It is interesting to note that two of the problem dates for Y2K are April 9, 1999 (it is the 99th day of the year and may be seen by some systems as having special meaning.) and September 9, 1999.

The dates for the NERC nationwide tests are: April 9, 1999 and September 8, 1999.

Perhaps this is also a contingency plan as well as a test. It would be logical to place people in specific areas to prepare for problems that would arise on these dates, but it is dangerous to test for too many problems at once.

The problem I have with this is that it is usually harder to fix two problems at once. It would be horrible to have to try to fix three or more all at the same time. I would suggest that they test compliance on different dates so as to not confuse "99" issues with "00" issues and make a real mess.

I think that even a short discussion with a technical person would reveal that testing for Y2K on these dates would be a mistake.

-- Anonymous, February 01, 1999

Answers

See my comments in previous posts on this issue. I believe they are "testing" on these critical dates to give the false appearance of being "in control" of the systems should failures occur. As far as I know they are doing nothing more than letting the clocks run and seeing what happens.

-- Anonymous, February 01, 1999

OK, this is a test. The the answer posted in the response to this (reasonable) question is based on pure speculation with no foundation in fact. The answer given shows a fundamental misunderstanding of the nature of the April drill. There HAS been an answer given to refute the reactionary claims of the respondent (by none other than Rick Cowles), AND cites a source for the true nature of the April drill. If this is indeed a quest for truth and not a collective hissy fit, will anyone respond to the above, correct the error, and allay the fears of Troy?

HINT: try searching this site for ftp://www.nerc.com/pub/sys/all_updl/docs/y2k/drills.pdf , and you will find the thread that should have answered this, if you are really interested.

PS: Using Adobe, you can access the ftp in the hint and you will see what for yourself what NERC is REALLY trying to accomplish.

PSS: Sorry for the tone of this. I am beginning to get frustrated by the possiblity that no fact, no report, will result in the changing of a single opinion. Certainly, the admirable goals and prudent planning of NERC in regards the drills can be accepted by reasonable people?

-- Anonymous, February 01, 1999

CL,

Chill out baby-doll ;-) No need for the hints and all that. I downloaded the pdf file and did a quick cut and paste for the forum goers.

Heres the scoop straight from the horses mouth!

Proposal to Conduct Two Industry-Wide Y2k Drills

One component of operational preparedness of electric systems is the conduct of training and drills. NERC proposed in the September 1998 report to DOE and the public that the electric industry should conduct two Y2k drills on April 9, 1999 and September 9, 1999. These dates have been selected to coincide with two lesser priority Y2k operating dates (99 th day of 1999 and 9/9/99).

This paper provides a first step by defining the scope and objectives of the April 9, 1999 industry-wide Y2k drill. The NERC System Operator Subcommittee is requested to use this scope and objectives statement to develop by December 31, 1998, a more detailed implementation guide for the April 9, 1999 drill. This guide will be delivered to the NERC Regions for development and implementation of the drill at the regional and company levels. It is expected that all ten Regions and all bulk electric operating entities will participate in the April 1999 drill.

This initial drill will focus on personnel and communications. The drill will assume partial loss of voice and data communications and partial loss of EMS/ SCADA functionality. Operating entities and Security Coordinators will be required to identify key operating facilities and information requirements. Properly trained personnel will be sent to key locations and will be required to identify and communicate critical operating information over backup communications systems. The goal is to demonstrate the ability to operate electric systems with limited voice and data communications and EMS/SCADA functionality.

The second drill on September 9, 1999 is expected to be a dress rehearsal for the rollover from December 31, 1999 to January 1, 2000. This drill may include reducing planned outages, modified commitment of resources, redispatch of generation and transmission loading, cooperation with electric market participants, and staffing of all critical facilities. The goal would be to simulate system conditions and operating plans for the Y2k transition as closely as possible without increasing risks to personnel and equipment safety or system operating security.

