Editor’s note: Guest author Kate Ray is the co-founder of Nerd Collider, a new Q&A website designed to bring a diverse group of nerds to a temporary page to solve a problem.
Data.gov is coughing blood. The budget being negotiated by Congress will reduce the Electronic Government Fund by 75%, and the same open-gov advocates who championed the project for two years are now taking up arms to save it by persuading people of its importance. Data.gov has had around 1.5 million downloads in the past two years, but the impact of those downloads hasn’t been made clear enough and the site is difficult to use.
This past week, I started a discussion on how to turn the project around. It drew great responses from people knowledgeable about Data.gov, but despite invitations, neither the Sunlight Foundation nor a representative of Data.gov took part. After a week of discussion, a series of common, simple questions emerged and as calls to #SaveTheData intensify, we still don’t know the answers to them.
Questions for Data.Gov
1. Who are your users?
Any startup raising money is immediately asked: Who is your market and how will you reach it? It’s unclear whether data.gov is aimed at journalists, academic researchers, entrepreneurs, or special interest groups. “Different user groups will have different interest in open government data and will have different barriers to using it. So I think the next step is for us to “unpack” the word user and have a serious conversation about who the intended users for open government data are and what their needs/issues might be,” wrote Justin Grimes, a PhD candidate at the University of Maryland researching open government data initiatives.
2. If your users are programmers, what’s in it for them?
Anil Dash just wrote a cogent post about some of the successes of data.gov (particularly health.data.gov) but wrapped it up with a burdensome, do-gooder imperative, “As a community of developers and technologists, we have to” build apps on the data. Reaching out to programmers by telling them what they should do won’t sustain the program for long. NYC BigApps – an app-building contest with cash rewards to increase the city’s transparency – is a good example of a sexier way to market gov data. Food+Tech Connect founder Danielle Gould suggested enlisting media companies and venture funds to identify startups that currently use data.gov and connecting with them through media and events to learn what they wanted.
3. If your users are Normals, how are they supposed to use your site?
Barriers to entry on the site are extraordinarily high. Data is available for download in XML, CVS, RDF, and other formats that most people wouldn’t know what to do with. One person who joined the discussion imagined a Kickstarter-style profile page that would showcase apps built on the data and let people make suggestions. “I like the idea of embedding or linking apps’ description pages through to something like challenge.gov with bounties,” added James Forrester from data.gov.uk.
4. Why is it so hard to find data on data.gov?
It’s difficult to visualize the full extent of data.gov’s database and to sift through it for the correct piece of information. Even data.gov’s ‘Internet Web Expert’ James Hendler said, “Searching for data on the site is tough — keyword search is not a good way to look for data”, and talked about efforts to use more metadata. Another respondent recommended products like MIT’s SIMILE as a way to look at all of the data as a cohesive set.
5. What are your metrics of success?
A huge reason that the debate over data.gov’s future is so contentious is that it’s hard to judge whether the project is succeeding or failing. A number of people argued that a graph of visitor traffic to the website was a false metric because success is represented in the number of downloads of data, or simply the fact that “the right people…do visit those pages sometimes and get the things they need to.” Just as data.gov was established to help us hold government organizations accountable, we need clearer benchmarks to hold data.gov itself accountable.