Y2k Consultant Doubles Personal Preparations

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

My family has decided to increase our personal preps "insurance policy" to 60 days.

For the past 2 years, I have been involved with a Y2k Think Tank and served as a Y2k consultant to many firms and organizations. My focus has been on embedded systems and personal computers. Paula Gordon has contributed to the TB2000 forum on numerous occasions, and her read on potential embedded systems problems is right on target.(somebody please provide links to some of Paula's contributions)

Our decision to increase personal preps is a result of my receiving a fax copy of a Public Summary of the Rotterdam Report. This independent report confirms that personal computers/desktops will slow down and cease to function after the rollover because of short date issues.

The Rotterdam Report and Royal Society Reports have been followed closely by NATO. http://www.nato.int/Y2K/tools.htm

I will post a link to the Rotterdam Report when it is available. Hopefully by the end of the week.

A presentation was made to the Royal Society in England on Oct. 5, 1999 explaining the technical aspects of this problem and the White Paper can be viewed at this thread:

http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=001qFx

It should be noted that the Rotterdam test was conducted using "vendor compliant" software and all available patches had been downloaded.

Could our friendly personal computers be the real Y2k achilles heel? Are personal computers ever used for "Mission Critical" systems?

-- Brian Bretzke (bretzke@tir.com), November 29, 1999

Answers

What was the name of your "Y2K Think Tank" and who did you consult for?

-- Friendly (question@s.com), November 29, 1999.

Never seen anyone post such a self-important post like this one before. But more importantly, the post tries to give support to a totally bogus idea that PC's slow down and crash if they dont get this special program to fix them.

We have run 15 testmachines (different motherboards, win3.1, win95 and win98) for more than 10 days after we rolled them into the year 2000 and nothing happened. Only lost a minute or two on a few of them but that is typical for most PC's. One crashed due to a CPU fan not working. We still have two machines running as 2000 dates and have not been booted or crashed for weeks.

-- hamster (hamster@mycage.com), November 29, 1999.


Hamster:

Have you turned them off, then turned them on again? Did the 2000 dates reappear (or was it 1984)? Have you entered yr 2000 data requiring manipulation from 1999 to 2000 (say payroll, tax witholding, etc), then export/import data packages? What were the results?

-- Nelson Isada (isada@alaska.net), November 29, 1999.


Thanks Brian. Humm.

Look forward to the later Rotterdam link.

Diane

Nato hotlink...

N A T O Year 2000 Support Site

http://www.nato.int/Y2K/ tools.htm

Royal Society in England on Oct. 5, 1999 ... White Paper thread hotlink...

The Rotterdam Report, MFX2000, Creeping Corruption. What do we know? Technical assessments requested.

http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id= 001qFx



-- Diane J. Squire (sacredspaces@yahoo.com), November 29, 1999.


Brian, thanks for your upkeep on this most important matter. PLEASE POST THE SUMMARY ASAP> i read this post after i sent mine in sorry for the question again....

I need the nails for some caskets of a few CEO's i know. LOL

-- Blue (bluefish@thepnd.com), November 29, 1999.



One reason that this issue may not have surfaced earlier, is that the "creeping corruption" problem is caused (they claim), by a gradual accumulation of unhandled errors caused by transactions involving short format dates.

Transactions. So if you park a bank of PC's on a bench, advance the clocks and wait a month, nothing will change. No problem. However if the software is *used* and transactions are created causing unhandled errors, then the problem gradually begins to appear.

If the Rotterdam test just advanced clocks and waited, it is IMHO meaningless. But if it somehow mimicked a real post-Y2K business environment, (including daily re-boots-important point),and came up with these startling findings, well...be prepared to bail out of software stocks.

And Brian asks: "Could our friendly personal computers be the real Y2k achilles heel? Are personal computers ever used for "Mission Critical" systems?"

Could be. Many mainframe apps are still in use in large orgs, but the old "dumb terminals" used to access the mainframe data are long gone. Terminal Emulation software running on Wintel PC's is quite common.

Thanks Brian for keeping us updated on this evolving issue. I look forward to the link.

Regards-

-- Lewis (aslanshow@yahoo.com), November 29, 1999.


Hamster and Friendly-

Skepticism is good. Unfortunately, we are beginning to get some scientific verification of what will happen after the rollover, and the results aren't good.

