Link

-- (@ .), February 17, 2000

Answers

cory hamasaki's DC Y2K Weather Report

February 17, 2000 - Y2K plus 47 days. - WRP135

http://www.kiyoinc.com/current.html - Final - $5.00 Cover Price.

(c) 1999, 2000 Cory Hamasaki - I grant permission to distribute and reproduce this newsletter as long as this entire document is reproduced in its entirety. You may optionally quote an individual article but you should include this header down to the tearline or provide a link to the header. I do not grant permission to a commercial publisher to reprint this in print media. This entire document is a Year 2000 Information Disclosure as defined in the Year 2000 Information and Readiness Disclosure Act, S2392.

--------------------tearline -----------------

In this issue:

1. Polly Moshe Was Right.

2. Where's the kaboom?

3. What Happened

4. CCCC

-- What Happened --

WRP 135 was scheduled for February 1, 19100, no make it 3900, ah, 199A? The original title was supposed to be "Polly Moshe was Right" as a tribute to a long running debate in USENET newsgroup comp.software.year-2000 with Moshe the mainframer polly.

Essentially, Moshe argued that Y2K would be solved by a combination of the remediation in 1998 and 1999 and by fixing the remaining few problems when they occurred in 2000.

What Moshe was right about is that the power stayed on, the water continued to flow, and goods and services have been available. The problems that we have seen, while strange and puzzling, are not over the Edwards 3 threshold, defined as 80 hour workweeks. It's only as you approach Edwards 5 that the scenario's morph to the "End of the World as We Know it".

So here we are, with more problems than anyone wanted but with the economy humming like the well oiled machine that it is, or something like that.

Late as usual, I'm pushing out WRP135 with two thoughtful articles and even more than a normal dose of my clueless rambling.

-- Where's the Kaboom --

"Where's the kaboom? There was supposed to be an Earth-shattering kaboom!"

-- Marvin The Martian.

I admit I expected a kaboom. I've got a right to expect a kaboom. I've architected the large systems - international, 24/7, fortune fifty, multi-billion dollar, globally scalable, multi-terabyte, fat pipe plumbing. Most of you mugs caught the fear second hand, but I've seen too many real big things really screw up big time with my very own eyes.

I started expecting the kaboom in June 1998. That sounds tame next to the folk who revealed last year they'd been remediating since 1995. I never figured out whether to believe in anyone paranoid enough to have started in 1995. But in June 1998 I put myself a hypothetical. What might go wrong with the systems I'd built during a global date rollover?

I don't need to restate my worries - anyone who wants a good chuckle can go read my 1998 Halloween nightmare at http://home.san.rr.com/merel/bs.html. Suffice it to say I didn't dream up any risk anyone else didn't dream up too.

Still I had good reason to believe in my nightmare. After two decades I may not be at the top of my profession, but I'm on a first name basis with the guys and gals who are. No one I trusted to know knew how to gauge the real risks. I sure didn't know how. Our ignorance was what scared me the most.

As each of 1999's dates of doom scrolled by without incident, my fears of a kaboom only slightly diminished. Like many of you I lived in a kind of schizoid world. I would read thestraight press taking seriously the daily nonsense, and then I'd check out Cory's site and Sanger's and TB2K, and scare myself silly.

I prepped for the worst. I don't think I went completely nuts. A few hundred went on junk I can't possibly need, but most of the other stuff turned out to be good fun anyway. I use the solar oven at least once a month and refuse to burn messy fossil fuels for bbq ever again. The CNG truck is my favourite vehicle. I swim every morning in the cement pond, and I'm amazed at how fast our little family is making a dent in the mountain of TP.

At least I didn't head for the hills. I always figured the hills would be full of folk like Milne and Infomagic with their uzis. But today, like many of you, I'm supplementing my diet with regular doses of crow and hat. When the lights stayed on I figured the first order of business was to commence drawing up a menu.

"Where's The Kaboom?"

* First, we programmers expect external events to be screwed up, so we put a lot of sanity and consistency checks in I/O and DBMSes that tend to stop bad data from propagating. So the little failures couldn't easily cascade to become big failures. Alarms were raised, but these systems were really never built to just keel over and die from dirty data. They get dirty data anyway from time to time, and they're built to sift it out.

