The federal government has the responsibility of taxing us and spending our money in beneficial ways. In some cases we are well aware of what we get for our money — most people use highways and weather forecasts on most days. But in other cases, the product of the government’s best efforts is hidden away and hard to find.
This is especially true for the huge number of datasets that the government collects, compiles and makes freely available. Navigating different federal agency web sites can be a daunting task. Which is why we were so pleased to find out that a new private web site has done much of that work for us.
The USGovXML.com site is the brainchild of Robert Loftin, a software developer with recent experience in the worlds of finance, telecom and pharma. Loftin has an Electrical Engineering degree as well as a Master of Science in BioMedical Engineering which explains his nuts-and-bolts approach to data access.
The goals for the site are modest yet potentially huge in scope:
USGovXML is an index to publicly available web services and XML data sources that are provided by the US government. USGovXML documents, in one place and in a uniform manner, the web services and XML data sources that are provided by the US government.
From this description, one might think that his site is in direct competition with the much ballyhooed Data.gov site. But Loftin’s approach to data access is from the standpoint of a technically savvy data consumer and therein lies a huge difference.
Whereas the Data.gov site appears to be targeted at the general public with the goal of displaying the breadth of publicly available data, the USGovXML.com site is decidedly targeted at data consumers with programming skills and helping them make use of data-oriented web services that already exist within different agencies.
Loftin’s site includes comments describing exactly how to access information from each site and “source code snippets to help developers better understand how the data sources can be used”. To a developer, this is where the rubber meets the road. Freely available data is great but how do we use it?
In essence, Loftin is single handedly putting together Application Programming Interface (API) documentation for a wide variety of freely available government data services scattered across different agencies. He’s not trying to force everything to fit into some new model. He’s merely documenting what is out there and how to work with it. Now that is a worthwhile effort!
We recently interviewed Loftin to better understand his motivation for creating the site.
He said he had been taking a sabbatical from consulting work and was using the time to learn more about web services and XML data sources. He thought government sites would have lots of freely available web services to work with. What Loftin found was that federal web services are scattered all over with little to no uniformity. He began creating his own database describing how to work with each subsequent web service and eventually realized that this could be a useful resource for others. After rejecting the idea of publishing a book with this information he set up the USGovXML.com site in November of 2008. Loftin has no current business model for the site and intends to keep it open as a useful resource for those hoping to create applications based on government data. A very noble effort indeed!
When asked for his take on the current state of affairs with publicly accessible government data and possibilities for improvement, Loftin’s immediate response was positive:
What I found made me feel good about how our Information Technology tax dollars are being spent.
He said he felt various agencies were working hard to provide access to their data but that he found himself doing a lot of reverse engineering to learn how to use those data services. (From our own experience we know that many federal agency web services are designed to work with software clients at partner agencies. There is often little encouragement and zero financial support to make web services available in a publicly useful manner. This is an area where funding priorities need to be re-evaluated.)
Even in cases where complete Web Services Description Language (WSDL) documents were available, Loftin felt he would have been helped greatly by seeing some example data. Better yet would have been some fully fleshed out code examples of how to work with the data — precisely what Loftin is providing in the USGovXML.com site.
When pressed to pick one area where he wished the government would impose a standard Loftin, after a moments thought, chose “Latitude / Longitude”. Working with the Recreation Information Database, Loftin found out the hard way why you have to be careful working with data compilations:
- lats and lons as degrees-minutes-seconds
- lats and lons as degrees-decimal minutes
- lats and lons as decimal degrees
- longitudes with W longitudes positive
- longitudes with W longitudes negative
This is why it is so important that data managers work directly with data consumers to validate data compilations. Only those with an interest in the data can provide the the kind of high level “does-this-make-sense” validation that moves Data into the realm of Information.