Project MyUSA

Last week we announced the MyGov Discovery Bar. This week we’d like to dive a bit deeper by taking a look at the API that powers it, with the hope that developers within the open-government community can help make it better.

Thy Discovery API exposes all the information that’s normally available via the MyGov Discovery Bar (code) in an open, machine-readable format. In fact, the Discovery Bar itself consumes the API — we eat our own dog food. Why would we do this? Two reasons: first, exposing an API to collect and retrieve information at the outset and then going back to build the layer to display it forces us to separate data from presentation and allows the application to be more flexible.

The Discovery Bar may be how the information is used today, but that doesn’t mean it’s always going to be that way and we encourage members of the community to build apps using it. Second, by using the API to power the Discovery Bar, we can see the data from the perspective of an outside developer and can hopefully, in the end, produce a better, more useful API.

Using the API

So what can the Discovery API do? Currently, the federal .gov footprint is organized around how government is structured. If a citizen wants to complete a given task, they have to know what agency to go to, and often times, what office within that agency. The Discovery API layers a folksonomy on top of the entire .gov landscape, allowing developers to query pages by keyword, or more importantly, to find content related to a given page.

Think about it this way: if you decide to change your name (perhaps you’re getting married), you’re probably going to want a new passport. But if we knew that task was also closely related to getting a new social security card or driver’s license, at least we could point you in the right direction. But that’s not the only way the Discovery API can be used. What if you want a listing of the top government content by tag, or a timely subject? Just query the Discovery API by tag.

Under the Hood

The Discovery API breaks the .gov space into pages, each represented by a uniqiue URL. Pages can have tags which in turn provide related pages. Each page is organized by domain (the part before the .gov in the URL), and domains can be queried by reverse domain name (e.g., gov.whitehouse.). This way, if you want all the subdomains of a given agency, you can get that list in a machine readable format.

The Discovery API launched with some core seed data, namely each federal executive .gov, and those pages available within one click of the home page, but over time, as people use the API, it can grow to become a comprehensive index of the most popular government pages.

Interested in giving it a try? Kick the tires in our API Sandbox or take a look at the code powering it.