* Second, once systems are successfully deployed the only real source of Y2K related errors is going to be in the components of the system. Component remediation was where all the money went, but what was worrisome, for me at least, was the lack of end-to-end functional testing of deployed systems. The engineers were spread too thin and it was too expensive to take core infrastructure systems offline to test. Even in the US this didn't get done. In the event, the deployments themselves weren't substantially altered and just didn't need end to end re-testing.

* Third, humans are more involved in making significant production and maintenance decisions than we usually think, and these humans are equipped with effective built-in BS detectors. Reports based on dates like 19100 were mostly disregarded or corrected long before they affected consumers. Furthermore, for most business tasks, a human wild-assed guess isn't so much worse than an informed response. Certainly it's not so much worse it makes enough difference to kill a business before the systems can be patched.

* Fourth it seems dates really aren't that important to most embedded systems. After the fact this seems kind of obvious. Dates used in these systems are for reporting, not logic. This might not be true of the database oriented machines, but for the embeddeds in utilities it seems like it was damn-near universal. To Cory's credit, he expected this. Point Tiwi was the exception, not the rule.

* Fifth, there just ain't that many mainframes in the world any more. There are a few screwing themselves up even now, no doubt, but so much processing has been pushed off onto distributed workstation/PC systems that the mainframe screwups just aren't biting manypeople. Perhaps if tax processing is affected this might change ... but even that seems unlikely now given the time available to fix on failure.

* Last and maybe best, we engineers are constantly screwing up and correcting our screw ups. The Y2K signal didn't rise above the background hiss of errors that go on in all our engineering efforts all the time. Y2K wasn't a big deal because we're simply far less competent than we usually think.

Now the question I get asked the most by friends and relatives, who have fair reason to wonder, is whether in my heart of hearts I'm not just a little sorry civilization didn't end up in the crapper. Wasn't I secretly looking forward to the big vacation, the return to barter and elbow grease? The enema that western civ. so desperately needs?

Hell No! If there's one thing 18 months of paranoia taught me, it's that civilization is something to love and respect. Sure the hive is incestuous and deluded, growing like yeast in a barrel, extincting species at 10,000 times the background rate just to build burgers and Buicks. But civilization is the cure, not the disease. Without it, life would be dirty, feudal, violent, and short. We engineers have our work cut out for us to fix the real problems in our society. We might still fail, but we need a kaboom like a hole in the head.

(c) 2000 by Peter Merel. This article is published as part of cory hamasaki's DC Y2K Weather Report and may be reproduced under the same terms and conditions. All other rights reserved to the author

-- What Happened --

What happened on New Year's Eve?

By Arnold Trembley, Mon-24-Jan-2000

Like many programmers all over the world, I spent my New Year's Eve at the office, watching my programs in case anything unusual happened. It was a pretty quiet way to celebrate and I only got an occasional TeeVee glimpse of New Year's celebrations in various time zones.

Things went better than I expected, no problems with electricity, telephones, natural gas, water, sewage, elevators, or cars. There were no bank runs or food riots. No national governments collapsed, and none declared martial law. The Dow-Jones Industrial Average didn't drop to 3000 points. Gold didn't rise to $800 an ounce. Oil didn't climb to $100 a barrel. The mainframe wasn't turned off and replaced by client/server. COBOL wasn't replaced by Java/Perl/C++/VisualBasic.

It's now three weeks since the rollover, and Ed Yourdon has a new essay on what we've learned from Y2K. Ed says we've learned that if it's absolutely critical, IT can meet a deadline. He goes on to say, if IT can meet a project deadline for Y2K, why can't they do it every time?

I would not have drawn that conclusion, having worked on more projects that met their deadline than were late or cancelled. Ed makes his living as a pundit, and he was just one of many on the internet who contributed to the debate, trying to raise awareness of the problem.

I got some good Y2K technical tips from usenet, but eventually the debate became extremely emotional, and feelings were hurt. Perhaps it's not surprising that some "doomers" are still convinced that technological civilization is about to implode. And if we have a recession in 2000, it will likely be blamed on Y2K, instead of four years of go-go economic expansion.

Oh, sure, there were a few problems. I know of a VSE shop that had to recompile 400 COBOL programs in mid-January to repair a day-of-week bug. But it's only batch. They fixed it before their month-end processing, and their customers will never know there was a problem.

