This detailed report was prepared by the Congressional Research Service of the Library of Congress. The CRS serves Congress. When a Congressman or a Senator asks the CRS to report on a topic, the CRS produces a report. This is its y2k report.
The report says that more than mainframe computers are at risk (p. 2). The y2k problem applies to mainframes, micros, minis, and networks. It also applies to "telecommunications systems" (p. 2). The report warns that fixing the problem is "a very complex management task" (p. 3).
Other problems include "the plethora of computer languages in existence today, the lack of source code and documentation for older software, and the shortage of programmers with skills in older languages" (p. 3). In short, there is no quick fix, no "silver bullet."
Then there is testing. "Testing is particularly laborious because the modified software must be tested in conjunction with all possible combinations of other software programs it interacts with to ensure functioning has not changed" (p. 4).
It lists problem areas: Social Security, the IRS, the military, the Federal Aviation Administration, state and local systems, banking, telephones, credit cards, drugs, medical records, and business. This is a comprehensive list (pp. 3-4).
Here is the big one, in my view: the Achilles heel. "Other systems that are not year-2000 compliant could send file information into the corrected databases, corrupting these databases. Flawed data can easily enter from the private sector into government agencies' database, and from foreign countries into US computer system[s]" (p. 5).
In short, if you don't fix all of them in a way that allows all of them to interact, none of them is safe. You must fix everything to be sure of anything. But it's impossible to fix everything.
What about the Department of Defense? ". . . for many DOD systems the work has not yet begun" (p. 5). The DOD has let every service solve its own problem (p. 5). There is no formal chain of command.
The absence of a chain of command afflicts private industry. Remember, this is a worldwide problem. There is no coordinator. The report says: "Most industry groups will need to make coordinated efforts to convert their software so that they can continue to interoperate as they do today. The securities industry, for example. . . . Other industries that must coordinate their year 2000 efforts include banks, insurance companies, telecommunications providers, computer manufacturers, and airlines" (p. 7). (The author left out railroads.)
"Because the United States is more heavily dependent on computers than other nations, the year-2000 is probably a greater challenge here than anywhere else. The economic impacts of businesses and government failing to correct the problem, both domestically and internationally, could be dramatic" (p. 10). Dramatic, indeed!