Question for FACTFINDER

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

I have read with great interest your debates with 'just another engineer'. Have developed considerable respect for the informative and generally civil nature of the discourse but have, like most, concluded that the only conclusion is that there is no basis for a conclusion. Have read and reread Frautschi, Beach, your posts and those of many others. I know, on a direct, personal level that mainframes are going down(the work either just wasn't done or done so poorly that it just doesn't qualify as remediated code)but cobol can,ultimately, be repaired if we have the infrastructure in place to enable the repair process to move forward. A long winded way of saying that I'm really sweating this embedded thing. You(Factfinder) have argued consistently and forcefully that basically its gonna be alright. I direct you to the excerpt from FORTUNE in the post below 'Question about General Motors' and the statements by Ralph Szygenda, CIO at GM. If, after reading this non-hypothetical, real-world scenario you still maintain your posture than I can only assume that you honestly believe that the same effort was expended to ferret out problems in all of the other entities that we depend on for our survival as was that expended at GM. If so, I believe that you are the kind of guy who consistently tries to fill inside straights. If that is your belief then I can only offer in response the stated hope that we all can improbably fill this mother of inside straights. Good luck and many thanks for the informative, interesting and well grounded posts and interchanges with 'just another'.

-- Get Real (gaf@mindspring.com), December 05, 1999

Answers

The Ralph Szygenda GM story is good news ... and bad news. Szygenda talked about embedded systems when their was plenty of time remaining to fix them. Since then many large companies have spent lot's of money to ensure their factories will operate in January. GM is not alone.

The bad news is after the FORTUNE article appeared, a GM public relations executive was assigned to handle ALL future contacts with the press for Szygenda. I'll bet he was never quoted again about any Y2K-related subject (if he wanted to keep his job)!

-- Richard Greene (rgreene2@ford.com), December 05, 1999.


Get Real, Sorry for the belated reply, I did see your message earlier but did not have time to respond. I saw the GM story some time back, and my impression after re-reading it is the same as the one I had when I first saw it - even though most embedded systems problems are minor, it appears that some can be quite significant. And the only way to be sure that your/my/our company is ok for y2k is to do a good job of inventory, assessment, remediation, and testing. The story, like most non-technical news stories also raises more questions than it answers. More about that later.

What I, or you, or anyone "knows" about y2k is limited according to: 1. What we read 2. What we hear from others 3. What we find out on our own (our own assessments and testing)

The key here, is this - is what we "know" accurate? Is it factual?

When I first started doing research into y2k for my job, I was suprised by the many claims regarding y2k problems in embedded systems. It only took a few months of researching in depth the nature of the y2k bugs with our plant equipment to determine that there was (and still is in many places) a huge disconnect between what was "speculated" as to what the nature of y2k bugs in embedded systems might be, and what the large number of problems actually were , which was in almost all cases minor. It was quite a while before I found even a moderate y2k problem, and even longer before I saw a PC based data acquistion system fail as I tested it (not installed plant equipment).

Let me point out one thing here, I stated in an another thread that there are no embedded systems experts. There are so many devices, instruments, systems, etc., that no one could never know everything needed to test evaluate etc all of these systems. So here is my own limited background -I have worked on y2k indirectly as the responsible engineer for several plant instrumentation systems, and directly on a y2k project at another plant. I have direct knowledge of, and had access to data, of the y2k problems found at four different nuclear stations having a total of 8 nuclear units. I have also had direct access through various industry resources to hundreds of reports from other power plants, and even the data of a few petrolium companies. In addition, I have done quite a bit of personal research into other sectors as well regarding y2k in embedded systems - particularly medical devices and building management systems/facility systems.

One thing I have learned - if statements are made regarding y2k in embedded systems by Y2K project managers or leaders with strict IT backgrounds, is that there's maybe a 50-50 chance they have their facts straight concerning embedded systems, depending on the level of contact they have with systems and equipment engineers/technicians who actually know and test these systems.

And if statements are made by companies who offer Y2K services, I drop it down to 25% chance of good information, and I'm being very generous at that.

If something appears in the mass media, newspapers, TV, etc., I drop the reliability factor to 10%.

And if it appears on the internet without a manufacturer or model number, I give it a zero until I get good technical information. I know this from experience, I once started tracking down y2k "failure" reports from a list posted here, and the first half dozen or so turned out to be "vapor" - the devices never really failed at all. So I approach claims here on the internet very cautiously.

Regards,

-- FactFinder (FactFinder@bzn.com), December 06, 1999.


Moderation questions? read the FAQ