Steve

-- Anonymous, February 01, 1999

I appreciate all of the responses. I did finally look at the NERC document and do have a genuine concern with the timing of the tests.

The document does admit that this coincides with "Y2k operating dates."

To further re-state my question(s),

How would it be prudent to test fixes on the same date that failures might occur? Wouldn't that increase the risk that the two may interfere with each other making the test results less conclusive?

-- Anonymous, February 01, 1999

CL & Others,

I apologize. There was a table in the pdf from NERC that I failed to include in my last post. I also apologize for not re-formatting my posting to make it more legible.

Anyway you would be best advised to visit the ftp site yourself and vew the document using Adobe's Acrobat Reader, which is free at many fine internet locations.

The following is the data from the table but please forgive the format. It looks better if you get it yourself. I do this for those who don't have, or cannot run Acrobat reader for whatever reason.

Electric Industry Y2k Drill 1  April 9, 1999

Scope * Industry-wide drill on operating electric systems with limited information and communications during Y2k rollover

Participating Entities * NERC and NERC Regions * Security Coordinators * Control Areas * Independent System Operators * Other Entities Operating Bulk Electric Generation or Transmission

Participating Personnel * System Operators * Security Coordinators * Substation and Field Personnel * Power Plant Operators * Computer and Technical Support Personnel

Drill Assumptions * Partial loss of primary voice communications systems * Partial loss of data communications systems * Partial loss of EMS, SCADA, and control system functionality * Possible bad or missing data in EMS, SCADA and control systems

Drill Objectives * Given the assumptions stated above: - Personnel are dispatched to key locations sufficient to allow system monitoring and operation. - On request, personnel identify, obtain and communicate key operating information sufficient to monitor and operate the system. - Personnel communicate using alternate (backup) communications systems.

Critical Communications Interfaces * Control area operators to plant operators * Control area operators to transmission operators * Control area/transmission operators to substation and field crews * Control area operators to other control area operators * Control area operators to security coordinator(s) * Security coordinators to other security coordinators * Other regional, interregional, and company critical communications interfaces

Duration * As necessary to meet drill objectives * Coordinate times as necessary with neighboring areas and regions

Preparations * Identify critical communications interfaces, operating facilities, and information requirements for operation with limited data communications and EMS/SCADA functionality and possible bad or missing data. * Prepare procedures/plans for operation with limited information and communications. * Identify and test backup communications systems * Train personnel * Coordinate resources to assure normal operations are not interrupted

Reporting * Report from operating entities to Regions and Regions to NERC by April 30 * Address nature of drill, objectives, success measures, and key lessons learned

Steve.

-- Anonymous, February 01, 1999

There's quite a few problematic dates during 1999. I think CL posted in an earlier response related to this topic that the desire might be to have people "on station", as it were, during these critical dates. That could very well be the case, I don't know.

If this were the case, though, wouldn't it be more prudent just to have more personnel on alert rather than stressing the system (and people) with simulations?

Hey, I dunno. It's just conjecture on my part - I'll try to contact someone from NERC directly this week and get a response. I'm sure there's a reason; these two dates were not picked randomly.

Footnote to CL: Troy is one of your bretheren - take a look at the bpa.gov email address. It's a good question; let's try to nail down a concrete answer.

-- Anonymous, February 01, 1999

Apologies - on re-read of CL's post, I realized the response wasn't to Troy's posting, but the first answer.

-- Anonymous, February 01, 1999

Very Good! The flaw in the response was that the objective was not to test rollover, but whether the system can be manually operated in a partial or total loss of voice and scada. This drill will answer many of your concerns I have seen posted.

Now, ask why a drill? To test whether our contingency plans have been designed to anticipate all problems and have the capacity to provide enough volume, speed and accuracy of communications to run the system. The objective is to 1. See if the contingency commo plan works to its design specs. 2. To stretch it beyond the design to find strengths and weaknesses. Strengths to share with other utilities, weaknesses to address before 12/31. IMPORTANT NOTE: Failures and shortcomings in the plan are not a BIG negative. This is diagnosing flaws in the contingincy plan and practicing emergency commo.

