WASHINGTON POST: "How Did It All Go Right?: Officials, Experts Happily Seek Answers to Y2K Riddle" - even the Washington Post is puzzled...:)greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread
How Did It All Go Right?
Officials, Experts Happily Seek Answers to Y2K Riddle
By Rajiv Chandrasekaran
Washington Post Staff Writer
Sunday, January 2, 2000; Page A01
Not so long ago, a Y2K computer catastrophe seemed inevitable. Few believed that the world's businesses, governments and technicians could band together to test hundreds of millions of computers and tens of billions of electronic devices, painstakingly fix any problems and then check them all again to ensure the repairs worked.
Even the most optimistic specialists predicted moderate disruptions: Some cities would certainly lose power. Phone systems might fail. Banking records could disappear.
But when clocks rolled into 2000, and officials from New Zealand to Hawaii checked their monitors, even the most sanguine forecasts seemed too dire, leading many to question why the much-feared Y2K glitch was such a dud--even in Russia, China, India and other nations seen as particularly vulnerable to failures.
So what went right?
Technology analysts offered a combination of answers yesterday. They credited the unusual cooperation among businesses and governments worldwide to address the issue. They cited the unprecedented mobilization of people, money and executive attention for the repair effort. And they wondered whether some of their previous assumptions--specifically about the technological dependence of less-developed countries--had been off the mark.
"We may have overstated the impact of technology on the infrastructure in a lot of developing countries," said Matt Hotle, a vice president at the Gartner Group, a technology consulting firm that predicted early last year that Y2K could create "significant disruptions" in such nations.
In Paraguay, for instance, where the government waited until mid-1999 to start tackling the glitch, the country's Y2K coordinator had predicted widespread power outages, water shutdowns and phone disruptions. But as of yesterday afternoon, all basic services, including electricity, telephones and water, were functioning normally.
"At this moment, everything is working," said Walter Schafer Paoli, the former coordinator. He attributed the poor, landlocked nation's success with Y2K to redoubled efforts to fix computers in the final weeks of 1999 and a discovery that many government services were less reliant on computers than initially thought.
"When they began to do the repairs, they found the problem was not as bad as they believed," Schafer said.
The same assessment was given yesterday by officials in Washington, who expressed surprise that there were no reports thus far of major disruptions in Russia and China. "I think the reason we're not seeing anything too serious there is that the systems in those countries were not highly vulnerable to the Y2K bug in the first place," said Bruce McConnell, the director of the International Y2K Cooperation Center, a United Nations-funded organization that has been closely monitoring date-related problems.
The civil infrastructure--the power, water and phone systems--in those countries often does not depend heavily on computers, he said. And in some cases where it does, foreign governments and businesses have been able to accomplish their repair work in a shorter time than their American counterparts because they opted for programming shortcuts that are less common in the United States, such as rolling a computer's internal clock back to 1979, because the days of the week match those of 1999.
With regard to the United States and other developed countries, industry executives and government officials cited the massive outlay of cash--estimated at $100 billion domestically and a half-trillion dollars worldwide--that was spent to hire programmers and buy new computers. They defended the spending yesterday as necessary to deal with the Herculean technology challenge.
"I don't know anyone who's spent any time on this problem at all who doubts that, had the effort not been made, had the money not been spent, we would be in a very different situation here right now," said White House Y2K czar John A. Koskinen.
In the District, which earmarked $140 million to wipe out Y2K bugs, city officials kept an old Department of Public Works computer application running into the new year--even though it had been replaced by a new system. Sure enough, just after midnight, it listed the year as 1900.
"The fact I think we haven't seen many problems was not because there really wasn't a Y2K issue," said Mayor Anthony A. Williams (D), who called the glitch "a once-in-a-lifetime event."
"It threatened the very foundation and infrastructure of the city, the basic delivery of services in the public and the private sector," Williams said. "It's very, very hard to say what cost is too high. You know no one's doing a cost-benefit analysis of Normandy. We had to do Y2K. It's on that order of magnitude."
Although much of the Y2K spending was devoted to simply making older computer systems operate normally this year--the electronic equivalent of patching a flat tire--some of the money went for new machines that likely will make businesses and government agencies more efficient. The repair effort also has given many organizations, for the first time, a census of all their computer systems and helped them weed out antiquated technology.
"We know ourselves technologically better than at any time before," said Rear Adm. Robert Willard, a Y2K coordinator at the Defense Department, which spent $3.6 billion on Y2K preparations.
