OpenSource Connections
Category: Uncategorized
On December 8, 2009, the Executive Office of the President of the United States issued a memorandum providing guidance and directives for implementing the Open Government policy. While the directive from Mr. Peter Orszag did direct some clear action to be taken in the immediate future, much of the directive simply instructed agencies on providing future guidance on how to implement the Open Government policy. Given that 11 months has passed since the initial Presidential directive, it seems optimistic to expect much movement in the next 90 days on implementing the policy.
There are some clear deliverables for agencies to meet:
- Websites which publish usable information to the public. Generally, this will follow the www.agency.gov/open format. Agencies will be required to provide a feedback mechanism for the public to interact with the agencies regarding what information is published. Agencies are also required to publish 3 “high value” data sets which have not been previously viewable to the public.
- What I expect. This is the clearest specific deliverable set forth in the entire memorandum. I expect varying levels of compliance, but I do think that most agencies will have pages up in differing states of completion. I also expect pretty serious data integrity issues. Recovery.gov had noteworthy data integrity issues, as documented in this Government Accounting Office report, and I anticipate similar issues within the agencies.
- Publication in open formats. While not directing what formats are required, the directive does specify that information must be published in open, searchable, parsable formats. The publication requirement is subject to “valid privacy, confidentiality, security, or other restrictions.”
- What I expect. The “other restrictions” requirement leaves a very large loophole for interpretation, so until more specification is given, I do not anticipate a flood of information flowing from agencies. There will also be a debate on the interpretation of open formats, so I expect publication in a wide variety for formats including, among others, XML, XBRL, PDF, and Microsoft Office formats.
- Integration with existing dashboards and reporting mechanisms. Agencies must publish information in a way that is integrable with dashboards such as Data.gov, the IT Dashboard, Recovery.gov, and USASpending.gov.
- What I expect. Until more specific guidance comes out, this may be a hand-waved requirement, particularly given the existing data integrity problems previously mentioned. The memorandum identifies information quality as a concern, but provides no concrete information on how to address the issue. Until there is a mechanism for vetting and validating information publication, this process will be subject to gaming.
- A presumption of openness. While this presumption is specifically laid out in covering FOIA issues, the theme runs throughout the directive – be more open. For the smaller agencies, this might not be as difficult to execute as for the larger agencies.
- What I expect. This will be a long slow road to go down. The technology exists today to make publication (and validation) of information viable. Its existed for quite some time. Technology is not the hurdle to overcome; culture is. The attitude of the administration is laudable – they are trying to bring transparency, accountability, and collaboration to the government for the people whom the government serves. Penetrating the entrenched layers of inertia and process will take more than speeches and broad, limited specification memoranda.
The effort towards more open government is a signal of the right things coming from the administration. However, particularly in understaffed agencies, this directive may create more wheel-spinning than action, or, as my old Ethics professor Ed Freeman would say, more heat than light.
Given the short timelines involved for some of the deliverables, I expect systems which can rapidly stand up and rapidly scale to gain ascendancy. This means, in my mind, either Microsoft Sharepoint portals will pop up everywhere (a real possibility given the conversion of recovery.gov to Sharepoint and the citation of a Microsoft executive in Washington Technology), or open source proponents Kundra and Chopra will get their way, and a system such as Drupal will appear in many .gov domains. Regardless, I do not see this as a great entry point for new services to the government.
There is also an opportunity for a data aggregation service to appear from the publication of previously unpublished information. Solr would be a great engine for finding information across agencies, and performing analytics against that information is only a step further. I expect lobbyists to demand value-added analytics of what comes out of the agencies as a result of this directive, and if the information is truly published in an open format, then providing the technology to meet the demand is not difficult. I do not anticipate a long wait for K Street to begin using agencies information to its advantage, if and when we truly start seeing accurate information published as per the intent of this directive.