A lot of Y2K optimists (I'm going to eschew the label "pollyanna" as counterproductive to dialogue) have been baffled by anybody's continued gloomy outlook on Y2K, especially with all of the "good news" pouring in. Many have simply chalked it up to Y2K becoming a quasi-religious sentiment with us gloomers ("religious" being synonymous with "irrational" in our culture). This post is my attempt to present some of my own reasons for remaining quite gloomy, as well as try to bring some segment of this forum back on topic.

These are in no particular order:

1) Missed deadlines: First it was Dec 31, 1998 with a year for testing. Just about everybody missed that. Then it was March 31, 1999. Lots and lots of folks missed that. Now it's June 30, 1999 with six months for testing. Care to bet a significant percentage will miss that too? Here comes September 30, 1999 with three months for testing. This highlights the fact that much of the "good news" that optimists base their position on is still projected good news; that is, it is not that all these companies and agencies are actually done with their fixes and end-to-end testing but that they are "on schedule" to be done by year-end.

2) The peculiar nature of software project deadlines: Related to point 1 is the what many of the software people on this forum (including me) have experienced; software projects are historically "on schedule" until just before the deadline, at which point it becomes clear that they are far from complete. So it's not necessarily that anybody is lying about this; they may genuinely believe that are on schedule. In many cases that will be the honestly held position, up until about December at which time the shortfall will come out. There is simply no rational reason to believe that Y2K projects are different from other software projects in this respect.

3) Federal goonvernment situation: We have verified instances of agencies lying about their progress. We have the spectacle of an ever-shrinking list of "mission critical" systems which, despite their declining numbers, are still not fully remediated. This subset of systems now comprises somewhere between 7-9% of total federal goonverment systems. We have a long history of federal ineptitude with respect to getting work done on time and in budget. And the current administration (not to speak of past administrations) has demonstrated its ready willingness to lie flagrantly to the public. In short, if the goonvernment tells me my name is David Palm, I'm checking my birth certificate -- I don't believe anything they tell me anymore at face value. Yes, I think they're lying about Y2K. So, I expect some nasty shortfalls in fed systems. And the extent to which our society funnels its lifeblood through various goonverment bureaucracies has me concerned about this shortfall.

4) Bloating budgets vs. optimists' claims that the budgets were overblown at the start: The optimists hooted that projections like Gartner Group's projections for the U.S. were ridiculously high. Now individual budgets have doubled and sometimes even tripled. All of this tells me that companies and agencies underestimated the scope of the problem and the difficulty of fixing it. That in turns tells me that there is no good reason to believe that they will hit the deadline for the fixes; they waited too long to allocate resources: Brookes Law.

5) The foreign scene: In a word, bad. Japan is a huge concern to me. Y2K failures on top of existing economic instability looks like a real disaster to me in many countries. Even if the world avoids the rise of military conflict(s) out of the disruptions, I just can't see that we can avoid an economic shockwave that will slam us plenty hard.

6) The convergence of failures in 2000: This point specifically addresses the optimists' "we have problems all the time and take care of them" argument. Of course we do. But we are currently staffed (usually understaffed, actually) in IT and engineering to take care of those routine problems. We are not staffed nor prepared with contingency plans to handle a huge spate of virtually simultaneous and ubiquitous problems. It's the convergence of computer glitches that could overwhelm our ability to fix them in a timely fashion. Does the relative ease with which companies and agencies have handled past "trigger dates" ease my concerns? A little bit. But I have always thought that the significance of those dates was overblown; they affect too small a subset of systems and operations. The simple fact haunts me: a company that started six months to two years late fixing its problems is not going to be able to patch any glitches "over the weekend."

7) Software metrics: I still have not seen a really compelling case that the metrics for software projects in general do not apply to Y2K. Even if we discount the figures a bit because, at least in some instances, Y2K remediation is more like maintenance than development we are still faced with the statistical likelihood of a significant percentage of not-fully-remediated systems. Any other position is IMO truly that of Pollyanna (ahhh, I thought I could go this whole message without using that term!).

8) Anecdotal evidence: This is tricky but still important. Anecdotal evidence goes both ways, of course. There are people on this forum who tell us that their company (not named) is doing great. Good enough. Again, we expect good news and lots of it. The metrics don't say that nobody will finish, just that some significant percentage won't. But there are also people on this forum who tell us of significant shortfalls. And it's not the systems that are fixed that are going to bite us, it's those that aren't fixed. The anecdotal evidence still points to some serious failures.

