Systems R&D focuses on response to terrorism
Connecting state and local government leaders
REDONDO BEACH, Calif.'In the wake of Sept. 11, government IT shops have changed more than just how they manage systems. They have also shifted the agendas of academics conducting research into governmental IT.
REDONDO BEACH, Calif.'In the wake of Sept. 11, government IT shops have changed more than just how they manage systems. They have also shifted the agendas of academics conducting research into governmental IT.
During last week's National Conference on Digital Government Research, Paul Rosenbloom, a computer science professor at the University of Southern California, described a workshop he'd held to explore research topics relating to disaster response.
'We held the workshop in New York, near ground zero,' he said, to underscore the urgency with which government agencies must improve their ability to respond. He, like others at the conference, hopes to get National Science Foundation research grants.
'Can we help with responses to unexpected events?' Rosenbloom asked. He intended to answer that question at the workshop, which was aimed at meeting NSF's often-overlooked national defense mandate. NSF, which sponsored last week's conference, expects to dole out $285 million next year to academics searching for IT solutions to government's problems.
Triage for disaster
Among the conclusions from Rosenbloom's workshop: 'Break down the wall between research and operational environments,' he said. And the federal government should consider establishing what he called centers of expertise for three classes of disasters: natural, accidental and terrorist.
Gary Strong, an NSF program manager, told of research into intelligence applications and how the terrorist attacks affected it. He cited legal impediments to the integration of intelligence databases, an effort that has become more urgent since Sept. 11.
'We can't have a single knowledge base. That's illegal,' he said. Cross-agency queries are also often illegal on Fourth Amendment grounds.
There are plenty of obstacles to using software agents to search databases as well. 'You can't send executable code across firewalls,' Strong said.
But, he said, the situation isn't hopeless. By using the legal grounds of probable cause, agencies might get around strictures against data sharing, at least in specific cases.
Strong hopes to find funding for researchers to create and work in a prototype intelligence environment. One technical hole is the lack of usable data. Real data is closely held by agencies. One professor, Bloom said, offered to create a large database of fake information, 'but the data wouldn't be dirty enough' to simulate reality, Strong said.
The DG.O conference, as the annual NSF conclave is known, had the usual science fair-like presentations of research projects.
For example, Howard Schultz, a research professor at the University of Massachusetts, demonstrated a system for using high-resolution video images taken from airplanes to populate geographic information databases'the opposite of the typical use of databases to generate images.
Working on the wireless
But Sept. 11 clearly has had an effect on IT research. Daniel Devasrivanatham, from the wireless systems group at Science Applications International Corp., described research to improve interoperability of wireless systems during disaster responses.
Since the terrorist strikes, 'we haven't done much for interoperability. We need a comprehensive approach to emergency communications,' he said. Research is needed in everything from who's in charge to overcoming interference among competing wireless systems, Devasrivanatham said.