Y2K - is K.I.S.S. solution too simple?

A friend of mine has an old computer (486) that he doesn’t want to upgrade. (He just does word processing and a little database management on it.) He had some Y2K concerns, so he had me look at it.

The database program (that I wrote) worked OK. It accepts 4-digit years, but defaults to 1900-dates if you only use 2 digits. I also changed the date to after 1/1/2000, and the program worked well.

The real problem was when I set the PC to 12-31-1999 at 23:59. When the clock rolled over a minute later, it said Jan. 4, 1980.

I told my friend to just adjust his clock New Year’s Day to the right date, and everything should countinue working well. But I have since learned that this problem has been noticed before, and the fix is to replace the clock chip.

The computer held the date when I chaged it specifically. It rolled from 1-1-2000 to 1-2-2000 and other 2000-dates as hoped. But will the date fail later on?

Is my solution too simple, or is it just right?

It looks like you’ve done some good Y2K testing and found a workaround. One extra step you should take it to check for leap day 2000. Some non-compliant systems won’t treat 2000 as a leap year or misinterpret it as 1900 which was not a leap year. Set the clock to 02-28-2000 23:59 and see if it rolls over to 02-29 as it should. Proper testing goes beyond that but for a word processing machine that might keep it useful for a while longer.

Well, here’s a follow-up question that I can’t believe I haven’t seen asked before:

Was there, during the 1900’s, a leap year that started on the same day of the week as the year 2000? If so, a lot of problems afflicting non-Y2K-compliant computers can be “worked around” at least long enough to sort things out.

(Do I have to explain this in more detail?)


Are you educated, erudite and maybe a bit eccentric?
Please help us test a new web game!

I scibbled it on my VCR manual, if I could find it, but I think the solution for VCRs was to use 1976.

***Timothy Campbell: Well, here’s a follow-up question that I can’t believe I haven’t seen asked before:
Was there, during the 1900’s, a leap year that started on the same day of the week as the year 2000? ***

Yes. 1972, 1944, and 1916. Every 28 years, the year starts on the same day of the week. And being a multiple of 4, years 28 years apart are either both leap or non-leap. Exception: if the 28 years spans a non-400-multiple century year.

***TC: If so, a lot of problems afflicting non-Y2K-compliant computers can be “worked around” at least long enough to sort things out.

(Do I have to explain this in more detail?)***

Um, yes. How does 1972 being the same as 2000 help?

Well it gets the VCR straight, anyway. The numbers and weekdays are right.