9) Miscellaneous, non-Y2K factors exacerbated by Y2K: War, terrorism, loss of civil liberties to federal goonverment grabs, economic instability due to overvalued stock market/personal debt, collective loss of a moral compass. All of these add up to make me much gloomier about the national and world situation than I was a couple of years ago. And certainly Y2K instabilities easily contribute to greater risks in all of those areas.

All of that being said, I am more optimistic today than I was six months ago. As I have mentioned on another thread this is due primarily to my opinion that we are much less likely to have a total power grid collapse. That has me wondering whether the alt-energy system I purchased was a wise investment. But frankly I'm still expecting a serious depression of at least 1930s proportions. That's quite bad enough. Given where our country is at morally and spiritually we could start there and it could get quite a bit uglier than that.

So, "Depression 2000 and Beyond". That's my rosier but still gloomy picture. And those are my reasons.

Fire away.

-- David Palm (djpalm64@yahoo.com), May 14, 1999

Answers

Amazing. A well-written, reasonable, on-topic post. Amazing. :-)

Brooks's Law: "Adding manpower to a late software project makes it later" (The Mythical Man-Month). More memorable: "The bearing of a child takes nine months, no matter how many women are assigned." And equally applicable, "More software projects have gone awry for lack of calendar time than for all other causes combined."

-- Lane Core Jr. (elcore@sgi.net), May 14, 1999.

What I'm doing to mess up and exacerbate my damage-share of the coming chaos:

Getting out of debt (excepting mortgage) before December. Surely hope everybody else doesn't do this: we'd see an awful contraction prior to the rollover. Instead of spending 10K on new, job- creating products and services, I'm sinking it into non-productive debt repayment. A gloomy, tight-belted proposition, but only sensible.

Picking out the items I'll want to buy next year at intensely deflated prices, should I keep my job. This is downright opportunistic and probably immoral, but hey. I went thru the '80's Texas oil & bank bust. There are fine people who still haven't recovered from that. There are rich people who were fortunate to be highly liquid when we bottomed out.

I am going to do far more community work than I already do (not a whole lot) should my services be required, and I suspect they will.

Yeah, I'm real short-term gloomy, but extremely optimistic about the future. Y2K has shaken me out of the rat-race, lemming mentality: I'm taking my life into my own hands from here on out.

I've (finally!) learned the details of finance, economics & govt. that I've always ignored. I feel like I could test out of 60 credit hours, entirely due to Y2K research.

Almost like freedom for the first time.

(I'm still terrified about Russians freezing to death, though.)

-- Lisa (lisa@work.now), May 14, 1999.

Good summary. I would add to that

an overall exponential increase in computer system complexity that has been responsible for software being delivered progressively later

See

Complexity, Testing, and Y2K: Why some engineers still Don't Get It

-- a (a@a.a), May 14, 1999.

Thank you, Mr. Palm. I have searched for an essay that spells out the problem, & my own attitude towards it, clearly & completely; yours is the best. I am giving copies to all my "polly" friends.

-- too lazy to (write@it.myself), May 14, 1999.

David:

I agree with virtually everything you wrote. I believe you have summarized the problem well, and few could disagree with your statements (though they will, of course). Thanks.

The only point we differ on is your "serious depression of at least 1930s proportions" prediction. While I don't think it will be quite that bad, I have no way to know for certain. I predict long-term economic problems, but I hope that the unemployment rate, bank failures, etc. of the '30's don't materialize. Time will tell.

-- regular (zzz@z.z), May 14, 1999.

regular --

The only difference between the coming depression and the depression of the thirties is that this one will be WORSE! Let me explain why --

We weren't 6 trillion dollars in debt as a nation.

We weren't individually in as much debt as today.

We are a nation lacking the "pen and paper" skills to fall back on (as demonstrated by many postings and sites on the internet.) Ever try to get correct change from the moron at Taco Hell when the register fails?

We won't have the funds to start a CCC or WPA to get the economy moving again.

We are unable to be self-sufficient. How many people do you know that can farm, hunt or otherwise produce their own food these days?

Have a nice day.

-- AriZONEa (sorry@work.com), May 14, 1999.

AND it is unlikely a World War III will boon our economy and pull us out of depression. WWIII will probably leave the planet a smoldering twisted toxic mess of bio/chemo/radio-hazards.

-- Leska (allaha@earthlink.net), May 14, 1999.