Now timing. There is a SLIGHT chance of Y2K related events on the April date. So, if you are going to have a drill - place the backup commo resources in place, is it better to do it on St. Patricks day or on a question mark date? To me, it makes perfect sense to mobilize forces on this date. (And yes, we will participate in whatever scope of drill NERC recommends).

Troy, I hope this answers your question. I am somewhat reassured that the members of this forum didn't let this particular response go unanswered. You see, you people here are not only learning the Y2K issue, you are impacting it. When you do not challange assertions of ill will and deceit on the part of the utilities and NERC, you actually SHAPE opinions. Imagine someone stumbling onto the above thread and learning that the NERC drills are designed to be the dark instrument of a cover-up. That person will now go to work, church, school and spread this error - CONTRIBUTING to the very panic that we are attempting to dispel by reasoned logic.

-- Anonymous, February 01, 1999

I do not mean to harp on this issue. The reason you see it raised so often is partly the mechanical nature of the treads in this forum. I find it difficult to keep track myself.

I am not saying it's not prudent for them to be on stand by, simulate communications failures, parade around in formation etc. on these key dates. However, let's suppose that a system failure does occur on one of those dates and it is in some embedded system that is not being affected by the tests, drills or whatever you want to call them. Will they say "Don't worry folks we are just doing some drills at the plant in preparation for Y2K. You have nothing to worry about here. We are in control of the situation." Or will they say "One of the embedded systems that we had not yet had a chance to find, test and fix just failed unpredictably and that's why the power went out today. But isn't it a good thing we coicidentally had our contingency team on alert and doing drills that day."

This is not conspiracy theory stuff. I am praising NERC for good crowd control.

Also, I am keeping an eye on the California PUC's investigation of the recent San Francisco blackout that was caused by "human error." Was it the fault of a really stupid construction worker or was it Y2K testing and remediation gone awry. The PUC wants to know. Here again "human error" gives the utility some level of control i.e. it is an isolated occurance and can be avoided. If they said it was a Y2K testing screw up (and shouldn't the utilities be testing their systems about now) would that not cause a panic.

It's plausible deniability.

-- Anonymous, February 02, 1999

Troy, Joseph, CL, Rick,

Thank you all for your vision on this issue. It made me think about our contingency plans here in the Netherlands. Is a y2k contingency plan for all of the critical dates or only for the rollover on M-night? That is the main reason I collect examples of failures in embedded systems at other critical dates. Till now I think that a y2k contingency plan has to be made for the rollover on M-night.

What to do on the other critical dates ? You can let a little bit extra people hanging around or you can use it for a kind of dress rehearsel for the rollover, also called a drill. In our company our first critical date is September 9, 1999 and parts of our contingency plan will be activated. I think that it is a good idea of NERC to do a drill on this critical date. The other critical date, 99th day of 1999, is not seen as a critical day in embedded systems in our view and is a little bit early within our project for drilling our personnel.

The next question is about external communication: What to say when a system failure does occur on a critical date? I think that indeed the whole world will think that it must be caused by the millennium-problem, but a system failure can always occur even on a critical date that is tested. We only have to be carefull with what to drill. It is a good thing to be extra alert on critical dates and have our contingency teams doing drills that day and that is also what utilities have to tell the people.

Be carefull with what to drill Take care.

-- Anonymous, February 02, 1999

CL,

When you do not challange assertions of overly optimistic statistics on the part of the utilities and NERC, you actually SHAPE opinions. Imagine someone stumbling into the NERC report and learning that "everything is going to be just fine". That person will now go to work, church, school and spread this error - CONTRIBUTING to the very mis-information that we are attempting to dispel by reasoned logic.

