NIST focuses on measurements for new IT technologies
Officials hope to develop predictability in information systems
- By Florence Olsen
- May 30, 2005
NIST's U.S. Measurement System Web site
What if it were possible to predict with scientific certainty that a complex information technology program such as modernizing the nation's tax-processing systems would be successful?
National Institute of Standards and Technology officials think it could happen if scientists make significant measurement advances in information science. "I hope it's possible," said Steven Ray, chief of NIST's Manufacturing Systems Integration Division. "Significant progress is possible."
Such is the thinking of NIST officials as they embark on a plan for meeting the nation's most pressing measurement needs in information science, nanoscience and other fast-growing areas of innovation.
NIST officials have dubbed the plan "Roadmapping America's Measurement Needs for a Strong Innovation Infrastructure."
"Measurements don't drive innovation, but they're certainly an important underpinning," said Mary Saunders, chief of NIST's Standards Services Division. Innovation, she said, is "more and more critical to U.S. competitiveness."
Standards developers, accrediting organizations, national laboratories, other science agencies and trade associations have a stake in the effort, Saunders said. NIST will host a series of workshops this year and a national measurement summit next January on information science and other areas of innovation, she said. At the summit, information systems experts, among others, will try to answer questions such as:
What do you need to measure that you can't measure now?
What can you measure but don't have confidence in its accuracy?
What can you measure in the United States but know that measurement is not accepted in foreign countries?
NIST officials have announced the first series of 10 workshops in which they initially will gather information about measurement needs in the magnetic data storage and broadband telecommunications industries, among others. They will create a measurement needs database to manage the information, Saunders said.
Ray thinks that information science cannot advance without a rigorous approach to measurements. "How do we engineer things in IT? Well, we engineer things very empirically," he said. "And how do we predict their behavior? Well, we don't predict their behavior. That's why we build complicated systems and hold our breath and do lots of testing before we've determined whether or not a thing is going to do what we think it's going to do."
The software industry is among those that could benefit from the NIST initiative.
"We're certainly strong advocates of good standards, and measurement does provide the building blocks for good standards," said Gregory Friedmann, director of public relations at the Systems and Software Productivity Consortium, which represents aerospace, defense, government and other industries that build mission-critical software and information systems. "To the extent that we understand what [NIST] is planning to do, it sounds like a very good initiative that our members should benefit from."
Ray has no doubt that NIST is on the right track with its measurement road map. "If we ever hope to be able to engineer ever more complex systems, which we absolutely intend to do, then we have to try to come to terms with some units of measure," he said. "Now, what are those units of measure? That's a completely wide-open question. I have some ideas. I don't have the answers. I have the questions."