Box 4-1 shows some of the criteria that I believe are important in evaluating cooperative university-industry research programs. For example, new measurement and modeling capabilities should be a major output. DSSC has produced a number of these in areas such as finite element modeling and instrumentation, which spin-off companies have commercialized. Of course new technologies, quality and quantity of the graduate students are all important. But it is very important that we make this transition in the United States to a systems-oriented education for our students in order for graduates to be instantly employable and useful.

As for the future, I will give one final comment. Although the ERC program has been highly successful, there are only a limited number of these in the United States. One of our problems as a country is expanding that impact to a larger group of universities. Currently, there are about a dozen U.S. universities with ongoing research in the data storage area and NSIC has put in $1.5 million dollars to form a nationally coordinated program.