After almost a full day to monitor computer systems, U.S. government officials yesterday reported a series of minor Y2K glitches scattered across the country as well as one potentially serious disruption: A military reconnaissance satellite system was inoperable for several hours last night.
The problem hit a ground-based computer system responsible for processing data from a network of intelligence satellites shortly after the rollover passed midnight Greenwich Mean Time (7 p.m. EST), which is the time standard for many satellite systems, Deputy Defense Secretary John Hamre said.
The processing station had been extensively tested, but not when the satellites were in the configuration they were in on New Year's Eve, defense officials said. Simulations did not pick up a problem that developed in the combination of the date rollover and positioning functions, the officials said.
By yesterday afternoon, the processing station was still not at "normal peacetime operation." Pentagon officials would not disclose the nature of the intelligence gathered by the satellites.
Other minor Y2K-related errors hit seven nuclear plants, including one in Arkansas where workers were denied entry through automatic doors because a Y2K software patch had not been installed on a radiation monitoring unit. None of the problems affected safety systems, officials said.
Wind-shear alert systems failed at airports in Tampa, Denver, Atlanta, Orlando, Chicago and St. Louis during the midnight rollover, but technicians were able to quickly re-start the systems to clear the error. Several electric utilities also reported glitches with clocks used to synchronize management systems. And in Omaha, a security access system malfunctioned at a federal building, freezing doors in the open position; to get the doors to shut, technicians reset the computer clock to 1972.
The absence of more serious disruptions pleased government officials. At a news conference yesterday afternoon, Koskinen gushed that the nation was in "much better shape than anyone would have predicted."
Still, Koskinen and others cautioned that additional problems probably will be discovered on Monday--and through the next several weeks--when companies and government agencies reopen for business. "It's far too early to declare victory," Koskinen said.
McConnell, of the international Y2K center, warned that organizations still could face some "inconveniences, headaches and hiccups over the next few days."
That was a message voiced even more strongly by people who believed the Y2K glitch would cause serious disruptions, even in the United States. Paula Gordon, a visiting research professor at George Washington University who earlier this year wrote a paper asserting that "Y2K is not a solvable problem," maintained yesterday that serious trouble could still occur over the next several months as hidden errors degrade computer systems.
"There will still be a major fallout," Gordon argued. "We should not be lulled into thinking this problem is over."
But other Y2K skeptics expressed surprise and relief yesterday about how smoothly the switch to the new millennium has gone, with no regrets about the hundreds of gallons of water they collected, along with the canned food, batteries and backup heaters.
In fact, some of the naysayers gave themselves credit for the glitchless Jan. 1. "If we had sat around and said nothing, we might all be sitting in the dark," said Burke resident Gail Fialkow, secretary of the Northern Virginia Y2K Community Action Group, which encouraged area residents to stockpile goods for the inevitable chaos it believed would occur when 2000 arrived.
Jay Golter, the group's leader, said that "in hindsight, some of the things I did were not necessary, but if you replay the tape, I'm not sure I'd do much different."
Golter, who said he was not surprised by how smoothly the transition went in the United States but had expected serious problems in other areas in the world, stocked more than 100 two-liter bottles of water and attached large barrels at the end of his four downspouts.
He now intends to use the water in his garden. And he's sure his family will use most of the canned food he bought, although he said he wouldn't have purchased a $600 wood stove if it weren't for his concerns about Y2K.
In a few weeks, his Y2K group will hold a potluck dinner. "Perhaps we'll have rice and beans"--commonly stockpiled fare--"or we might use that as an occasion to drop off the extra food at a food bank," he said. After all, he noted, his family is not fond of all the bags of dried beans he bought as a precautionary measure.
Staff writers Stephen Barr, John M. Berry, William Claiborne in Chicago, Caroline E. Mayer, Sylvia Moreno, Don Phillips, Michael D. Shear, Alan Sipress, R. Jeffrey Smith in Rome and Roberto Suro contributed to this report.
-- John Whitley (firstname.lastname@example.org), January 02, 2000
Now, THIS is ironic!!! Koskinen is essentially saying the same thing that we doomers are: It's great that we came through the first hours unscathed, but it will take WEEKS to really see how troublesome Y2K is going to be. Whereas the pollies, who have always been stooges for Koskinen et al, are trying to tell us that Y2K is over, done, finished, kaput!!
