Re: What happened with the Y2K Problem?

Haha! I recall being told I gave one of the better explanations of the 
whole Y2K concept here on this forum. That was before I got married and 
changed my name, though...

Anyways, I think what happened with the Y2K thing is that it got way 
overblown out of proportion, what with all the computer consultants and 
companies trying to make some easy money, especially here in Seattle. It 
really wasn't that difficult of a problem at all, coding-wise. I've seen 
solutions in C, C++, Perl, Pascal, PL/I, Java, FORTRAN, LISP, and COBOL
(!), and none of them had solutions requiring more than 50 lines of code 
to fix a typical Y2K problem.

Many of the major companies I spoke to about the Y2K thing (this was 
before I answered the question the first time around) had already 
finished whatever upgrades were needed long before the moment arrived. At 
least two of the companies I had spoken with told me it was nothing more 
than prefixing all year codes with '19'. One company told me they had 
never needed a Y2K fix as a number of their data records required year 
codes prefixed with '18' and even '17', so they had taken that into 
account before they even coded their system. Another company told me that 
some time ago one of their programmers had decided to code date values as 
a Julian date, which meant the Y2K problem didn't even exist for them. 
Another company used the standard UNIX time() value for the date, which 
was even better than the Julian date solution because it not only got the 
date right, but the time of the transaction down to the second.

I think that computer technology advanced quickly enough to easily 
circumvent the Y2K problem. In particular, hard disk storage capacity, 
which was essentially the root of the whole problem. Two extra bytes per 
record for 100,000 records would be 200KB. In 1980, your typical hard 
disk drum was 5 or 6 LP-record-sized platters on a spindle, and about as 
large as a birthday cake. One of these could store up to 10MB. See the 
problem two extra bytes could cause if enough records were created? 
Especially if some system designer wanted to remove those two 'redundant' 
bytes?

I saw some of those specials on the Y2K problem. I thought the ones 
arguing religious significance were especially hilarious, since it 
doesn't take a lot of research to find out that when Pope Gregory I 
revised the calendar, the dates October 5, 1582 through October 14, 1582 
(inclusive) were deemed to be non-existent in order to restore  the 
calendar to accuracy. So much for that argument, eh?

Hope this helps you out!