David -- Excellent. I would add one more to the mix and that is the downplaying of actual preparation even BEYOND the silly stuff officially proposed. Think: the scenario you describe is a real possibility and, in the face of that possibility, the populace is rearranging deck chairs and dancing in the ballroom. That can only make the actual outcome major-ly worse than it had to be.

-- BigDog (BigDog@duffer.com), May 14, 1999.

Thank you.

I tend to ramble on and my posts wander. My convictions on this issue run pretty much along with yours. Well done!

-- Kristi (securx@Succeed.Net), May 14, 1999.

More reasons this depression will be worse than the 30's:

- Higher crime rate.

- Collapse of family values

- Much more dense population centers

But as Milne says, "Go ahead. Don't prepare." Then, as Cory says, in 5,722 hours it'll be "Squee-squeee-squeeel! I want canned eeel!"

-- a (a@a.a), May 14, 1999.

Actually I tend to agree with ariZONEa. As I mentioned here and also in the thread with Drew Parkhill on the stock market, I see so many terrible underlying economic conditions that Y2K could easily create a situation worse than 1930s. I agree also with "a" that other social factors make the prognosis much worse.

I'm going to cross-post this on some of the anti-doom sites and see that they have to say.

-- David Palm (djpalm64@yahoo.com), May 14, 1999.

David ... excellent job!

Quick sound bite ... RISING PRICES.

The KPIX Channel 5 CBS evening news (http://www.kpix.com) just mentioned that the S.F. Bay Area has experienced a 45.8% rise in gas prices in the past 2 months!!!

*Big Sigh*

See what shortages can do?

Wonder what just ... Y2K fuel shortages ... and resulting higher cost/ prices will do to local economies?

(Naw. Global shortages wont have a local economic impact, will they?)

I see Y2K as more catalyst, to exacerbate pre-existing problems, than prime mover.

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), May 14, 1999.

David - a truly excellent summary IMHO.

However, a few more things to add to the mix:-

The very real danger of "poisonfire" - accidental/deliberate release of nuclear/chemical/biological hazards into the environment ... - think Chernobyl, Bhopal, Three Mile Island, Porton Down in the UK...

Government spin - dissuading folks from preparing, actively IGNORING y2k (our internet inventor Gore (and Billy Jeff) - heard from him on y2k lately??)

Government spin [Greenspan, the Fed, FDIC] on the dangers to our financial systems - if these tank the civil unrest will make the aftermath of the Rodney King trial look like a Disney(tm) production...

GREED - JQ Public is basically greedy, apathetic, spoiled, lazy, ill- informed...

Uh, that's all for now :)

-- Andy (2000EOD@prodigy.net), May 15, 1999.

!!!!!!!!!!

Well said !!!!!!!!!!

Damn !!!!!!!!!!

Oh, shit man !!!!!!!!!!

<:)=

PS - !!!!!!!!!!

-- Sysman (y2kboard@yahoo.com), May 15, 1999.

Sorry, forgot the comma, it just doesn't sound right...

Oh, shit, man!!!!!

<:)=

-- Sysman (y2kboard@yahoo.com), May 15, 1999.

this post will be entirely ignored by the "bump-in-the-road"crowd

-- zoobie (zoobiezoob@yahoo.com), May 15, 1999.

ANdY!!!!! YoU FOrgOT TO MentION ThE COmInG REPtiLE alIEn invasION!!!!! wHY The ovERsiGHt??????? iT WiLL Be a fACTor, WiLL it nOt??????

-- Dieter (questions@toask.com), May 15, 1999.

Yes, Dieter, the coming reptile alien invasion will also be a negative influence on TWAWKI. Thanks for not letting us overlook it.

-- Nabi Davidson (nabi7@yahoo.com), May 15, 1999.

Yep. That's pretty much how I see it will go. Very astute (naturally because I agree with you)... :)

But I don't discount the possibility that things could end up working out for the better. If we all generally stick together, keep cool heads, work our butts off, elect prudent and practical leaders, bite the bullet, tell the truth, try to get it together, and be kind to our neighbors, we'll end up with merely a "bumpy road"--ie., the unpaved kind (no not a single "bump in the road" or 72h hurricane), with few people hurt but all of us inconvenienced. I mean an initial mild inventory recession combined with later spillover from severe global recession, producing a breakdown of JIT a la Yardeni's predictions for a year or so. Then we get a huge resurgenence of the New Economy a la _Wired's_ "Long Boom" article from a few years back.