The following is from the IEE site. It explains how the same basic information on embedded systems can be presented in drastically different ways.

The Institution of Electrical Engineers W151512 Counting systems or components

Embedded systems vary widely in size and complexity. Those with potential for date problems will usually incorporate a number of components each of which may be itself be regarded as an embedded system. In relation to the business what matters is the system i.e. the plant, machine or piece of equipment. For the purposes of examining and testing it is necessary to identify each component of the system separately. The statistics obviously depend on which is used as the baseline. As a hypothetical example, the XYZ company operates 10 systems each of which it regards as business critical and each of those systems has 200 embedded system components. It might regard itself as having 10 embedded systems or 2000 embedded systems. If one component system fails in each of two systems the failure rate is either 2 out of 10, i.e. 20%, or else it is 2 out of 2000, i.e. 0.1%.

CL, I value your input but your basic premise seems to be everyone at this site would like to prove ill will on the part of the NERC. Not so.

The NERC is not outside, over or above the electric utility industry. They have their agenda. Presenting the best picture possible is within the scope of their interests. For Rick or anyone else to show the very same information in other ways does not constitute a conspiracy or unconscionable act of any kind. Pointing out errors or biases in methodology of analysis and reporting practises is VALUABLE! I see no one else looking over their shoulder.

Exaggerating one way is just as harmful as exaggerating the other way.

Steve

-- Anonymous, February 02, 1999

Some find it hard to follow the information written here sometimes but I guess you could say I am a "regular" kind of US citizen that would like to understand all of this. This will make you crazy but the original question is still unanswered in my view. Why test on a date that has a higher percentage of risk than another? I agree that being present on these days is, no less than smart, but why would you not check this before those dates too. You would have some problems found and will be ready for any new ones that could crop up. It's a plain simple question. It is reasonable and I WILL wonder why you can't answer it. Leave me with what you are offering and I will indeed NOT trust you. We are a simple people and if you can't come down to a simple level to explain it then don't expect any respect from me or a lot of other people. The impatience with "US DUMMIES" is getting old too.

-- Anonymous, February 02, 1999

Story time (my attempt to explain myself):

My furnace went out for an entire week. They repairman kept putting transformers in and they would blow out. He would replace a part and it would blow out again. He kept ordering parts and replacing them. After a week of trying, he had three people in there with meters and parts. Finally, they fixed it (last night; I feel so much better.) The owner of the company was very sorry and explained that nothing like this had every happened to him. Two key parts of the furnace were bad which was affecting another part (the transformer). He explained to me how hard it is to fix two problems.

Testing for the problems in isolation is best.

**However**, when testing contingencies, there's nothing like a little bit of a problem to make you really feel the need!

The other story:

Last week, I got to use my "No Heat" contingency plan (which I had never tested). I had wood for my fireplace and two electric space heaters. I thought to myself, "I will try to heat this house." Well, it didn't work. But, I survived without so much as a cold! We would sit arround the fire at night (kind of a neat thing actually) and we would change with a heater nearby. We would keep busy in the colder rooms. We learned which doors should be open and which should be shut. The result: feel better about being without heat. This was not a test of the furnace, it was a contingency test for heat. They were working on the "real tests" at my furnace while I was performing contingencies...

Alot later than we expected things were better than they were before.

-- Anonymous, February 02, 1999

To my understanding, this is NOT a rollover test. That is being done seperately. Analogy - When civil defense drills test emergency response, they develop a hypothetical scenario to stress the infrastructure - communications, hospitals, emergency response, traffic and crowd control. The drill SIMULATES a disaster - say a plane crash. The purpose of the drill is not to test the immunity of airplanes to crashes, it is to stress the ability of the system to respond.

^b There is no rollover testing planned for April. No one (that I'm aware of) will be changing dates on systems. This is to measure preparedness to operate the power system blind with automation out of service ^/b

This isn't creating or simulating problems, it testing the RESPONSE.

