DOD looks to semantics for better data-sharing, cost savings

In its ongoing quest to catalyze cost efficiencies and improve information-sharing, the Defense Department is increasingly looking to IT to solve problems of all sizes. The latest bid involves high-tech search capabilities, interoperable data and a futuristic, data-rich internet known as semantic web.

In a new RFI, the Defense Information Systems Agency and Deputy Chief Management Office are looking to strengthen interoperability and data-sharing for a vast array of requirements through an enterprise information web (EIW). Their envisioned EIW is built on semantic web, which will allow better enterprise-wide collection, analysis and reporting of data necessary for managing personnel information and business systems, as well as protecting troops on the ground with crucial intelligence.

“At its heart, semantic web is about making it possible to integrate and share information at a web scale in a simple way that traditional databases don’t allow,” said James Hendler, senior constellation professor of the Tetherless World Research Constellation at Rensselaer Polytechnic Institute.

One way semantic web helps is by standardizing information to enable databases to better communicate with each other – something that could be particularly helpful for DOD’s diverse systems and lexicons.

“The information necessary for decision-making is often contained in multiple source systems managed by the military services, components and/or defense agencies. In order to provide an enterprise view or answer questions that involve multiple services or components, each organization receives data requests then must interpret the question and collect, combine and present the requested information,” the RFI reads.

“EIW, through semantic web technology, should allow the DOD to have visibility of personnel, pay, logistics and other pertinent data at the enterprise level using authoritative data from each of the services,” the RFI read.

That’s key for an agency that has thousands of business systems and data bases – many of which are not interoperable – and happens to be facing some of the most dramatic budget cuts in department history.

“DOD historically spends more than $6 billion annually developing and maintaining a portfolio of more than 2,000 business systems and web services. Many of these systems, and the underlying processes they support, are poorly integrated. They often deliver redundant capabilities that optimize a single business process with little consideration to the overall business enterprise,” DOD Deputy Chief Management Officer Beth McGrath said in an April 4 memo. “It is imperative, especially in today’s limited budget environment, to optimize our business processes and the systems that support them to reduce our annual business systems spending.”

To do that, DOD must break down all of the data elements and identify what’s mission-critical and what’s redundant, according to Brand Niemann, senior data scientist at Semanticommunity.net and former senior enterprise architect and data scientist at the Environmental Protection Agency.

Once the mission-critical is identified, building semantic web on top of it would mean that data can be easily searched and translated – for example, numerical data can be pasted between documents as actual tables rather than jumbled characters lost in HTML translation. Pieces of data become apps that can integrate, Niemann said.

“This is about making apples and oranges into apples and apples – they can talk to each other. That’s semantic interoperability,” he said.

For example, in a collaboration on lead poisoning between the Environmental Protection Agency and the Centers for Disease Control , lead may be referred to in some documentation as “lead,” or it might be labeled as its atomic symbol, “Pb.” With an agreed-upon mapping with semantic web, information could still be accessed and used despite the incongruent terminology, Niemann said.

“Most people are familiar with the web we use today, which is about documents – like HTML or PDF. Semantic web is all about data and putting that data on the web,” he said. He added that semantic web data is accessible in four clicks: access, filter/sort, download and share, including via mobile device.

It’s proving invaluable for government sites like Recovery.gov and Data.gov, which are using semantic web to create dashboards rich with information that can be downloaded and is easy to access and understand.

Now Niemann is using an inventory of DOD’s thousands of information systems, to create a dashboard that categorizes the systems and plans for each, such as whether to transition, retire or rebuild, he said. Such a dashboard could play an integral role in combing through DOD’s glut of systems to target redundancies and mission-critical systems and data.

“We are the biggest, and so have failed in the biggest way in eliminating duplication,” Dennis Wisnosky, DCMO CTO, said at the Enterprise Architecture conference in Vienna, Va., on Nov. 7. He added that DOD must focus not on picking winners and losers, but on looking inside systems to determine their functions.

“It is the only way [we’re] going to make this small enough to be managed,” Wisnosky said. “It all begins with understanding the data; that is what the semantic technology push is all about.”