Blog post

When It Comes to Open Government, Nobody is Perfect

By Andrea Di Maio | March 15, 2010 | 1 Comment

open government data

Last Friday, the US Government Accountability Office (GAO) issued a report on the Implementation of the Federal Funding Accountability and Transparency Act of 2006, where it highlighted some shortcomings of USAspending.gov, the government web site providing transparent information about how the US federal government spends its money.

The report says that:

While USAspending.gov currently contains required fiscal year 2008 information on federal assistance awards from 29 agencies, 9 agencies did not report a total of 15 awards. These agencies, which include the Department of the Treasury and the U.S. Election Assistance Commission, stated that they plan to report future awards as required. Nevertheless, OMB has not implemented a process to identify nonreporting agencies as originally planned and instead has relied on agencies’ voluntary compliance with OMB guidance to ensure complete and accurate reporting. Without a more effective approach to ensuring that all agencies report applicable awards, the utility of USAspending.gov will be impaired by gaps in the required information.
In a random sample of 100 awards, GAO identified numerous inconsistencies between USAspending.gov data and records provided by awarding agencies. Each of the 100 awards had at least one required data field that was blank or inconsistent with agency records—or for which agency records lacked sufficient information to evaluate their consistency with data on USAspending.gov. The most common data fields with inconsistencies or omissions included titles describing the purpose of the award and the city where award-funded work was to be performed. These errors can be attributed, in part, to a lack of specific OMB guidance on how agencies should fill in these fields and how they should perform the required validation of their data submissions. In addition, publicly available information that OMB provides on the completeness of agency-provided data does not address a required data field relating to the city where work for the award was to be performed. Until OMB and agencies better ensure that complete and accurate information is included on USAspending.gov, the Web site will be limited in providing the public with a view into the details of federal spending.

This follows another report on Recovery.gov, where the GAO highlighted some “discrepancies between ZIP codes listed in the multiple federal databases that recipients use to enter project location information” (as reported by NextGov)

Now, I am sure that some critics and cynics will point their finger at this, saying that also those who are meant to be a  beacon for openness (as they lead the charge for open government) and have used USAspending.gov and Recovery.gov many times as examples) fail being open enough.

My take on this is that nobody should criticize anybody else. The journey toward transparency, participation and openness is not one in which government should travel alone. Nobody has the perfect recipe for sustainable improvement in data quality, completeness and transparency. Open government is an endeavor where the so-called “perpetual beta” approach is a necessary evil: everybody involved – individual government agencies, government 2.0 working group and taskforces, activists, associations, vendors – must work together to figure out what open government actually means.

In an earlier post I did mention that using an open source development approach to open data inside government would make a lot of sense. We need to move from expecting that government has complete and accurate data which we need to ask it for, to looking for versions of data that may be only partially complete, or partially up-to-date. The scenario highlighted by Nat Tokington, where government, third parties and the public at large will collaborate on gathering, cleansing, and processing data won’t be reality in the near future. However we need to start understanding the trade offs between openness and accuracy, and change our expectations about how precise or complete open government data will be.

The Gartner Blog Network provides an opportunity for Gartner analysts to test ideas and move research forward. Because the content posted by Gartner analysts on this site does not undergo our standard editorial review, all comments or opinions expressed hereunder are those of the individual contributors and do not represent the views of Gartner, Inc. or its management.

Comments are closed

1 Comment

  • Craig thomler says:

    The only reason this criticism can be levelled at the US government is because they are becoming more transparent. Gov 2.0 inevitably leads to greater criticism and media exposure as more of the mistakes and inconsistencies are available for public scrutiny.

    This is a good thing as it helps government improve.

    Just as making it easier and more acceptable in society for those facing discrimination or abuse to step forward appears to increase the perceived level of abuse, exposing and discussing government flaws appears to increase their number. But the outcome in both instances is a fairer and more equitable society (and longer term lower rates)