This paper examines trends in computer usage and the effect on productivity growth for a cross-industry panel of firms during the period 1977-93. We link firm-level computer asset and financial data from a variety of public and private data sources, including Computer Intelligence (a market research firm), the Census Bureau's Enterprise and Auxiliary Establishment Surveys, and Compustat. We find that computers-especially personal computers-contributed positively to productivity growth and yielded excess returns relative to non-computer capital, providing further evidence refuting the Information Productivity Paradox. These results appear robust with respect to econometric specifications and to choice of data. Moreover, our results suggest that the excess returns from computers first increased and then decreased over our sample period, and reached a peak in 1986 or 1987. In addition, we find evidence that computers are complementary with skilled labor and that they help reduce inventory levels.

The oil spill trajectory prediction for the barge Buffalo 292 spill was provided by NOAA and TGLO. The bulk of the 5000 barrels of IFO 380 that was leaked moved rapidly through the Galveston Channel entrance and into the Gulf of Mexico as a result of a strong meteorological event. Because of the nature of the product, it was possible to track the resulting slicks for more than 3 weeks. Initially, the oil trailed east away from shore. Changing winds and currents moved the oil south and west, leading to sporadic impacts along the shore from east of Galveston to south of Corpus Christi. Trajectory forecasts were used to alert response personnel of impending beach impacts and to direct offshore skimming operations. Real-time current and wind meters, oil-tracking drifters, visual overflights, and remote-sensing observations provided an unusual amount of calibrating data for trajectory forecasting. This fact, along with detailed analysis assisted by computer models, allowed for a surprisingly high degree of accuracy in trajectory prediction in spite of complex current and wind patterns and changing wind drift factors for the product as it weathered. In this paper, these favorable results are compared with results of an earlier spill in the region where fewer resources were available for trajectory analysis.

A software development team in the Hazardous Materials Response and Assessment Division of the National Oceanic and Atmospheric Administration (NOAA/HAZMAT) implemented software usability engineering techniques to design a new version of NOAA's oil weathering and spill management program, ADIOS II™. Their purpose was to ensure that both novices and experts can easily use the program for their own work. To achieve this end, team members used design features intended to help novices, without impeding experts. These features included progressive disclosure, incorporation of documentation into the program interface, and error-checking safeguards. The team empirically checked the effectiveness of their design in a series of usability tests, in which they observed spill responders and planners using a prototype to complete realistic tasks. Team members recorded problems encountered by test participants, as well as participants' suggestions and comments. The team then used these findings to improve the ADIOS II interface. The NOAA team believes that usability engineering has proved to be an effective tool for ADIOS II design.

The authors use a hypothetical spill incident 10 years in the future to examine the possible advances of spill response technology. The status of remote sensing at present, as well as its capabilities a decade hence, are discussed. The authors examine spill communication improvements, speculate on the use of the Internet to disseminate spill information, and examine electronic database systems for slick management. Progress in effectively using alternative cleanup strategies such as in situ burning and dispersants are reviewed, along with some of the likely impediments to their use in spills of 2011. Spill trajectory and behavior forecasting techniques of tomorrow are discussed in light of the expected continuing advance in computer technology. The authors review the likelihood that these new capabilities would actually be implemented. The resulting picture is a mixed one. Possible positive and negative scenarios are described.

No abstract