But it could also be worse. Much worse. If we've been lied to, people WILL panic. If we continue to be perceived worldwide as the #1 Asshole of the Globe, one day our actions will come back to haunt us like spitting a huge hocker into a gale force wind.

I don't like the way the idiots in our government are ruining our future. Do you think they can try any harder to START WORLD WAR F***ing THREE?? (Or at best provoke a few stray terrorist bio-nuke-cyber-chem attacks on major population centers...?)

I will rest a bit easier when or if 2000 comes and goes with little pain. But not much easier. Sometimes I think a bad domestic crisis is exactly what we need in order to return us to the more sensible taoist, live and let live, less-imperial values of the founding fathers--minus of course their slavery, their denial of the womens' vote, their ethnic cleansing of the Indian and so on...A Y2K crisis, I'm ashamed to say, might possibly teach us all a badly needed lesson in humility, community, and frugality. But seeing that this babbling monologue is straying farther and farther away OT, I'll shut my mouth now.

-- coprolith (coprolith@rocketship.com), May 15, 1999.

Dieter,

Do lizards in the backyard count?

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), May 15, 1999.

David, thank you for having the guts to post your courteous and thoughtful essay on the BFI forum. Predictably, this was the response from a major player at Der Boonkah:

I for one, don't much care. You are entitled to try to re-inforce your meme by infecting others.

Forum: Gary North is a Big Fat Idiot Forum

Re: Why I am Still Gloomy (David Palm)

Date: May 14, 21:59 From: cpr

MESSAGE:

SHORT and sweet: Year 2000 is OVER as a business/computer problem. Its now even downhill has far as the hysterics are concerned. Daylight is coming all over the place and soon it will be obvious even to the "Clueless General Public" Yourdon, Hyatt and North PREY UPON.

LONG:

You can look for all the "reasons" to continue to justify your beliefs all you want but you, like Yourdon, are WRONG.

FLAT OUT WRONG.

Because you buy YOURDON's CENTRAL FALLACY that he so incompetently defends in his silly "Deja Vu";..... you buy the rest of the NONSENSE.

His Central Fallacy is what leads you astray.

He continues to insist that Y2k is a software project and use the metrics of Capers Jones plus his annoying and overbearing and never ending "I am the 30 year Expert and I do not expect a Real Estate Broker in Texas to understand this".

WELL HE IS WRONG ON THAT ALSO. I DO UNDERSTAND.

He neither knows nor cares about my background but ANYONE CAN SEE HIS ERROR: Y2k is **not** a software project: it is "maintenance". That has proven to be the case, he knows it and still he MISLEADS OTHERs.

Bill Urlich makes a big deal about 'residual errors' as does Yourdon. BUT LOOK AT URLICH's NUMBERS: 100 /errors in a million remediated LOCs. "Now that the results of independent code audits are coming back, organizations are finding on average about 100 date-related errors per million lines of 'fixed' code," says William Ulrich, president of Tactical Strategy Group Inc. of Soquel, Calif. .

THAT is going to end the world?? THAT IS 1/10,000 LOC ....OUT OF WHAT 10-30% OF THE STARTING LOCs OF THE OVERALL PROJECTS??

AFTER REMEDIATION.....?? WHICH ASSUMES THAT SOMEBODY INVENTORIED AND ASSESSED BEFORE SELECTING THE CODE TO BE REMEDIATE??/

BECAUSE THAT WOULD GIVE WHAT??? A BLUE PRINT OF WHERE THOSE STUPID 100 "RESIDUAL ERRORS" ARE!!!! AND IF I HAVE A WORKING INVENTORY BLUE PRINT OF THAT I CAN WHAT??????????????

DEBUG IT PRETTY FAST??

YES OR NO??

ANSWER IS OBVIOUS BUT YOU AND YOURDON ARE WRAPPED IN THIS "MYSTICISM" OF THE "IMPOSSIBLE JOB AHEAD OF US" MANTRA OF THE LOSERS OF THE WORLD YOURDON CATERS TO.

As for "ITS SYSTEM" that's crap also. I'll leave that for COBOLLER: http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=000pbw

NOW, all YOURDON&TOAST.ED has left is his "demands" for IV and V and even that few are listening to.

He has managed one great Y2k Project. I will give him credit for that.

In less then 3 years, he has managed with great skill to completely destroy his own reputation in I.T.

