Why not just subtract 10 from all dates?

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

X - 10 beats Y2k! Why wouldn't this work? I call it the Jakielski solution.

It's relatively easy to identify all of your date fields in your files. Certainly this is one of the preliminary steps in Y2k fixing. Through utilities like SQL or programs you can subtract 10 from the year digits, buying us 10 more years. How many businesses or process systems have meaningful records from 1900-1909 that would get lost?

What we need is a data processing holiday(s) to do this. Then maybe the only crisis is in Times Square on 12/31/99, I mean 12/31/89. We would get to live the nineties over again, maybe for the better.

What I don't know is howand whether embedded chips can be set back.

-- Rich Jakielski (rich.jakielski@mapics.com), May 22, 1998


Well, to start with the calendar would be all wrong! 28 years is better, except that MSDOS and similar thinks the world started in 1980 so you can't set the clock to 1972.

Then you'd have to find absolutely EVERY date and subtract ten. Don't forget things like filestore dates, which get used by the backup program.

Then you'd have to find every place where a date is output and fix that to add ten back (or explain the wrong date to all your staff, your customers, their lawyers, and probably worst of all your lawyers.

Then you'd have to cope with the consequences of all the dates that got missed, getting ten-year-wrong data into your database.

And you'd have to work out what to do about the lady born in '09 who you pay a pension to and won't when she hasn't yet been born.

etc. etc.


Clock setback may be a limited solution, especially for self-contained systems that don't exchange date data with other systems. A traffic light controller that thinks it's 1972 or may be quite acceptable. The IBM dinosaurs running flight control might just about be given another 28 years by making 1/1/2000 equal "Air date 010172" or suchlike.

But in general ... THERE ARE NO MAGIC BULLETS. There's no answer except the hard one - fixing those bugs, one by one.

-- Nigel Arnot (nra@maxwell.ph.kcl.ac.uk), May 22, 1998.

When I explain the Year 2000 problem to a non-specialist for the first time, this is almost always the first reaction. Even before I complete my explanation, someone will say :

"Why don't we just change the calendar."

Suggestions vary from voting by decree that the year 1999 will have 99 months to regressing the time back to 1900. While time regression may work for some independant electronic devices, the year 2000 problem is a world wide global problem. Almost everything in our world is now interconnected. This kind of solution would have to be applied by everyone everywhere at the same time. This is impossible for the simple reason that adapting all our systems to a new calendar would require MORE adaption work not less.


Daniel Cormier dcormier@parousie.com Y2K Water Discussion Moderator y2k-water@parousie.com

-- Daniel Cormier (dcormier@parousie.com), May 23, 1998.

Couldn't we get Michael J. Fox to make a new film, "Forward to the Past" where he would just go back to the seventies and convince the offending management of the day that they were (are) making a mistake?

-- Carl Chaplin (chaplin@lillonet.org), May 24, 1998.

Moderation questions? read the FAQ