Having fixed my share of Y2K bugs, I would say that Y2K was primarily a maintenance project rather than new development. Fixing Y2K had a higher profile than a normal project, and in some ways it was a larger project, but most of us just did our jobs as best we could.

Y2K was different because it was a widely dispersed problem that the non-programming public could almost-but-not-quite understand. They could never determine the extent and

consequences of possible failures. No one knew. Just as no one knows how many computers are doing productive work, much less how many programs or lines of code. So it was easy to believe that the worst possible outcome that could be imagined was the most likely outcome.

The next watch date is Tuesday 29 February 2000, and after that Monday 01 January 2001. I'm sure there will be a few problems somewhere. Somebody's beeper will go off. Somebody may have to fix a bug in the middle of the night. It's hard to see how it could be any worse than the problems we've already encountered and solved.

Y2K isn't quite over yet, but most programmers have declared victory and moved on. It's been an emotional ride with a big letdown at the end. If the programmers did their jobs right, that's the way it was supposed to happen.

(c) 2000 Arnold Trembley This article is published as part of cory hamasaki's DC Y2K Weather Report and may be reproduced under the same terms and conditions. All other rights reserved to the author

--cccc-- Cory's Closing Clueless Comments

What happened to all the mainframe failures?

Two c.s.y2k posters, Johnny Yuma and DRACON301 may have found some, Johnny cited week long computer problems at Ft. Meade. DRACON301 provided a link to an article about DeeCee's financial system failure. This week, IBM announced that they are shaking down the deadbeats at DeeCee for 28 million bucks for Y2K work.

What was my line? Oh yeah, Edwards 3.75, system failures like we've never seen before. Odd stuff, geeks in overdrive. You tell me what's going on under-the-hood in Johnny Yuma's Ft. Meade or DRACON301's DeeCee city government.

Keeping this about mainframes. Some have forgotten my public questioning of the embedded, secondary clock, power industry issues and have forgotten that the WRPs presented different sides of the issue.

Steve Heller and I have received some clueless natterings from polly- come-latelies as a result of the January Y2K article in Wired!. I managed to turn that off by pointing the Wired! readers to the "If you don't have a sense of humor, turn back now." banner. Spare me the natterings of the pollies.

Interestingly, the original Y2K Alarmist article was published by Wired! It was far more doomer than the 1997 "The Day the World Crashes" article in Newsweek. Wired! more than any other publication, made the case for the end of the world as we know it. This was their 1998 piece that featured pictures of the DMR Telco consultant in his survivalist bunker.

In contrast, the WRPs ran knock-down duels between InfoMagic and Bruce Webster, where Bruce argued for a resilient network v. Infomagic's Charlotte's web.

I guess the publishers and editors of the Washington Post agree with the financial columns, the horoscope, Dear Abby, or the Op-Ed pieces. I suppose that when the Post reports on a horrific auto crash or cautions that an intersection is unsafe, it means that they want automobile accidents.

If something's happening like Yuma's post implies, gold plated, guarenteed not to fail under any circumstances, systems failed and failed and failed again. Of course, the weather might have something to do with it. Maryland is under a weather alert so maybe it wasn't Y2K. Maybe it was the weather. We know that 12, 18 inches of snow can take down mainframes and keep them down for a week, even if there are no power problems. (I don't know how but that seems to be the new style of discussion in c.s.y2k. The polly-come-latelies have been saying, "of course they knew, because.")

Did anyone notice the Internet outage on Friday January 28, between 2:00 PM EST and, well, later that night? Was that the weather too? If so, I'll add it to the Weather Report.

Eastabrook 2.5 or so in preps, 6 D-cells, tuna purchased at half price. When the ice storm came through last week, I had everything I needed already. I even popped open one of my half dozen MRE's to see what was inside. I found a teeny-tiny bottle of Tabasco sauce. There was other stuff, crackers, candy, a plastic fork.

The snow on Tuesday was a surprise so there were no problems at grocery stores. This Sunday's light dusting of ice is another matter. This was announced on Thursday and Friday, so by Saturday, the TeeVee was filled with reports of traffic snarls caused by people shopping for preps, there were 2 and 3 hour long lines at the stores. (Reported on WBAL).

The oil, gas, natural gas, diesel price rise was caused by the weather and not computerfailures. Put it together, every year we have this thing called winter, the weather gets cold and like clockwork, every year about this time, mid-winter, there's a fuel shortage and gasoline shoots up in price. It's happened every year for the last 20, 30 years.

