Stanford University

News Service



CONTACT: Stanford University News Service (650) 723-2558

Stanford computing: productivity change is also slow here

STANFORD -- Timothy Bresnahan is on his 17th computer operating system in a 25-year academic career.

"People who work in universities, where the computing environment is catch as catch can, have worked on every imaginable kind of computer," says the economics professor who studies computer use in bureaucracies.

The lack of a centralized information technology purchase plan is just one of the ways in which his own workplace is different from those he studies. But like a number of those organizations, Stanford is attempting a major overhaul of the way it handles administrative computing and President Gerhard Casper has appointed a commission to look at how information technology will change two parts of the basic product ­ teaching and learning. Bresnahan is not completely convinced such a strategy will work.

"Teaching has never been subject to much technical progress in the past," he said. That doesn't mean it won't be in the future, but in his view, history suggests predictions in that area should be put in the category of "I'll believe it when I see it."

Stanford economist Nathan Rosenberg's work on the history of technical change has shown that "early guesses about how technology will be used are not completely wrong," Bresnahan said. "But they can be completely wrong in the sense that if you followed them, you would lose a fantastic amount of money.

"Ward Hanson at the business school has just written a paper that compares radio in the '20s to the World Wide Web today. People were not thinking that broadcast radio was going to be commercial. They thought it would be an artistic medium with lots of interesting things, and one of the things it was supposed to do was reduce the number of universities we needed. Ex post, you can recognize an element of truth in what people said then, but if you took it at all literally, they were completely wrong."

In the administrative area of Stanford, Bresnahan expects to see more change. "I don't know the details because I've managed to dodge that faculty advisory committee, but I support their effort because we have problems. What a hellacious mess it is to figure out where you are budgetarily at the end of the year."

In his research, Bresnahan has found a number of companies who don't want to invent their own administrative computer systems. A German company called SAP has become quite successful selling off-the-shelf bookkeeping and other business systems, which require the companies who buy them to handle information in the same way. In exchange for that, Bresnahan said, they get better support from the seller, they avoid the expense of creating their own systems, and employees don't have to relearn as much when they change jobs or employers.

But Stanford is probably among those large, white-collar bureaucracies that are not able to use many off-the-shelf administrative systems, he said.

"Many of our administrative systems are not appropriate for using standard business software because we're non-profit or because we have a peculiar organizational structure compared to a company. It's not hierarchical. There are centers that are different from departments which are different from programs, and we have different revenue streams."

Yet, he said, "there is probably too little sharing in general, which is one of the reasons change is going so slowly."

While he thinks the budgeting and payroll computing systems at Stanford could be improved, Bresnahan said that "there are some things very specific to our environment which are terrific. Axess, the computer program that studen