So EST will be 19:00 for effect start on 12/31/1999, not midnight.

-- ..- (dit@dot.dash), November 01, 1999

Answers

Nineteen o'clock?

LOLOLOL!!

(Maybe you meant nineteen hundred hours, or seven o'clock pm)

-- (sorry@couldn't. resist), November 01, 1999.

If this is so, what is the Zulu time or GMT (what ever civilians call it these days) at the International date line (or Guam) where they are suppose start watching it for "signs" of the times. Understand CNN is doing the round-the-world watch same as CBS and some of the others are doing. By the way, puts us on the West coast at 1600 (4:00pm) - maybe in time to prevent some people from going to New Year's bashes if TSHTF then.

-- Valkyrie (anon@please.xnet), November 01, 1999.

We've been telling people to be offgrid by 3:00 PM. That gives them time to start a fire in the fireplace, get out the battery radio, settle down, etc.

-- bw (home@puget.sound), November 01, 1999.

"By the way, puts us on the West coast at 1600 (4:00pm) - maybe in time to prevent some people from going to New Year's bashes if TSHTF then."

Well at least there will still be daylight here.

Dang, right before supper though. Hate to have my dinner disturbed! :o)

-- Brian (imager@home.com), November 01, 1999.

The grid is not tied to GMT. It dont work that way.

-- The Engineer (The Engineer@tech.com), November 01, 1999.

Engineer, I'm not qualified to say whether the grid is or is not using GMT in its various controllers and switches. I've read many opinions that it is, but can't double check the qualifications of each speaker.

Doesn't really matter. GMT controls perhaps millions of devices, and it's probably safe to say that if a device is not set to local time, the overwhelming odds are that it's set to GMT. So midnight of 12/31/1999 GMT is likely to mark SOME disruptions, whatever local time that translates to anywhere on the globe.

So we tell people to be offgrid before GMT midnight, not because we know that our local grid uses GMT (don't have a clue) but because that's what Y2k preps are all about. Minimize unnecessary risks, sit back, watch the developments.

-- bw (home@puget.sound), November 01, 1999.

bw, it does matter.

You say: Engineer, I'm not qualified to say whether the grid is or is not using GMT in its various controllers and switches. I've read many opinions that it is, but can't double check the qualifications of each speaker. 

Then you go on to say that GMT controls millions of devices. Can you name me a few?

You give advice to others on what to do based on your (total) misunderstanding of how things work.

There is no time what so ever in any of the switches. Zippo, nada, none, zilch. As for the controllers  depending on what you mean some may have a clock function for recording info but the clock doesnt control anything.

How do you know that being off the gird wont increase the risk, perhaps from a fire or electrical shock, rather then decrease it?

-- The Engineer (The Engineer@tech.com), November 01, 1999.

Engineer - Nope, can't (won't, actually) name you a single one. We found some in our inventory where I work, but don't have the names handy. And I post using a pen-name to keep my employer out of this, so I'm not likely to be more specific. I could give a descriptive label for the device, but the cat would be out of the bag then.

I advise people how to be safe, given that we don't know what's going to happen, given that nobody understands how our infrastructure fabric is going to respond to Y2k.

My misunderstanding isn't total, it's selective. I'm not a well-rounded fool. What I DO understand is large scale mainframe database systems, and PC OO programming, because that's where my career has focussed for over 25 years. I've been doing Y2k remediation since 1994, and been writing Y2k compliant code since about 1986. I know LOTS about how that stuff works (and what I know makes me REAL nervous), so my advice gets more specific in those areas.

"There is no time what so ever in any of the switches. Zippo, nada, none, zilch." Flat statements have rarely impressed me. You MIGHT be correct, but you might not. We have reports (on this forum and elsewhere) of devices at the bottom of drilling rigs, of devices without documentation, of manufacturers long gone out of business, of two or three different chips from different sources doing a single job under a single part number. (Sorry, don't have the part numbers for you, either.) And you are sure enough to say "Zippo, nada, none, zilch"? Thanks for your input.

I repeat: it doesn't matter. The highest risk of time-related disruptions comes when the clock rolls from 99 to 00. That happens at two times for every place on earth: midnight local time, and midnight GMT. Whether the number of time-related functions in embedded devices is low or high, those that will fail due to time-related functions are most likely to fail at one of those two times.

So minimize the impact that failure can have on you, just to be safe. If people believe you, and you are wrong, they can get badly hurt. If they believe me, and I am wrong, they spend the evening in front of the fireplace. There is a risk in that, too, as you point out.

In a completely separate arena are the big-mainframe failures, which are likely to show up in following days and weeks. Chips with date problems are going to give a short, sharp impact. Mainframe dates will bleed us to death. Want to talk about dates in indices and sort keys? Now THAT'S exciting!

-- bw (home@puget.sound), November 01, 1999.

Hello BW. I liked your answer. I work at a major university. I study Information Management. In the elevator this afternoon I asked a COBOL professor from the Computer Science Dept. what she thought about y2k. She did not say much, in fact she gave me a dirty look. I'm interested to see how bad you really think this will really be. Thanks

-- jb smith (joebobsmith@yahoo.com), November 01, 1999.

1) GMT no longer exists. The new (internationally politically correct) name is Universal Time Coordinated (UTC). Still Zulu.

2) Many systems (UNIX) run under UTC and convert to local. I personally am concerned about power (but not on Kauai), and civilian telecommunications the most.

-- Mad Monk (madmonk@hawaiian.net), November 02, 1999.

That 'Engineer' sure is an arrogant bastard. I find him somewhat irritating. Reminds me of a fellow I used to work for. He tried to BS his way through a large thunderhead in a Cessna 182. The thunderstorm didn't care how smart The Engineer (oops, I mean this fellow) thought he was. It crumpled that plane into a little ball, and slammed it into the ground like a large hailstone. They had to scrape up what was left of him with a stick and a spoon.

Godspeed,

-- Pinkrock (aphotonboy@aol.com), November 02, 1999.

BTW, Engineer, what do you use to synchronize your frequency? Is it time? What sort of units do you use?

-- Pinkrock (aphotonboy@aol.com), November 02, 1999.

The frequency is synched to time. However not the way or as continually as you think. One utility is designated the official time standard and you synch to them. We use to use our own atomic clock but have since gone to GPS. However the time code is not constantly needed. The clocks are so accurate that once you have them synched and up and running if you lose the (satellite) time signal they are accurate to microseconds after several weeks. We need this when we do end-to-end line testing where you need the clocks in an area that the signals cant reach. Deep inside of a plant for example.

The time only has to be corrected if it is off by more then two seconds. Usually it is done more often then that but that is what the rule states.

We could synch to the standard clocks in Virginia or Colorado if we had to.

-- The Engineer (The Engineer@tech.com), November 02, 1999.