Friendly- The Think Tank is called "Master Y2k Think-Tank".

A partial list of Associates:

William M. Ulrich. Authored one of the first textbooks on the Y2k issue.

Steven L. Hock. Widely known as the first attorney in the U.S. to recognize and call attention to the legal implications of the Y2k problem.

Comdisco, Inc. One of the world's leading providers of technology services. Comdisco's support of customer-declared disasters include: The World Trade Center Bombing, the Chicago flood and the 1989 San Francisco earthquake.

Best Solution 2000- That's me.

Thelen Reid & Priest LLP. Thielen's Year 2000 Team is one of the nation's top providers of Y2k legal services. Clients include Charles Schwab & Co., Federated Stores, Visa, Federal Home Loan Bank of San Francisco and Phillips Petroleum.

Triaxsys Research LLC. A cross discipline research organization that provides statistical analysis of Y2k disclosure statements.

Scott J. Nathan. Attorney specializing in complex technical issues, president of dispute Resolution 2000, Inc., a forum providing Y2k mediation and arbitration dispute services.

Triennium Corporation has provided business technology solutions to more than 2000 of the leading 1,500 enterprises in North America.

Marketing Services Group- Has served as a strategic advisor to leading professional service firms including KPMG, Arthur Andersen, Andersen Consulting.

Hamster- A presentation on the short date problem with PCs was given to the President's Council on Y2k several months ago. Since we haven't heard anything about it, I can only assume they agree with your assessment of the problem.

-- Brian Bretzke (bretzke@tir.com), November 29, 1999.


I've got a bad feeling about this...

-- (popgoes@the.bubble), November 29, 1999.

We did do rolls and reboots first and rolls without booting, RTC problems were noted on 1 486 machine but a reboot cleaned it up. The machines we rolled were used daily. They were not used for accounting but were used for normal applications like image processing, internet, cdrom burning, editing, etc...

I am skeptical about the claim that almost all pc's will slowdown and/or crash if not fixed with this program.

Its known, under demo conditions, this program can find date usage in obj code, but thats all I saw written about it. I didnt see it used in a live production envirnment nor did we see any end result of having a PC fully patched and then re-intruduced into a stream or live production.

Likewise if a application was written to use 2 digit dates AND 4 digit dates as a work around it was not noted how the fixer would be able to tell that a 2-digit date handling was desired, that is the fixer doesnt know the logic of what the application does with a date rather that it simply uses 2 or 4 digit dates.

-- hamster (hamster@mycage.com), November 29, 1999.


Well DONE. Looking forward to the summary pages. If their is a fee $$ involved with the release of the report I'll gladly pay...

Blue

-- BLUE (bluefish@thepond.com), November 29, 1999.



hamster, thanks for the details. Your results are encouraging.

One thing that concerns me about this, is that there may not be enough time for anyone else to run a similar set of tests for independent verification of the problem. This could breed alot of FUD about the effects of Y2K. (as if we needed more.)

We'll all be working on test benches soon enough..

I guess we could just re-install the OS, applications and patches on every PC in our organization every 30 days.

How hard could that be?.....

-- Lewis (aslanshow@yahoo.com), November 29, 1999.


Brian Bretzke, I guess you're getting past your brazen, "The 1999-2000 rollover will be a non-event" prediction then, are you?

Hope so. Too many weak links in that ohso important infra-structure IT chain.
Cheers,

-- faith'nhope (y2kaos@home.com), November 29, 1999.


Brian Bretzke, I guess you're getting past your recent brazen prediction,
"The 1999-2000 rollover will be a non-event", then, are you?

Hope so. Seems an extremely poorly determined conclusion. Too many weak links in that ohso important infra-structure IT chain.
Cheers,

-- faith'nhope (y2kaos@home.com), November 29, 1999.


Faith- I try to remain as optimistic as possible about the actual rollover event. Still stick to my guns that the actual rollover will be pretty anti-climactic. I see the Y2k problems as a cancer, not a heart attack. Have seen lots of paper clip and bubblegum contingency plans that could keep the dam from bursting for a while.

-- Brian Bretzke (bretzke@tir.com), November 29, 1999.

I'm hoping this PC study is wrong.