And I trust the Industry never forgets it nor the PANDERING to the fears of the Technically un-enducated. In that regard, Yourdon is worse and far worse than North whose goals are rather clearly stated.

I called Yourdon's bluff and I continue to call his bluff.

NOW, I am not alone and that is reward enough for me: to see Edward Yourdon continue to make a complete fool of himself.

IF....you wish to join him that is your right. DO not think you will INFECT OTHERS with his disease of Fear Uncertainty and Doubt (FUD) or his incessant ramblings about that which he has no expertise from Finance to now: World Affairs re: his view on Kosovar.

Y2K IS.......REMEDIATION. A FIX. IT IS NOT A DESIGN PROJECT FROM SCRATCH. CAPERS JONES' METRICS OF PROJECTS ARE USEFUL FOR ENTERPRISE "DESIGN AND BUILD" PROJECTS OF THE PAST.

DO THEY APPLY TO "CHANGING DATE FIELDS AND LOGIC IN CODE??" We need not even mention anything about the Spurious hoax known as "embedded systems" and those billions of chips he and others moaned about.

I will restrict this to SOFTWARE:

GET THIS......YOU... AND YOURDON ONCE AND FOR ALL:

Y2K WAS AND IS A "FIX UP JOB". CLEANUP.

PROOF: ?? WHO NEEDS IT?? YOUR ADJUSTING DATE FIELDS AND LOGIC OF PREVIOUSLY BUILT PROJECTS. THAT IS "INSPECTING A BUILDING FOR FLAWS FOR RE-SALE" IT IS .....NOT...... NEW CONSTRUCTION FROM THE DIRT.

CAN THAT BE SAID ANYMORE CLEARLY???

THOSE METRICS DO NOT APPLY TO Y2K WORK. THAT IS WHAT IS NOW BEING DEMONSTRATED.

ALL THE REST OF IT IS OUT AND OUT "POLEMICS" TO SUPPORT YOURDON'S CENTRAL "FALLACY" AND THAT IS WHAT IT IS.

FALLACY.

AND IF A BUSINESSMAN IN DALLAS CAN SEE IT.....I FULLY EXPECT THE DAY WILL COME THAT ONE DAVID PALM DOES ALSO.

Projects slip because humans do them. When they "must" be finished, they get finished.

We now know from one report this week that 24% of the I.T. people surveyed reported "FINISHED PROJECTGS". It is now MAY.

Game time is 1/1/2000. The "mantra" for 1997 was "finish code remediation by end of 1998 to leave one full year for testing".

-- OutingsR (us@here.yar), May 15, 1999.

Begging your pardons, OutingsR failed to make clear those are the views of an optimist from the camp dedicated to saving us from our selves.

-- OutingsR (us@here.yar), May 15, 1999.

Wow!

There is one Fearful, Uncertain, and Doubtful (FUD-e-dudy) poster!

Wonder if he/she ever reads the United Nations web-site? Or downloads a GAO pdf file reports, or focuses on the Senate testimony, or checks out a 10-Q filing.

Naw. Probably reads the Y2K newsprint at the bottom of a messy birdcage. You know ... the one with the caged golden eagle.

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), May 15, 1999.

CPR is kind of interesting. If you omit all the ranting and attacks against Yourdon's observations, you are left with the statement that Bill Urlich has found 100 date errors per million of fixed code. OK, that's pretty good. Not great, but not bad either. If those errors are minor, then Urlich's customers will be in good shape (at least for the code Urlich examined). If they are serious errors, at least they have been found and will presumably be fixed.

Now, IF Urlich had examined ALL the code being remediated worldwide, and IF his examination finds ALL the errors in that code, we could all sleep easy. Perhaps CPR would like to speculate on just how much code has been submitted for IV&V, and why some IV&V outfits are finding higher error rates. And maybe CPR might like to guess about the reasons why so little code has been verified (maybe the inhouse guys are so good that the effort isn't worth the money?)

What CPR provides is ONE little piece of debatably positive news. How he can decide that this error rate is so small as to be meaningless, and how he can extrapolate this to ALL other code, is a mystery. I doubt that anyone not willing to make these leaps of faith is 'infected' with anything more serious than common sense.

-- Flint (flintc@minspring.com), May 15, 1999.

That non-argument from CPR pretty much confirmed my opinion of Biffy. That's the last time I'll stop by there. I think the religious argument cuts the other way with those folks; it's the audacity, yea blasphemy, to question the integrity of The System that irks them. "The System is strong, the System will never fail, blah, blah, blah..."

