Year 2000: real crash/virtual crash?

Philip Greenspun's Homepage : Philip Greenspun's Homepage Discussion Forums : Ask Philip : One Thread
Notify me of new responses
Given the mixture of rumour and hokum surrounding talk of the 'year 2000 bug', and the fact that no-one really knows what will happen when those clocks tick over into the next century, do you think that there _will_ be a millenial meltdown? Or even that fear of the potential costs of the bug may cause runs on the stock market long before the end of 1999? Especially since enough people will be watching the activity of the tech stockholders like hawks to see if they jump ship early.

It's an interesting thought-experiment, at least: the Y2K bug may be a more effective prophecy of doom than Nostradamus's for 1999...

-- Nick Sweeney, December 7, 1997

Answers

My favorite cartoon on this subject is pinned on someone's office here. The Deep Blue team at IBM have gotten a message from Gary Kasparov. He says he'll be happy to play a rematch but insists it start on December 31, 1999.

Practically the whole of humanity is working to maintain business computer systems right now. I can't imagine how Y2K bugs will make things that much worse. Especially given that almost all larger companies are converting to enterprise software systems such as SAP or Oracle Financials (see http://photo.net/wtr/software-industry.html for my take on how the Web world will do the same thing eventually). Also, remember that Y2K bugs will in fact appear gradually. For example, people taking on-line orders now are already having to deal with credit cards that expire 05/00 or 03/01.

-- Philip Greenspun, December 14, 1997


You asked Phillip, not me, but I don't know why the 2000 factor would affect tech stocks in a negative way. It's causing people to invest in new applications software, which probably trickles down to new systems software and hardware. Certainly it's been a boon to the software consulting industry already.

Mostly I think a lot of people's accounting will get screwed up, bills and checks will be sent out wrong, or not at all, that sort of thing. A few businesses will probably manage the whole thing badly enough that it'll actually affect their earnings significantly. I don't think most real-time systems that control critical things like air traffic control, power stations, etc, particularly care what the year is, but just to be safe, I don't plan to be on an airplane late in the evening of 12/12/1999.

-- Mark Ciccarello, December 9, 1997


You mention Oracle Financials in your reply Philip. While it's true that the very latest versions of this software are "Year 2000 compliant", the versions that most people run (prior to 10.6) are not. Personally I'm quite happy about that, as I'm a contractor helping implement/upgrade such systems - lots and lots of work out there for the next couple of years.

-- Steve Graham, December 17, 1997

Also remember that Unix stores the date as the number of seconds since the epoch (January 1, 1970) so somewhere around 2040, this number will become too big for the variable it's stored in (a 32-bit long int? I don't know) and break.

-- Ken Oeser, February 15, 1998

Most of the panic-stricken emails and web pages I see on this subject eventually boil it all down to one thing: the power grid. Supposedly everything from Nuclear Power facilities to the meter on your house are riddled with embedded microchips, and many of these chips are supposedly not Y2K compliant. Can anyone point me to any *hard facts* about this? I have a hard time figuring out why *any* part of the power grid would check to see what day it was to decide whether or not to shut down....it's not like we Americans have occasional power holidays!

-- Glenn Dixon, August 5, 1998

A late addition to my earlier question: I read the other day that the major UK tour operators have agreed not to charter flights for 31/12/99 or 01/01/00 because of worries over air traffic and the like. Now, given that there are likely to be a larger number of people than usual in the air around these days, in search of Pacific islands etc. close to the first seconds of 2000.... hmm. I still think it's Chicken Licken territory, though.

-- Nick Sweeney, August 14, 1998

Dan Lyke has an informed rant about the Y2k problem (especially regarding embedded systems at Reasons I'm not worried about Y2K.

There's also a number of LuseNet boards over at Mark Dalrymple, August 19, 1998


It's been my experience that the more people worry about some possible "doomsday" scenario, the less likely it it to happen. At the rate we are going, this one should be an incredible yawner.

My sole preparation for 2000 will be a bottle of champagne -- which I've already bought, just in case my local liquor store's cash registers stop working. ;-)

-- Frank Wortner, June 10, 1999


My bottle of Baron Philippe De Rothschild, Mouton-Cadet, Bordeaux Blanc from 1982 may not be year 2,000 compliant but I'll bet it's going to be very compliant with my tongue. So who cares if the lights go out (snicker) ;o)

(but then again, it just may turn out to be really great salad dressing in which case I better get some candles so I can see the salad while I'm eating it....)

-- Marika Buchberger, July 5, 1999


Well... I guess this question has now been answered.

-- Matthew Durell, January 9, 2000