Jennifer Zaino
SemanticWeb.com Contributor

At last week’s International Semantic Web Conference (ISWC 2009), semantic web application development vendor TopQuadrant announced oeGOV. The initiative aims to create an open, W3C Semantic Web standards-based set of ontology models to encourage and facilitate the use of linked government data.

SemanticWeb.com conducted an email conversation with TopQuadrant’s Dean Allemang, chief scientist, and Ralph Hodgson, co-founder and CTO, to learn more about the effort to help the government sector embrace semantically linked open data.

Semanticweb.com: Why does TopQuadrant see a need for this initiative?

TopQuadrant: Open Linked Data in the government is taking off in a big way. But there is a lot of data out there that has to be presented in a coherent, reusable way. Ontologies help that happen by providing support for aggregation, provenance and data quality – aggregation through everything having URIs and controlled vocabularies, provenance in terms of who was the source and when did the data appear, quality through units of measure and data types. Government data is available to the public, but not easily accessed at the moment. It isn’t even very big data – there aren’t any serious technical barriers to making it more available. We feel that now is the time to start up an initiative to bring public data to the public.

Semanticweb.com: There’s been a lot of discussion of greater transparency in the government and certainly efforts like recovery.gov support that. But when it comes to linked open data specifically, do you see any hesitance to embrace the vision or are there other underlying conditions that have so far kept the vision from being fully realized?

TopQuadrant: Quite the contrary. Every presentation we see about open initiatives (like data.gov, recovery.gov, etc.) from government spokespeople makes a point of mentioning the Semantic Web standards (usually RDF and OWL, occasionally SKOS). When we talk to our own contacts within the government, they tell us that receptivity to linked open data initiatives is better now than it has ever been, and that in fact, genuine efforts are being made to open up data. This effort is serious. We have seen a real sea-change in how government agencies view open data in general, and linked open data in particular.

Semanticweb.com: Could you provide a laymen’s interpretation of the stated plans of the initiative to “interpret government data through precise semantics and controlled vocabularies, including geospatial and temporal aspects. Correlations and comparisons of data will be possible through the standardization of neutral models for data exchange, units of measure and transformation, along with an ontology for units of measure.”

TopQuadrant: Publishing linked open data is one thing; making use of it is another. Just because we publish data as LOD, doesn’t make it any easier to understand ‘insider terminology’. It isn’t really ‘insiders only’ – it is all published on various government sites. But we are making it available for use with all the government linked data. This makes it easier to linked information sources in meaningful ways.

Semanticweb.com: How did TopQuadrant decide what would be the first published ontologies it would provide (the OWL representations of about 500 US Government civil bodies, models of the Federal Enterprise Architecture (FEA), developed with GSA and the QUDT ontologies for units of measure, developed with NASA ARC). Are there important reasons to start here in terms of driving this movement and if so can you give some details on that?

TopQuadrant: There are a lot of good starting points; any model that has a potential impact on several other efforts is a good candidate. This choice was a combination of utility and opportunity. Identities of agencies (with official names etc.) have been available for ages, but not in a way that could be easily queried. It is a useful and easy starting point. The FEA is a bit more complex, but we have the advantage that we have already done a lot of work with this vocabulary. The additional effort we did to publish it to oeGOV was to simplify its presentation by rendering it in SKOS. The FEA is particularly interesting when dealing with budget information; for the past several years, every budget request has been required to include alignments to the FEA vocabulary. This gives its applicability a very wide scope in the US government. We have even seen interest in the private sector in the FEA. We think that making this available in every form possible will be a great boon for information management in both the government and private sector. The FEA is a good example of ‘insider terminology’ that isn’t really just for insiders; anyone in the world can look up the FEA. Now it is possible for linked data to use it for cross-reference.

Finally, the QUDT units might seem to be an odd man out here, since they do not refer to government organization or spending. But on the operational end of government, in the DoD, NASA, DoE etc., a great deal of what these agencies do is technical. Information about engineering units (as well as dimensions and quantities) has been available in engineering and scientific communities for generations, and is a key part of just about any technical work. Working with NASA, we developed a production quality version of this well-known information. Just as with the FEA, we expect this information to have utility that goes well beyond its government roots.

Semanticweb.com: Has anyone else signed on to participate? What would you like other parties to bring to the table?

TopQuadrant: All initiatives move from, as a colleague Jack Ring puts, “Awareness, through Appreciation to Preference”. The oeGov initiative has just started and we are very much at the “Awareness” stage – people are beginning to discover oeGOV. This said, we do have some people at the “Appreciation” stage with participation from colleagues at GSA and NASA. Preference will be evident when government bodies express their organizations on the web in machine-processable ways – instead of just text on HTML pages.

We’d like to see a wide range of contributions. At the level of data, any agency is encouraged to provide information about the agency itself, its projects, connections to other agencies, etc. In terms of vocabularies and schema, we welcome contributions from any group who advances notions of openness in data, in particular government data. In particular, we feel that organizations like The Open Group (e.g., TOGAF) and OMG have charters that are very much in alignment with the goals and values of oeGov, and we would welcome contributions from such arenas.

Semanticweb.com: What do expect to be the ultimate outcome of the project in terms of the benefits it will deliver both within the government and to the citizenry?

TopQuadrant: We feel that the whole Linked Open Data trend in government will lead to a greater awareness, first by ‘watchdog’ groups, and eventually by private citizens, of what the government is doing with their money. There is a perception in this country – and indeed around the world – that the inner workings of government are inscrutable and inaccessible to ordinary people, because of the high complexity of political and fiscal processes. But just as the advent of the World Wide Web has changed how ‘ordinary’ people relate to information about just about every aspect of our lives, we expect the linked open data web to change the way private citizens can view their government. We hope for a day when wheeling and dealing about government spending becomes so transparent that lawmakers will have to be just as responsible to their electorate as they are to the private interests who fund them.

What role can oeGov play in this process? We aren’t turning all of the government’s data into linked data form, nor are we organizing efforts to hold government. We aren’t even providing cool applications that use this government data. Our vision is to build a community devoted to enabling this kind of use of government data. We have begun by providing a few key building blocks that illuminate the meaning of the currently available data sources – we want to build up to a full community of contributors, to make oeGov a hub where anyone can go to find resources that will help them to make sense of the stream of open linked data.