But hey, c'mon Flint, Poole, Decker, Paul D., et al. Wazupwidat? That's not the best you guys can do, is it? No answer to this at all?

-- David Palm (djpalm64@yahoo.com), May 16, 1999.

Yeah, talk about your "memes",

"The system is all, it can not fail, the system is always right, it will flex but never fall, all hail the system!"

HIZZAH!!! HIZZAH!!! HIZZAH!!!

-- Unc D (unkeed@yahoo.com), May 16, 1999.

Let me see if I understand the person refered to as CPR. His (her?) contention is that (paraphrasing) Y2K fixing is not as hard as writing code - it is mere maintanance, any errors generated by the fix can be "DEBUG IT PRETTY FAST", which he (she - though I'll assume he is male since women don't usually have to SCREAM to make a point) summarizes as "Y2K WAS AND IS A "FIX UP JOB". CLEANUP."

Ie., Y2K fixing is easy. If this is the case, why did S.S.A. which started in 1989, not finish in the early 1990's? Why are all these corporations that projected "fix by 12/31/98 with a full year of testing" miss that deadline as well "Y2K Deadline II: The Sequel" of 3/31/99?

If this "ain't rocket science", why doesn't it fly?

-- Ken Seger (kenseger@earthlink.net), May 16, 1999.

I'm not a septic Unc, but isn't it HUZZAH!!! x 3

-- Andy (2000EOD@prodigy.net), May 16, 1999.

Ken,

CPR is indeed a male. And you're right, females seldom do so much handwaving and screaming to cover the fact that they have no case.

Flint, of course, nailed CPR's "argument" completely. Ulrich reviews a few of his clients, finds such-and-such a percentage of residual bugs (and if you read his article you'll find that he's quite disturbed by it) and CPR declares VICTORY (all caps in honor of our esteemed real estate salesman). One man's tiny perspective of residual bugs and CPR calls it a Universal Truth. Better stick to real estate, buddy. Sheesh.

Unanswered. I would say that pretty adequately sums up the response from the "optimists". It's enough to make me start calling 'em Pollyannas again. At the very least it seems that they aren't really interested in engaging the central issues.

Ok, this is at the top again. If nobody engages the argument then the thread dies a victorious death (but who actually wants to be right on this?).

-- David Palm (djpalm64@yahoo.com), May 17, 1999.

Well, not wanting to disappoint one David Palm, let's have a go.

1) Missed deadlines: First it was Dec 31, 1998 with a year for testing. Just about everybody missed that. Then it was March 31, 1999. Lots and lots of folks missed that. Now it's June 30, 1999 with six months for testing. Care to bet a significant percentage will miss that too? Here comes September 30, 1999 with three months for testing. This highlights the fact that much of the "good news" that optimists base their position on is still projected good news; that is, it is not that all these companies and agencies are actually done with their fixes and end-to-end testing but that they are "on schedule" to be done by year-end.

This argument around Dec 31, 1998 has been brought up numerous times. For myself, the line "and a full year to test" definitely implies that those who used this line meant code remediation, not that they would be completely done. And we have virtually no way of knowing if this type of deadline was missed or not. I seriously doubt most companies will create large fanfare about making an intermediate project deadline.

I've had a real problem with those that expect completion prior to a June/July timeframe. Once complete, the systems will in essence be "frozen", and even 6 months is quite a long time to expect a major corporation to not allow modifications. A year, at least in my mind, was totally out of the question.

2) The peculiar nature of software project deadlines: Related to point 1 is the what many of the software people on this forum (including me) have experienced; software projects are historically "on schedule" until just before the deadline, at which point it becomes clear that they are far from complete. So it's not necessarily that anybody is lying about this; they may genuinely believe that are on schedule. In many cases that will be the honestly held position, up until about December at which time the shortfall will come out. There is simply no rational reason to believe that Y2K projects are different from other software projects in this respect.

The assumption here is that the deadline is Dec 31, 1999. Outside of a very few exceptions, I've seen none that have set the end of the year as the project deadline. Your own use above of projected deadlines illustrates this fact. Assume, for the moment, that the past metrics strictly apply (see below for more discussion). 20% will miss the June 30th deadline. But how many of those will complete in the remaining months? This is probably one of the major fallacies in the use of past software metrics; applying them to an assumed 12/31/99 deadline, as opposed to the actual deadline the individual companies are working towards.

