Can anyone name a single embedded system that has been tested and come up faulty in a critical infrastructure system?

If not, this problem is only alleged, and no one should panic when there isn't a shred of evidence one way or the other.

If you plan to answer, please don't use unnamed sources, generalities, info that must be purchased, or probabilities when answering this, or I plan to ignore it. (Since that would reduce its evidence value or importance)

I have been looking for about four days on the Internet and have found mostly allegations and what I consider to be unsubstantiated proof.

If there is really a global problem that is not panic based, why have you not been able to come up with actual FREE evidence. Money makes this all sound like a panic that has been used to gain income.

This evidence is REQUIRED by engineers like myself, BEFORE I panic and cause others to do so.

Just like the engineer in the movie "Titanic", I must make sure that "going down" is a "mathematical certainty" before I can convince my boss to get on a lifeboat, let alone save others.

-- Anonymous, May 03, 1998

Answers

Response to Mr. Cowels and all, can you name one chip that has a problem?

I was wrong. I have continued to work on this problem, and have found evidence that there is an infrastructure problem. It is not in the electricity industry, but it certainly could be in the electricity industry just as easily.

At the "House of Commons" web site, a report was generated by a group doing research on behalf of the government of the UK. (www.parliament.the-stationery-office.co.uk) . Here is what was stated to the British government in the "Select Committee on Science and Technology Second Report, Chapter One: The Century Date Change Problem, item number 5":

"5. The majority of our witnesses, including many from well-known organizations such as Shell UK, SmithKline Beecham, Barclays Bank, Sainsbury's, and the BBC, agreed that the century date problem posed a genuine and significant problem... For instance, Morgan Stanley told us that 'to date, we have encountered Year 2000 date problems in nearly all of our internally developed systems ... we have also found that networks, telecommunications infrastructure, and office building systems are affected. In fact, the central building management system in our ... office which controls and monitors fire alarm, water detection and other safety systems has defective embedded chips"

I am confessing that the tone in my question (and the spelling of names) was wrong. I want to thank Mr. Cowles for his efforts in this very difficult time. I am sure that more certain embedded problems will surface (as soon as people start looking into it!) if what was shown to the House of Commons in the UK was correct.

Thank you all for your patience,

Troy

-- Anonymous, May 04, 1998

Response to Mr. Cowels and all, can you name one chip that has a problem?

The question seems simple enough. Just find someone with the courage to state that they "have" had critical problems, and this is exactly what they were. However, we are now entering the "cover-up" stages, where most of the critical agencies, whether Utilities, Finances, or Defense, etc., will more apt to be silent. I think we can surmize that there will be actual failures, as the embedded chip manufacturers have already produced listings of both compliant and non-compliant controlers.

Perhaps there will be someone who will risk their jobs to inform us, or we can spend the money to buy the information.

-- Anonymous, May 04, 1998

(I know this is an old thread but thought I'd add this comment).

Here is Motorola stating pretty clearly that these seventy chips have a Y2K problem and won't be fixed.

http://www.mot-sps.com/y2k/black_prod.html

And here is another chip they just added on 12/04/1998:

http://mot-sps.com/y2k/yr2knote.html

Here are the numbers for embedded CPUs that were shipped this year (1998) alone. Only a small percentage of these embedded CPUS (MCUs or microcontrollers if you prefer) will be tied to an RTC and have the potential for a Y2K problem. Nonetheless, these numbers are large and a small percentage of any of these numbers is still a large number.

4-bit MCUs Approximately 1 billion 8-bit MCUs 2 billion 16-bit MCUs 800 million 32-bit and greater 200 million Total Most likely more than 4 billion

--AJ

P.S. Motorola isn't the only company that makes chips with an integrated RTC.

P.P.S These are numbers that are publicly available on the Internet. They do not include chip counts that are hard to quantify, such as those used in military applications. Many divisions of the military and different security agencies have their own chip fabs.

-- Anonymous, December 14, 1998