It's my understanding that many large-scale embedded systems (LSES) are, in fact, also PC-based. This includes Distributed Control Systems (DCS) and SCADA (Supervisory Control and Data Acquisition) systems used throughout the power, chemical and petrochemical, oil and natural gas, and basic manufacturing industries. But perhaps these PC-based embedded systems wouldn't be processing very many date fields and so wouldn't be susceptible to this type of "creeping corruption"?

-- Don Florence (dflorence@zianet.com), November 29, 1999.



I read the Rotterdam Report as posted here a few days ago and a lot of it looked bogus to me. Like a sales pitch. If PC's are rebooted periodically, that will correct any "creeping corruption" from errors. Any kind of error can leave a PC operating system unstable - that's what the reboot button is for. The instability is not permanent and is easily correctable. I use my reboot button all the time due the inherent instability of the Windows operating systems. Also, the use of two-digit years in data files, or any other type of file, will not neccesarily cause any problems. Nor will the use of so-called "non-compliant compilers".

Incidently, as a programmer, I know that the Y2K bug is ubiquitous and that the many claims of government and corporations having "fixed" the problem are greatly exagerated. I am certain that much software is going to fail in unpredictable ways in the near future. What will result from this, I don't know. Since all sorts of manufacturing and supply disruptions COULD result, it seems prudent to prepare. Cheap insurance.

I still think the Rotterdam report looks bogus though. How about some comments from other programmers?

-- A Programmer (notmy@address.com), November 29, 1999.


A Programmer,

There are so many little things that bug me about this 'miracle program'. For instance what if a program MUST be able to read the dates in 2 digit format for handling 'legacy' data from archival storage or non-compliant data streams. Does this program just go helter-skelter through all apps and move them to "compliancy". Supposedly it asks you if you want to do it, but the complaint is that you cannot tell where exactly in the logic the fix is being done. Lacking a xref to the source can you really tell if you arent just making a bad thing even worse by using this 'wonder tool'.

I am shocked that NATO finds this to be a recommended program, but reading the full review they seem to think its a 'better than nothing' post-failure recovery tool for low-level-language saavy programmers.

What say you all?.

-- hamster (hamster@mycage.com), November 29, 1999.


Programmer- It's good to be skeptical. Here's a link which gives an unclassified NATO test assessment of this "bogus" stuff. They say Highly Recommended. http://www.nato.int/Y2K/tools/y2ktests/mfxtst.htm

BSI, which established the accepted standards for Y2k, uses this "bogus" stuff.

Be skeptical, but don't be close-minded!

If you have a better Y2k PC solution (software,apps,data,os), please let me know immediately. I'll gladly change paddles in the middle of the stream.

-- Brian Bretzke (bretzke@tir.com), November 29, 1999.


Brian,

I can't address why someone at NATO would recommend this software. I don't know. But there is just too much stuff in the report that doesn't make sense to me, for me to take it seriously.

Another thing that does not make sense to me: The number one recommended use by NATO (see your link) is for patching programs for which source code is not available. I just don't think this is possible. No tool such as this can know what date formats are anticipated or acceptable in a particular program, or how they are used. Many programs require 2-digit years. They can only use 2-digit years. How can you "fix" that without understanding and changing the source code?

Sorry if I appear closed-minded. These are just my opinions, based on working with and around the "Y2K problem" for 20 years.

No, I don't have a better solution. If I did I would use it on my own software, which I will be fixing, one error at a time, well into the year 2000. I don't think there is a "solution".

-- A Programmer (notmy@address.com), November 29, 1999.


Hi Brian, thanks for the info. I'll try and synopsise for the less techically minded: some applications that are very poorly written AND are not Y2K compliant, may experience a new form of corruption that no one else has thought of, therefore give us money. ;)

Serious evaluation: the report is thought provoking, and contains some very useful statistics and remediation advice. However, it only deals with one very small aspect of Y2K. And it IS a sales pitch.

Thanks again Brian, it was an interesting read.

-- Colin MacDonald (roborogerborg@yahoo.com), November 30, 1999.


Brian, I have tried to follow all the threads about this possible pc problem and there are several statements I do not understand so I don't think my pc will have any problems. About the overall Y2K situation, I am not sure about the interconnections and possible failures there also. Kinda complex so I don't think any thing serious will happen.

What do you think about the National Football League ?? I think Boston will come from behind and win. Duh, Have a good day.

-- KickedOutofMensa (Tuttie@FruttieToo.com), November 30, 1999.


Moderation questions? read the FAQ