Doomers have small brains with weak memories so they forgot that last January and every January before, unleaded spiked up to $1.40 gallon and home heating oil goes to $1.85 when you can get it. Odd, my records show that I paid 84 cents/gallon for unleaded last winter. Odd too that propane is expensive and in short supply. Is propane made from oil?

They also forgot that Ft. Meade has a regular 3-4 day computer failure about this time and that DeeCee always gets its financial system in a tizzy, something about hiding all the winter holiday gifts for the Mayor's girls.

I've been sitting here, blankie pulled around my shoulders, paper bag of used Kleenix, still not at 100%, not that I'm ever at 100%.

Anyway, looks like the pollies were right. None of these problems happened. Rail is running perfectly. No problems with trains, no derailments, collisions, or missing freight. Planes aren't snarled, grounded in Australia, crashing other places.

We've got the usual winter layoffs right on schedule. Layoffs at Lockheed Martin, Amazon.com, ValueAmerica.com, just like normal.

The economy is fine. Makes sense for a bunch of dot coms to spend millions of dollars advertising on the superbowl. Lifeminders.com, register with them and every year in December, they'll send you an email reminding you to buy a present for mom. Now there's a valuable service.

One dot com, a forty person company, dropped 2 million bucks on an ad, I think they sell office supplies on the web. Whooo-weee, skinny- minnie, I'm sure Quill and Office Depot are scared. Quill has terrific prices and fast delivery on common office supplies. I got vue 3 ring binders from Office Depot for a buck on Saturday, they had 'em on sale.

No unusual problems. Stock market's down a little, but that's a buying opportunity. The pollies were right; you'll see, any day now. Things'll be fine.

Some goo-ood information on the casino called the NASDAQ, check out http://www.itulip.com, you can't win if you don't bet your 401(k). Every player is a winner, put your life savings down on a spin of the wheel.

CAUTION: itulip.com is the financial twin of the DC Y2K Weather Reports. It says "Parody" at the top, the authors have an edgy sense of humor. (they've decorated their home page with spinning heads showing a bald spot.) They're like the DC Y2K Weather Reports. They don't tell you what to think, like the pap called the popular press. They show you what's under the hood and give you the power to assess the risks yourself.

This week, I was lunching with a geek-bud, he whispered, I've been thinking, the stock market could crash and not recover for a long time. This guy is in his late 30's, has an advanced degree in software engineering, about a quarter million dollars in his 401(k), minimal debt, a $350K house in Fairfax.

He said, there's nothing behind most of these companies. Why do people buy them? AOL, Yahoo, Amazon.com, Cisco.

Badboy that I am, I pointed out that his 401(k) is likely invested in the very companies that he was panning.

Cisco is an interesting case. They have real products and a near lock on a critical piece of technology, today. It's not clear that they will dominate optical networking, nor is it clear that Internet usage will continue to grow.

Last week, as part of my NXTract effort (http://www.kiyoinc.com/nxtract.html), I ordered Oracle for Linux, try 'n buy. (http://www.oracle.com) For less than 40 bucks, I got everything I need to run Oracle on Linux. Webtools, DB clients, etc. If I find it useful, I can purchase a commercial license. I could have downloaded the files but I don't have a DS3 so I eCommerced the order.

40 bucks, a couple days later, UPS delivered a box with 11 CDs. Who made out?

UPS made maybe five bucks, maybe 10 tops.

The CD pressing house made another 10, maybe 15.

couple bucks for packaging,

Oracle marketing soaked up the rest.

The portion of the eCommerce that went to Cisco and all of the Internet infrastructure was down in the pennies.

Betting on Cisco makes no sense at all.

Oh but it's a fundamental part of the Internet infrastucture. Well, fiddle-de-dee. What about Bell and Gossett and American Meter Company? My basement is full of Bell and Gossett valves. These control the flow of hot and cold water. A couple years ago, the gas company installed an electronic, radio transponder gas meter made by American Meter Company. They'll be back in a couple years to service and renew the meter.

Water and gas are the real infrastructure. Bell and Gossett and American Meter Company play the same role in water and gas that Cisco plays in the Internet.