-- Anonymous, February 02, 1999

It is a measure of how far short we are from having a government that can protect public health and safety that we are essentially spectators in what amounts to national and local planning for risky tests on the electric grid. We have two kinds of y2k risk periods during 199-2000 and beyond: the well-known y2k-risk dates such as 9-9-99 and the testing times that corporations may choose to use in between those dates.

Grassroots "citizen Y2k awareness and community action" groups are beginning to spread all over, with a primary goal of raising concerns and seeking candid and detailed information on critical topics such as y2k compliance in electric utilities, water and sewer plants, chemical and nuclear facilities, and community contingency planning. They will inevitably turn to local officials for leadership in identifying and remedying identified problems.

It will be critically important for local officials to foster an ethic of candid truth-telling and to show an active public sector role in getting and passing on crucial information. In these Right- To-Know times, citizens who are patronized with phony reassurances or kept in the dark by corporate or public officials will not be happy campers as they dust off their Coleman propane lanterns for power outages and water shortages of uncertain duration.

Beyond information gathering, a vigorous local government will take on itself the "Right To Act" in protecting public health and safety. It will endorse the right of citizens to have a new kind of "informed consent" regarding the risks of y2k, both on the critical risk dates and in the y2k testing by agencies and companies which during 1999 will inevitably result in a certain percentage of y2k-related failures and accidents that could impact workers, the public and the environment. For example, one would not want to allow the local chemical plant to be doing what might be characterized as "Stealth" testing of its most risky full-system y2k problems and fixes on the very weekend that the local hazmat team is off at the annual state convention. Major chemical corporations realize that workers will be at high risk from facility upsets caused by y2k problems, either on the difficult dates or when the facilities are undergoing the necessarily but risky full-scale, on-line testing during 1999 to determine if y2k fixes are successful. The British government's Health and Safety Executive, much more proactive on y2k issues than U.S. EPA or OSHA, has published detailed guidance which recommends to companies testing for y2k progress: "A 'Permit for Test' procedure, where all [necessary safety] measures are checklisted and agreed before testing, is essential." U.S. workers, take note.

A coincidental but powerful new development may stimulate more precautionary actions. The most dangerous oil and chemical facilities, 66,000 of them nationwide, are now calculating their hypothetical "worst case accident scenarios" for fires, explosions and toxic gas clouds that can impact populations off-site. This new and often scary information must be released before the 1990 Clean Air Act's deadline of June 20, 1999. Facilities such as refineries, propane sites, water and sewage treatment plants using gaseous chlorine, and chemical warehouses must do a thorough self-examination of possible disasters and prepare for the first time to disclose potential off-site disaster impact zones ("vulnerable zones") to the public.

The Chlorine Institute, for example, has recently re-published its Pamphlet 74, "Estimating the Area Affected by a Chlorine Release". It tells chlorine users (many of whom are municipal water and wastewater plants) that a "worst case scenario" from just one standard 90-ton chlorine railcar must be reported as stretching out 3 miles wide and 41 miles long into the community. Hundreds of communities will be surprised by this new information.

Congress's explicit intent here is not primarily to drive emergency planning, but to stimulate intense and well-informed community dialogues and steady public pressure over time for serious reduction of major chemical disaster risks in communities at significant risk. The new accident risk information will put dangerous facilities and their circles of potential impacts "on the map" in the media in a new and vivid way - Chemical Week magazine (July2/9, 1997) said that for chemical executives this is "the most daunting public relations challenge this decade."

How will this information hit the public in tandem with the 1999 developments in the y2k arena? Computer programs and computer chips embedded in millions of control devices worldwide may fail massively and simultaneously on any of several dates in 1999, and especially at the most risky time at the turn of the Millennium on January 1, 2000. While no one knows for sure how bad the problem will be, the results could include catastrophic chemical releases putting thousands at risk and damaging the environment. In the words of one major chemical company's y2k manager: "In this job you spend a lot of time being rocked back on your heels!"

