Originally created 02/08/97

$2.3 billion estimate to fix computers draws scoff



WASHINGTON - For the first time, the government has put a price tag on fixing a turn-of-the-millenium problem that could cripple its computers: $2.3 billion. It says the crisis can be averted in the 35 months left.

But the technology industry calls that evaluation optimistic and the cost estimate "ridiculously low" - evidence that Washington is responding too casually to a looming information age disaster.

"The cost estimate does not pass the laugh test," Harris Miller, president of the 11,000-member Information Technology Association of America said Friday.

"They've created the impression there is a relatively cheap and easy fix that can be done with existing funding. There no real effort to put in additional resources to solve the problem."

The problem is that most government computer software reads the last two digits of a date. When the year 2000 is entered into computers, they will malfunction "unless they are fixed or replaced," the Office of Management and Budget said in a report ordered by Congress.

"They will reject legitimate entries or they will compute erroneous results or they will simply not run," the report said. "The potential impact on federal programs if this problem is not corrected is substantial and potentially very serious."

The OMB dismissed the suggestion that a single piece of computer software might be developed to correct the problem.

"There can and will not be a single solution," the report said. "Solving this problem requires technicians and engineers to write or revise software code and to replace hardware. A `silver bullet' is a logical impossibility. There is only a need for hard work, strategically directed, and plenty of it."

Still, the 11-page report reached an optimistic conclusion: "The new CIOs (chief information officers in every agency) are working hard to accelerate agency activities to address this challenge, and we are confident that the problem will be solved without disruption of federal programs."

Gartner Group, a Connecticut consulting firm, last year estimated that $30 billion would be required - $1 for every line of computer code that must be examined. Miller said he lacked enough information for an independent estimate.

"But everything we know from private industry and outside analysis says that the government number is ridiculously low," Miller said.

"I'm still hopeful," he said. "There still is time. It's going to take leadership from the top. It may mean the president and vice president and Cabinet secretaries must get involved. Those companies which have solved the problem or are close to solving it are those where the leadership has come from the top."

He said the OMB report suggested that only the Defense Department had made a detailed analysis. The Pentagon's estimated cost is nearly $1 billion. The OMB said the government would spend $529 million on the problem in the current fiscal year.

"Awareness is only the first stage, we've found," Miller said. "Commitment is the second stage. Candidly, we haven't seen the commitment yet."

Criticism came, too, from Rep. Stephen Horn, R-Calif., chairman of the House Science subcommittee on technology. He said the report showed the government "may gravely understate the costs and difficulties of averting an electronic disaster."

Miller said he feared that inadequate time to test solutions will be allowed. "In most cases, these big complex programs interact with other computers and other programs," he said. "You have to make sure that happens."

The OMB's $2.3 billion estimate covers only the costs of identifying necessary changes, evaluating the cost effectiveness of fixing or scrapping computer systems, making changes, testing the systems and contingencies if failures occur. It does not cover the cost of fixing state and local computer systems for welfare, food stamps and unemployment programs that Washington is obligated to underwrite or share.