Posted on 08/11/2018 12:29:20 PM PDT by SamAdams76
I've been doing control systems of all sorts since 1972.
I read that when 1000 AD was to dawn, it was the Millennium and everybody would go to heaven or be destroyed. People burned their houses and possessions or otherwise prepared themselves and went up on hilltops to await events.
When 12:01 am arrived, they all went home to whatever was left.
TG used the word "problem", not "catastrophe".
Vidka? Sounds like somebody hit the sauce a little early!
However, that is an intriguing idea to inject alcohol into seedless watermelon. Come to find out, there are tons of videos on YouTube describing exactly how.
big scare push by Gary North.
Cobol developers - the dinosaurs - were back and we were hungry.
Being a cobol “tester”, my best earning years were 97 thru 99.
Life was good.
I was the head of it at two companies at the time. It was my job to do Y2K remediation. What frightened me was that when I ask the power company to give me proof that they had done a remediation effort to avoid the bug they said they did the best we can but we can’t guarantee anything basically. That got me worried.
Instead we got a somewhat subdued celebration as lots of people decided to hunker down rather than go out and celebrate.
So ironically it was the nerds that saved the world in 2000, but people still didn't trust them enough to go out and celebrate the non-nerdy millennial.
Ummm. I’ve been writing control systems and designing such hardware since 1972. Some of that equipment is medical apparatus. Control systems do not generally care about date, only time intervals. I can think of no instance where any control system would be impacted by date. My systems sometimes kept date, but only to timestamp entries in a log file. Even as far back as 1972 we wouldn’t have done the 1900-to-2000 screw-up. Business programmers, sure. Control systems programmers, no. Any control system that keeps the date keeps it in the number of seconds past some epoch. I based my early systems from 1970, which gives perfect behavior until 2108 assuming a 32-bit time. Every ten years I would change the base date. My systems are now good until 2138. Control systems are usually obsolete and replaced long before their date bases expire. Still, control timing isn’t relative to any date. It is relative only to the clock that rolls over every 653 seconds. No control loop cycle last anywhere near that long. The next major hurdle will be in the year 2400. I suspect the circuit foil will come off the circuit board before then. Non-volatile memories are good for only 20 years.
Killing people is a “problem”?
That is probably caused by the EEPROM contents of BIOS fading away. Those devices will retain programming for only twenty years.
“Y2K related problems are more severe for Embedded systems than computer based systems as
replacement of embedded system is sometimes not possible due to high cost, obsolescence, inaccessibility,
interruption in process etc. Test or simulation of embedded system is not always possible, as they may be
inaccessible, for example, may be due to radiation hazard or chemical hazard etc.
Most of the embedded controllers utilize two digits for year representation. “
Impact of Y2K problem on Physical Protection System
“That is probably caused by the EEPROM contents of BIOS fading away. Those devices will retain programming for only twenty years.”
Lots of 20+ year old cars on the road ...
All of those are written by hypertensive people back then who did not write software for control systems. I do write software for control systems, and have so for a very long time. A control system's view of time is in short intervals with no respect to date. If you can prove otherwise, please do.
Take an elevator, for example. The worst that would happen to an elevator that had the wrong date is that it would cause the elevator cars to take weekend parking stations during the week. But wait! Day of week can be computed infinitely without regard to date simply because the length of a week is always seven days of 24 hours each.
Hype from people who wanted to make money off of ignorant people. It worked, and it worked well.
“Of greater concern to the electric
power industry are embedded com-
puters—small electronic chips or
control devices. These chips are
used extensively in all parts of the
electric power industry including
generating plants, transmission lines,
distribution systems, and power
control systems.”
Investigating The impact of Y2K problems - Utilities
I was badly fooled. I distrust computers so thoroughly, I just assumed there would be minor issues like stuck elevators and broken ATMs. I didn’t want to even be out and about, so I hunkered down on the most exciting New Years Eve in my lifetime. The Turn of the Century. Not one single issue occurred. It was the greatest non-event in world history.
While you might be right, I’m a bit skeptical. With millions of programs being revised for Y2K, not a single burp occured. You can say it is because programmers took it seriously. But when has new software EVER been free of bugs? Never. There are always hiccups. That not a single hiccup took place from a single bug in a single fix for Y2K tells me that for the most part, nothing was going to happen. If we were facing serious problems for Y2K with the old 2 digit dates, then the program fixes would have had some bugs and there would have been a few minor failures. Instead we had none. No failures whatsoever. That flawless level of service has never occurred in the history of programming. That can only mean that the buggy fixes were a non-issue, which tells me the whole thing was a non-issue.
The probable result of not fixing anything would have been a bunch of thrown error codes in the error files while everything continued to work as was.
NO EMBEDDED PROGRAM USES DATE TO SAMPLE DATA OR PERFORM CONTROL LOOPS.
Any information to the contrary is hype, usually from those who could scare ignorant managers into spending lots of money to fix NOTHING.
That is because it was a fake emergency hyped by scammers who could talk ignorant managers into fixing issues that were not really critical in the first place. Ignorant managers were willing to spend the money simply because they felt vulnerable...better safe than sorry.
I worked for a division of a large corporation. Our division created custom semiconductors for imbedding in other equipment. Our operation depended upon the proper operation of about 600 pieces of processing equipment of which about 200 had been created by our operation and not purchased.
I had oversight of our Y2K effort. I had a list of about 250 Y2K issues. In the simplest cases we identified a non-compliant piece of software and it was merely necessary to update the version or substitute a compliant competetive product.
The most challenging issue was for an elaborate custom lithography system which, for reasons that I was never made aware of, had been designed not only to do the processing on the chips but also to track and schedule the products upon which it operated.
Had we waited until a failure was detected we would have been a month or two without product and would have affected the delivery of several billion dollars of finished products.
There was no delay due to Y2K because we identified the problems, assigned resources to solve them, and implemented the solutions.
My daughter joked about my assignment one time by pointing out that the older engineers and managers were assigned to the Y2K problems. I pointed out that this was because we knew intimately how our systems worked and were the most prepared to solve the problems.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.