The massive, although uneven, corporate and government efforts to begin to identify and remedy the millions of y2k-related computer program and embedded chip problems may eventually cost $1 Trillion worldwide. Despite these efforts the situation remains perilous and poses a real chance of social breakdowns in the U.S. and in many other countries, mainly because of our profound dependence on cheap and available electric power.

The two major streams of powerful new risk information may converge in a way that may dramatically increase the power of each to move the public and local officials to new levels of concern and, one can hope, to vigorous risk reduction. Finding an appropriate level of community concern, however, and taking the necessary risk abatement measures, will be a most severe challenge to local community democracy, creativity, compassion and resilience.

(Sorry, I don't know how to prevent such unlovely formatting here when I paste in some text from elsewhere...) Fred

-- Anonymous, February 02, 1999

Fred,

"Finding an appropriate level of community concern, however, and taking the necessary risk abatement measures, will be a most severe challenge to local community democracy, creativity, compassion and resilience."

I'd like you to expand on "vigorous risk reduction" as an intended result of moving public and local officials to new levels of concern.

What kind of risks do you intend to reduce, business as usual or the health and welfare of my children?

They're not necessarily incompatible. But, how we go about planning for each may be a qualitatively different activity.

~C~

~C~

-- Anonymous, February 02, 1999

Sheila, there is a method of deduction called the "Occam's Razor" test. The basic idea of Occam is that the simplest explanation among many is usually the correct one. The simplest reason I can come up with for scheduling the response tests on risk dates goes like this:

NERC decides a couple of industry wide drills will be beneficial. NERC is also getting flak from utilities in the industry over the time and paperwork required by overseeing agencies, because it takes people away from regular or Y2K work. To do the drills, people have to be positioned at various sites, including outlying areas of possible risk - and for that day they will not be doing their normal work. So NERC decides to kill two or more birds with one stone; get the drills done, have people on alert for the risk days, AND only distract from regular work for two days instead of more.

The whole plan is very likely to be simply a compromise between what NERC wants and what demands the least extra time, distraction, and expense for utilities. Whadda ya think? Makes sense, huh?

-- Anonymous, February 02, 1999

Bonnie, Yes Indeed, and thank you.

The problem for me was that it appeared as if they were going to do a rollover at the time of the test. I did read the text, more than once, and I did understand that they were going to do a simulated communications problem, but I did not recognize the rollover as a simulation too.

This problem is hard on all of us. We must first remember our limitations and be kind to one another.

Let me say, that I am sorry for the impatience I have dished out here. And let's all hope we can come out of this a better human being, no matter what the outcome.

And I AM glad you are here to help me understand, even though it takes a brick sometimes.

-- Anonymous, February 03, 1999

It is a measure of how far short we are from having a government that can protect public health and safety that we are essentially spectators in what amounts to national and local planning for risky tests on the electric grid.

^I No, what you have is an industry planning to perform tests that it deems necessary in the manner that provides the best results at the least risk. You seem to think that the government is more capable than the engineers of the industry. If the industry is so incompetent, please explain the history of reliable, cheap electricity  reliability rates in excess of 95%? ^I

We have two kinds of y2k risk periods during 199-2000 and beyond: the well-known y2k-risk dates such as 9-9-99 and the testing times that corporations may choose to use in between those dates. Grassroots "citizen Y2k awareness and community action" groups are beginning to spread all over, with a primary goal of raising concerns and seeking candid and detailed information on critical topics such as y2k compliance in electric utilities, water and sewer plants, chemical and nuclear facilities, and community contingency planning. ^I And their role in community planning is proper and prudent. Which grassroots group to you contend is more capable than the utilities at developing contingency plans for operation of the utility? You have the right to know what to expect in the way of electrical reliability  this is necessary for your community contingency planning efforts. This is being developed and was reported by NERC. The problem is your refusal to accept the summary reports and insistance on detailed information. Remember, the vendors are still operating (for now) in a free market, capitalistic environment. They are more willing to cooperate (read as: HELP US SOLVE THE PROBLEM) with us when THEY have assurances that no proprietary information will be leaked to their competition. How about I send you the tech bulletins, schematics, and application drawings for all embedded chip devices. You and the grassroots organizations can analyze them  call me when youre done.^I

They will inevitably turn to local officials for leadership in identifying and remedying identified problems.