You've seen Cisco ads on TeeVee. I understand Pepsi and McDonalds ads but Cisco? Who are they selling what to? No one will be influenced by a Cisco ad to buy a router. Every one of the few thousand real network engineers in the country already has decided on Cisco or whoever. Cisco advertises on TeeVee to sell you their stock.

Go into your basement and look at the things down there. Multiply that by a hundred million households. That's the real infrastructure.

Fuel prices, odd little problems; the heaves and lurches of the mainframe, back office world will continue for another 6 months or so. Right now, it's running about an Edwards 3, code-heads cranking to get the reports fixed before the problems affect customers.

The aggregate of the errors is not enough to tip over the economy but I expect more price rises and pockets of unemployment. I was prepped for an Eastabrook 2.5. I'll be looking for ways to prepare for continued economic problems.

I expect continued labor shortages for technical specialists and swings in the stock market. I still believe that there is a 5-10 percent probability of a rerun of something like 1929.

Consider January and February to be a lucky break. I'm not complaining that I have a bunch of cans of tuna purchased at half price. I haven't broken open the packs of 6 Alkaline D-Cells yet.

If this calms down to an Edwards 2 or lower, we'll have a terrific time at Paul Milne's party in southern Virginia. I'll help out with it.

cory hamasaki

--Members Only--

Given the good outcome to date, we will scale back the WRPs to a monthly newsletter. We'll continue to operate the listserv at dc-y2k- WRP. This is a really nice group that discusses family, caring relationships, gardening, independent living, etc.

As I receive information on problems, solutions, and items such as near free Linux, Oracle, and Sun products, I'll pass these on through the WRPs or the members only site.

I still owe several people pocket knives, LED kits, and gardening books. Please drop me an email if you asked for these and I haven't gotten them to you yet. My direct mail address is HHResearch Co., 3503 Norris Place, Alexandria, VA 22305.

I hope that everyone will take advantage of our collective good fortune to prepare for a better, safer, and more sensible world.

Fine Print -- Subscriptions --

You don't have to subscribe to the WRPs, you can keep reading the issues on the web or in Usenet or however it gets to you. For most people, subscribing is a way of helping spread the world. I've used the subscription money to print thousands of copies of the WRPs.

2000 Subscriptions

While contributions are always welcome, we have a new subscription policy. No additional charge - shareware memberships

No additional charge - small corporate

$5,000 - Fortune 5,000 or Government agency.

Fine print -- publishing information --

As seen in

USENET:comp.software.year-2000

http://www.elmbronze.demon.co.uk/year2000/

http://www.sonnet.co.uk/muse/dcwrp.html

http://GONOW.TO/Y2KFACTS

http://www.ocweb.com/y2k/weather.htm

http://www.vistech.net/users/rsturge/cory_launch.html

http://www.kiyoinc.com/current.html

Please fax or email copies of this to your geek pals, especially those idiots who keep sending you lightbulb, blonde, or Bill Gates jokes, and urban legends like the Arizona rocket car story.

If you have a webpage, please host the Weather Reports.

-- Hokie (Hokie_@hotmail.com), February 17, 2000.

There is a thread way below on this. For me, I'd have a lot more respect if Cory had simply said "I was wrong. The Baron sucked." I was a top end doomer and I still anticipate bubble.com to collapse so I'm glad I have what I have.

It seems a little late to say Webster was right and Infomagic was wrong and try to shine the whole thing on.

Todd

-- Todd Detzel (detzel@jps.net), February 17, 2000.

Thanks Hokie!

-- Carl Jenkins (Somewherepress@aol.com), February 17, 2000.

Cory isn't about to say he was wrong. Of course it's going to happen, it's just late. It's like all those other dates that didn't bring disasters. Just keep watching the news. It's happeing! Dont you see it????? Sure it is! And my electric lights are just figments of my imagination.

It's hard to say you're wrong. But I'm telling you I was Wrong. I'm not sorry about the food or the Berky, but the rest of my preps will be sitting here useless until the local Animal Sanctuary has a big yard sale this summer and I can donate this stuff to them.

Want to see a doomer nut! Step right up folks! Here she is! Get your bargains! Batteries 10 cents each, fruit jars cheap! Cheap! Lamp oil by the bottle, jugs of clorox, soap, crackers, rubber gloves, galore.

-- gilda (jess@listbot.com), February 18, 2000.