After months of work and billions of dollars were spent worldwide to correct Y2K bugs in computer programs, few of the expected disasters materialized. But the work that was done to prepare computers for the new millennium wasn't wasted, according to two Cornell experts. On the other hand, they add, fixing the old problems might create new ones.
In the process of upgrading their software to handle 21st century dates, many businesses also upgraded to better software and can now begin to make better use of the Internet, said Fred Schneider, Cornell professor of computer science. Schneider specializes in the study of computer systems that are linked to one another and in computer security. He served as chair of the Committee on Information Systems Trustworthiness of the National Research Council and is the editor of the committee's final report, "Trust in Cyberspace."
"It's hard to know if spending money was a good thing because the risks associated with our growing dependence on computing systems are not known," Schneider said. Important glitches might actually have been avoided because of the massive effort businesses and government put in, he said, but institutions don't like to publicize their problems, so we'll probably never know the extent of the problems that were avoided or that might still be lurking.
As for the possibility that Y2K bugs will surface in the future, Schneider said, "We haven't gone through full cycles yet. There will be bugs, but there probably won't be catastrophes."
Some have said there will be problems on Feb. 29, because 2000 is a leap year and millennium years usually aren't leap years. Schneider dismissed that by saying, "Software's been crashing on leap years forever."
Meanwhile, he added, the Y2K work everyone has done will pay important dividends down the line. "The thing that isn't discussed in the press is that every once in a while it's a good thing to clean out your closet," he explained. "This forced people to replace some very old applications with applications that are interoperable and therefore can grow in a networked world in ways they couldn't grow before. Now companies are better able to get into e-commerce."
Schneider also points out a downside: that tinkering with old computer code to fix Y2K problems may have introduced new and different bugs -- almost inevitable in any complex programming job -- or even that bugs might have been introduced deliberately by some of the thousands of programmers hired for the fix.
"The most conspiratorial take," he said, "is that maybe some of the people rewriting systems took this as an opportunity to put in trapdoors. There are rumors of that, although nobody will confirm that it has occurred."
"Trapdoors" are features a programmer could write into a program that would allow someone to hack into the system from outside at a later date and make changes or commit vandalism. Schneider noted that a lot of rewriting was outsourced to foreign companies, including companies in India and the Soviet Union.
Alan McAdams, associate professor of managerial economics in Cornell's Johnson Graduate School of Management, agrees that the overall result of Y2K work will be positive, and he says the greatest benefits will come from the software updates businesses have made.
"It was absolutely vital that we took [Y2K] very seriously," McAdams said. "And in taking it seriously, we have probably accelerated change in the way businesses carry on their operation by a couple of years."
McAdams has served as a senior staff economist with the President's Council of Economic Advisers and has been a consultant to industry, government, the Ford Foundation and the Institute for Electrical and Electronics Engineers (IEEE). In October he chaired the joint IEEE-USA/Cornell "Workshop on the Future of the Telecommunications Infrastructure."
"One of the most important advances in the way businesses do business has been the creation of what's called the 'value web,' a way of sharing information among companies in a supply chain," McAdams said.
Most manufacturers now use software that monitors what's happening on the production line, keeps track of parts inventory and orders parts when needed. But it is no longer enough, he explains, for a company to know what's going on within itself. Managers must know the status of suppliers and, in turn, of the businesses they supply. Compaq, for instance, needs to know if Intel is having supply problems that will delay the delivery of chips that go into Compaq computers; retailers that buy from Compaq need to know what's happening on its production line, and production managers need to know how sales are going. The Y2K problem highlighted these dependencies, McAdams said, and new Internet-ready software makes this kind of sharing possible.
"I like the analogy of a spider web," he said. "The spider sits in the center and knows instantly if an insect lands anywhere on his web. Information lands on the value web with the same impact. What Y2K did was move it forward in time. One of the most important things to come to the fore was just how dependent businesses are on their suppliers."
Many businesses, he points out, realized that even though they had fixed their own Y2K problems, they would still be in trouble if suppliers hadn't done the same, and this forged closer ties between companies, which led to more sharing of information.
McAdams sees a downside as well, but not a conspiratorial one. When company A develops such a networking relationship with company B, he said, it may be locked into using just that supplier. While company C might offer a better deal or faster service, its management software might not be able to link to company A's.
Some companies are dealing with this problem, McAdams said, by creating "server farms," where companies working together can all obtain and use compatible software modules.
| Cornell Chronicle Front Page | | Table of Contents | | Cornell News Service Home Page |