modules on handling open data. One such example called the Urban Data School has been piloted in Milton Keynes, UK, which researchers found to be a success. They note that of course teachers need to be trained in the first place. So, why not extend adult training to wide ranges of the population?
3. Contextualise Data doesn’t always make sense taken in isolation. Providing context on a raw dataset can render it intelligible and make its practical relevance more obvious. One example is to present transcripts of assembly hearings together with a note on who speakers are, including their political membership, links to lobbies, financials contributions they’ve benefited from, or yet what work they’ve been responsible for. DigitalDemocracy.org does this, relying on the dedication of political science students. Yet one can imagine configurations where contextual information is automatically bound with data, e.g. via the adjunction of news feeds.
4. Answer data demand An excellent way to create databases that are actually used is to let users have a say about what they want in it. A straightforward means of doing that is freedom of information (FOI) requests . Groups or individuals make a claim for punctual data and access is provided on a case-by-case basis. The problem here is that FOI procedures are mostly unpopular or just unknown, and the result of a claim –be it positive or negative– lacks publicity. Hopefully, civil society can rise up to the challenge. The website WhatDoTheyKnow facilitates FOI on top of being itself a repository of collected answers. Taking this logic one step further, one can imagine a crowdsourcing of problem and data definition.
