Hey, this should be a simple prediction - will power work, will phone lines work etc. All we have to do is pre-test the infrastructure. Why hasn't anybody run any trial tests and published the results. For example, what happens if a local utility sets the clock to '00'. Could it be that 'they' really have tested, and the results are too disasterous to publish. Me thinks so. Its WAY to quiet, and the rollover dates are fast approaching.

N/A

-- n/a (look@thereality.com), January 12, 1999

Answers

I believe you're referring to the recent Frisco and LA power outages. They aren't totally ruthless. They'll wait until it warms up a little, and then we'll be having power outs galore.

-- (@@@.@), January 12, 1999.

When you're running a public network, you can't do a full-scale test. Even assuming you could simultabneously set ever clock in the network forward, that would just precipitate the crisis a year early without the benefit of whatever remediation and contingency planning can be done in the meantime.

Of course, they can and do test individual bits of equipment, and are busy right now replacing or upgrading the things which failed. The big worry is that individual tests, or tests in a building-sized "test" network, won't reflect reality. However, it is the best that can be done.

Results have been published. Go to any manufacturer of communications equipment's web page, and you'll find Y2K status reports for all current equipment and most past products. Power utilities have tested inividual power stations. I've seen two reports, both of which encountered trouble, but which in both cases was transient (ie if the operators knew in advance what was going to happen, they could have used manual overrides to keep the power station online).

At the end of the day, *nobody knows*. When you look at snow on a mountain, can you tell if it's an avalanche about to happen? When you look at a small tropical storm, can you say if it's a hurricane in the making or not? Neither is reliably predictable -- and neither is Y2K.

-- Nigel Arnot (nra@maxwell.ph.kcl.ac.uk), January 13, 1999.

"Its WAY to quiet, and the rollover dates are fast approaching."

Indeed it is, what, 250 working days, will be gone in *no time*.

-- Andy (2000EOD@prodigy.net), January 13, 1999.

No systemic tests have been run - no body (at any level) wants to risk "turning off" the state, city, or county long enough to find out.

Yeah - its stupid and shortsighted. What else do you expect from our "fearless leader" in Washington? You expect improvement now?

-- Robert A. Cook, PE (Kennesaw, GA) (cook.r@csaatl.com), January 13, 1999.

Actually it may not be stupid and shortsighted. The work involved in doing a large-scale test of the sort you envisage is almost certainly greater than the amount of work needed to do the remediation. Also if the test encounters problems, these may be of a nature that makes it impossible to get back to the pre-test state in a reasonable amount of time. In other words, trying to do a large-scale test would guarantee that things turn out worse both in the short and in the longer run.

Provided there is a full-scale search-and-remediate program going on at your local utilities, this is the best you can hope for. These big public networks are designed to be fault-tolerant; lots of things go wrong daily and you usually don't notice. Every bug removed before it activates (usually 1/1/2000) is one less thing to burden the network when that fateful day arrives. If they started early enough and fix anough in time, things may be OK-ish. If they started too late, nothing can be done about it now. If you know they're slacking, blow the whistle; otherwise, the die is already cast.

-- Nigel Arnot (nra@maxwell.ph.kcl.ac.uk), January 14, 1999.