SAN MATEO (01/01/2000) - Having enjoyed a smooth transition into the year 2000, IT operations are girding for the resumption of business on Monday and are remaining watchful for date-related glitches.
The date rollover exceeded even the increasingly positive predictions that led up to the event. However, analyst forecasts have called for a relatively quiet holiday weekend, to be followed by potentially more than a year of sporadic problems.
"The issue must be managed throughout the year. Most organizations realized that and have plans in place," said Matt Hotle, vice president and Y2K research director at the Gartner Group, a consulting and market research company based in Stamford, Conn.
Estimates of the price of preparedness vary. While noting the impossibility of pinning an exact dollar amount on year-2000 expenditures, the Gartner Group estimated worldwide spending on software corrections at between $300 billion and $600 billion.
This amounts to a very expensive but -- many managers assert -- necessary insurance policy.
"The general consensus is that it's like insurance. It's a cost of doing business," said David Whitton, IT director at Mitsubishi Silicon America, a Salem, Ore.-based silicon materials supplier.
The direct economic impact of year-2000-related problems is estimated to be far lower than the cost of preparation.
International Data Corp. (IDC), a market research company based in Framingham, Mass., this weekend put the figure at $21 billion, down slightly from its earlier estimate of $23 billion.
Whatever the final tallies, businesses are readying systems for their routinely used, date-sensitive applications, such as the waves of business-cycle reporting.
Mitsubishi Silicon America plans to step up its testing over the next several days, according to Whitton.
"We have a pretty elaborate quality-control department, and we are really going to fall back on quality control to audit our procedures. We're going to be using quality-control principles and disciplines, and use auditing and testing," Whitton said. "It will be ongoing. It's the same testing we've had, but with a higher degree of sampling. For the first 48 hours [of 2000], we are going to look at everything -- every report, every label, and every secondary and tertiary piece of documentation."
As for the durability and reliability of the various remediation methods in use, Whitton said most, if done properly, should leave organizations in the clear.
"In my opinion, obviously, code changes -- replacing applications with all new systems, are probably the most durable, and then you have people who modified code, which is probably pretty durable, if they went in and found everything.
And, if someone used windowing, it should be effective for at least long enough to [run applications safely until they can be replaced]," Whitton said. "As for durability, I haven't seen anything that's just a patch that would work for only a short time."
The arrival of the new year went off without incident for the University of Toronto, according to Norman Housley, manager of network design and services.
"Things appear to be fine. We did all the things the application provider and systems providers said to do and at the moment things look OK. But, I kind of think we'll be more confident once things really start to roll and we're doing the normal day-to-day work," Housley said.
"Our business applications are not really being tried today, so the proof of the pudding is a couple of days away. Probably by Wednesday, we'll be able to say things are working normally or not. People expect there will be a lot of irksome tings, but not a lot of serious things," Housley added.
The University kept its roughly 55,000-account e-mail system and approximately 20,000-account dial-up network access service running through the date change and without incident, according to Housley.