^Perhaps Big Brother is more competent than even the grass-roots organizations?^I

It will be critically important for local officials to foster an ethic of candid truth-telling and to show an active public sector role in getting and passing on crucial information. In these Right- To-Know times, citizens who are patronized with phony reassurances or kept in the dark by corporate or public officials will not be happy campers as they dust off their Coleman propane lanterns for power outages and water shortages of uncertain duration.

^I Please show me the constitutional basis for an absolute right to know. As a matter of fact, one the functions of government may be in PREVENT the undesired disclosure of corporate info to foreign governments and competitors.^I

Beyond information gathering, a vigorous local government will take on itself the "Right To Act" in protecting public health and safety. It will endorse the right of citizens to have a new kind of "informed consent" regarding the risks of y2k, both on the critical risk dates and in the y2k testing by agencies and companies which during 1999 will inevitably result in a certain percentage of y2k-related failures and accidents that could impact workers, the public and the environment.

^I Inevitably? Really? Support that assertion with the facts  I have a right to know. Informed consent? This paragraph is complete bullshit. Do you advocate that the maintenance schedules of utility companies be reviewed by the bozo politicians at the federal or the yahoos at the local level? What do a bunch of lawyers know about the maintenance requirements of a complex power system. THIS IS ADVOCATING COMPLETE SOCIALISM. This model you present has been tried. I suggest spending New Years eve in the former Soviet Union. With their government control of industry they MUST be way more reliable that we are. OR, maybe we could round up all the utility engineers and management and put them in a gulag - the government would then be free to make the appropriate appointments (cousins and in-laws) necessary to solve the problem.^I

Congress's explicit intent here is not primarily to drive emergency planning, but to stimulate intense and well-informed community dialogues and steady public pressure over time for serious reduction of major chemical disaster risks in communities at significant risk. The new accident risk information will put dangerous facilities and their circles of potential impacts "on the map" in the media in a new and vivid way - Chemical Week magazine (July2/9, 1997) said that for chemical executives this is "the most daunting public relations challenge this decade."

How will this information hit the public in tandem with the 1999 developments in the y2k arena? Computer programs and computer chips embedded in millions of control devices worldwide may fail massively and simultaneously on any of several dates in 1999, and especially at the most risky time at the turn of the Millennium on January 1, 2000. While no one knows for sure how bad the problem will be, the results could include catastrophic chemical releases putting thousands at risk and damaging the environment. In the words of one major chemical company's y2k manager: "In this job you spend a lot of time being rocked back on your heels!"

^I NO ONE KNOWS FOR SURE. But NERC has a good guess, you just wont believe it. The only way to end the conspiracy is to have the government take over the utilities. Yea, the Clinton administration  sources you can TRUST.^I

The massive, although uneven, corporate and government efforts to begin to identify and remedy the millions of y2k-related computer program and embedded chip problems may eventually cost $1 Trillion worldwide. Despite these efforts the situation remains perilous and poses a real chance of social breakdowns in the U.S. and in many other countries, mainly because of our profound dependence on cheap and available electric power.

^I How much are you willing to increase your taxes to pay for the costs of government mandated testing? Previous paragraph states outcome unknown, but this one states there is a REAL chance of social breakdowns? Quite a leap. Perhaps we already see evidence of the social breakdown.^I

-- Fred Millar (fmillar@erols.com), February 02, 1999.