-- King of Spain (email@example.com), January 02, 2000.
Why computers did not crash!
greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread
The Computer programmers told us that on January 1st the computer date would go back to 1900 and that would confuse the computer and cause the systems to crash. They were wrong.
In Russia where they did not do any repairs, electricity stayed up. Why?.... Simple. Two days ago the date was 1999. The computer thinks today is 19100. What is the number that comes after 99? It is 100. Next year the date will be 19101. We may see that date on statements from businesses, but who cares? We know that is means 2000.
I checked several news websites and they all show the date as 19100. The programmers were wrong and I'm glad! Now we can go on with life as we were acustomed to. YIPPEEEEEEE!!!
HAPPY NEW YEAR "19100"!!!!
-- bbb (firstname.lastname@example.org), January 02, 2000.
The author of this piece is the guy who was so crochety when people accused him of not covering the story. Now he talks about how a Y2K computer catastrophe seemed inevitable. I don't believe he mentioned that before.
-- Mara (MaraWayne@aol.com), January 02, 2000.
Hold on! How can a computer program which is only alloted two digits for the date go from 1999 to 19100? Or even from 99 to 100, for that matter. I'm no computer whiz (far from it), so please explain in verysimple terms
-- jumpoff joe a.k.a. Al K. Lloyd (email@example.com), January 02, 2000.
I am a computer programmer and have seen a lot of terrible code. The quality of programmers is the same as any other profession. It varies from supurb to stupid. It is not reasonable that all the programmers hired to fix the Y2K bug were experts who could understand the junk code, repair, and test it in such a short time. My guess is that a lot of the code is still untested junk. Anytime a date is converted to a string using the internal short date format on a PC of "MM/DD/YY" and sent to a function that compares two dates, it should fail. Embedded chips have a similar problem and lots don't have practical tests. I am amazed that the doomers were wrong. I see a lot of criticism of people who used their knowledge to be prudent and protect their families with preps. I think the criticism is unwarranted. "How it went right?" God only knows.
-- Don Hopeful (Don@Hopeful.com), January 03, 2000.
Jumpoff Joe In answer to your question,it can't. There is a link somewhere on this site to an article which listed the date as 19100 instead of 2000. It does, however, demonstrate the type of junk code I was talking about in my earlier answer. For a little background; somebody came up with a gimic called "Windowing" to fix the Y2K problem. He even patented it and is now trying to collect millions for anyone that used it. The fact that it is severely flawed didn't seen to stop anyone in upper management so almost all programmers were forced to use it. What it does is give a date range for 100 years. Microsoft Windows defaults to the range 1930 to 2029. so the programmers uses a "Pivot Date" of 30. If the 2-digit year is 30 or greater it must be the 1900's and less than 30 the 2000's. The fellow that did the programming to get 19100 probably split the date into a 2-digit century and a 2-digit year. Then he added 99 + 1 and got 100. Then he changed the two values to strings for display and concatenated them to get a date. Assuming some pivot date, say 30, 100 is greater than 30 so the year must be "19" & "100" = "19100" The guy that did this is not a moron. It is a "normal" coding error. What he should have done is use the right 2 digits. A similar error might take "20" & "0" = "200" instead of "2000". I have seen this one coded. The point I was making earlier is that errors like these are a normal part of coding and won't fail until the clock rolls over AND the code is called AND someone is looking for this error. For Visual Basic the cost of testing software is prohibitive. It also requires such a steep learning curve that few people use it. So, I think we have a lot of bad code out there that should be failing. In my earlier example of using "MM/DD/YY" as stings, the standard functions may fail. 01/01/00 - 12/31/99 = about -100 years. Some of this has been reported. As far as the windowing fix for Y2K, that is trouble. Most of the pivot dates are being hard-coded. Let's examine using 30. Guess what happens when we get to 01/01/30? Since this is equal to our pivot year it must be 1930. So 01/01/30 - 12/31/29 = about -100 years if the code is perfect. We haven't fixed the problem, we have just pushed it out 30 years. Apparently there is no world-wide standards agency so every company in world came up with its own pivot year and soon data will be corrupted as it is passed between companies. These problems are pretty obvious to a software developer but we are not the ones making the decisions. I have been telling my bosses about Y2K problems since 1993. I don't think they will ever be fixed.
-- Don Hopeful (Don@Hopeful.com), January 03, 2000.