^I Fred, I cant tell if you are the author, or just pasting from another source. These themes frighten me as an American. Any comment was intended for the author, and the author only, and only to the extent that I VIGOROUSLY disagree with the foundational principles that the words betray.^I

-- Anonymous, February 03, 1999

Should we support martial law at rollover to prevent "civilians" from overloading communcations infrastructure with unnecessary Y2K internet activity and trivial phone calls to remote relatives at midnight of 12/31?

-- Anonymous, February 09, 1999

Just a little tech talk to shed some light. The 9/9/99 is an old problem in the mainframe world. This should have little or nothing to do with embedded systems.

Robert Egan wrote:

In all my years of IT work, I never saw 9/9/99 used as a "sentinel" date. I never even heard of it, either as a real issue or as a gag, until I started reading the posts here. Frankly, I find it hard to believe there's more than a dozen cases in the entire world.

Cory Hamasaki responded ...

It was common in 2nd generation (before my time). Some systems and languages didn't have EOF sensing. There is a huge inventory of code that pre-dates databases and a significant inventory that pre-dates EOF sensing. In the mid 1970's I saw 9's cards being taught as the proper way to terminate files.

Will this problem occur in production systems somewhere? Absolutely. Is it common? I'd guess not but I don't know.

Can it be fixed on failure? Depends on the expertise in the shop, do they have the source? Will they notice the problems? How long before Sam complains about lost data.

I get back to the 50,000 IBM style mainframes and millions of legacy systems. COBOL (and assembly language, RPG, DYL-260, MarkIV, Nomad, PL/I, and hundreds of other languages and databases) runs the world.

Best to you.

-- Anonymous, February 11, 1999

I'm going to stick my neck way out, and say that the often-repeated problem with Sept. 9, 1999 is pure balderdash...an urban myth.

To you and me, the date is 9/9/99, or 9999, only because we intuitively and arbitrarily discard the implied zeros in front of the day and month. Computers don't do intuitive or arbitrary things, so the date field in a computer is 090999, or some variation thereof. If a programmer needed to invent a code to act as a "flag" with a special meaning, he would not select "090999", he would select "999999". No legitimate date would ever be mis-interpreted as the flag value.

For a more complete discussion on the subject, reference Lane Core's excellent article at http://www.y2ktimebomb.com/Computech/Issues/lcore9846.htm

If anyone has an explanation on *exactly* how Sept 9, 1999 can cause a problem I would like to hear it.

-- Anonymous, February 11, 1999

David, Obed, and All:

These are great answers that I am going to incorporate into my knowlege on this issue. As I see it, there is risk on April 9, 1999 and September 9, 1999; however, I am beginning to realize that this risk is minimal

Since I asked this question, I have come to realize that the '99 problems are far fewer than the 1999/2000 date math problems. I would even venture to guess that they are 1000/1 (Y2K to '99). In this light, the '99 problems should not be compared to year 2000 problems; however, these problems do happen at a poor time. This was the orginal reason for my question. But, that was before I understood the nature of the testing.

Also in question is the robustness of our infrastructure. The '99 problems may not be large, but how strong is our infrastructure?

The matter really comes down to the fact that this is not a Y2K test. It is a contingency plan test. Calling this a Y2K test makes it sould like it has something to do with fixing the Y2K problem. Even as a contingency plan test, this is testing the telecommunications contingencies which could be problematic even if an astroid hits a satellite, or if a war started in the U.S. (God forbid) It makes good sense; it just doesn't communicate well to me to call it a Y2K test.

-- Anonymous, February 11, 1999

I didn't realize until recently that April 9th is a Friday. NOW the drill timing seems bad. It will not give sufficient time for a proper de-brief and critique until Monday. All should take care that meticulous notes are taken so that recollections will be easily refreshed after the weekend, OR do the critique before leaving for the weekend.

The potential for messing up or delaying my weekend is the BEST reason so far for NERC to change the drill date.

-- Anonymous, February 16, 1999