3) Federal goonvernment situation: We have verified instances of agencies lying about their progress. We have the spectacle of an ever-shrinking list of "mission critical" systems which, despite their declining numbers, are still not fully remediated. This subset of systems now comprises somewhere between 7-9% of total federal goonverment systems. We have a long history of federal ineptitude with respect to getting work done on time and in budget. And the current administration (not to speak of past administrations) has demonstrated its ready willingness to lie flagrantly to the public. In short, if the goonvernment tells me my name is David Palm, I'm checking my birth certificate -- I don't believe anything they tell me anymore at face value. Yes, I think they're lying about Y2K. So, I expect some nasty shortfalls in fed systems. And the extent to which our society funnels its lifeblood through various goonverment bureaucracies has me concerned about this shortfall.

I'll assume for the moment you are not talking of the FAA as your verified instance. I can't remember one; care to clarify? If there are, it is because the government found them through audits; i.e., the system is working.

Personally, I think the process the government is going through for Y2k could be a very good thing. My feeling is anything that causes a review of the bloated bureacracy to determine what truly is essential is a positive.

4) Bloating budgets vs. optimists' claims that the budgets were overblown at the start: The optimists hooted that projections like Gartner Group's projections for the U.S. were ridiculously high. Now individual budgets have doubled and sometimes even tripled. All of this tells me that companies and agencies underestimated the scope of the problem and the difficulty of fixing it. That in turns tells me that there is no good reason to believe that they will hit the deadline for the fixes; they waited too long to allocate resources: Brookes Law.

There is quite a bit of evidence that some companies at least are being grossly overcharged for Y2k work.

My personal experience is that Y2k budgets contain a large amount of slack, in particular redundantly addressing systems. In general, using budgets, particularly % of budget spent, is almost useless as a gauge to project status. Budgets will contain personnel costs throughout the coming year, with particularly high head-counts in December-January, as a contingency for fixing problems. I'm no accountant, but there are multitudes of ways of allocating costs, especially in the area of capitalized purchases, such as hardware, etc. Also, balloon payments to outside firms where used is another factor.

As for Brookes Law, its application depends in large part on having the project fully staffed to begin with, and the current phase of the project. There are definitely areas and opportunities within a project where adding staff does, in fact, reduce the time. No, not directly linear, but a reduction all the same.

5) The foreign scene: In a word, bad. Japan is a huge concern to me. Y2K failures on top of existing economic instability looks like a real disaster to me in many countries. Even if the world avoids the rise of military conflict(s) out of the disruptions, I just can't see that we can avoid an economic shockwave that will slam us plenty hard.

I'm definitely no economist, so I tend to stay away from predictions. But my own, admittedly simplified view is that a large window of opportunity will open for those that have, in fact, addressed Y2k. Call it the whole "Flight to Quality" theory. I'm sure we'll feel an economic effect, just have now way of estimating it. We (the US) seem to have survived quite a few "foreign meltdowns" in the past.

6) The convergence of failures in 2000: This point specifically addresses the optimists' "we have problems all the time and take care of them" argument. Of course we do. But we are currently staffed (usually understaffed, actually) in IT and engineering to take care of those routine problems. We are not staffed nor prepared with contingency plans to handle a huge spate of virtually simultaneous and ubiquitous problems. It's the convergence of computer glitches that could overwhelm our ability to fix them in a timely fashion. Does the relative ease with which companies and agencies have handled past "trigger dates" ease my concerns? A little bit. But I have always thought that the significance of those dates was overblown; they affect too small a subset of systems and operations. The simple fact haunts me: a company that started six months to two years late fixing its problems is not going to be able to patch any glitches "over the weekend."

Again, another fallacy is this "business as usual". No company that I am aware of is planning on handling the rollover as just another day. I fully expect an "all hands" approach to the rollover. From observation, maintenance staff is usually less than 50% of IT resources, the others being in development. Having those resources available increases the staffing level.

7) Software metrics: I still have not seen a really compelling case that the metrics for software projects in general do not apply to Y2K. Even if we discount the figures a bit because, at least in some instances, Y2K remediation is more like maintenance than development we are still faced with the statistical likelihood of a significant percentage of not-fully-remediated systems. Any other position is IMO truly that of Pollyanna (ahhh, I thought I could go this whole message without using that term!).

The most cited source of these metrics, Capers Jones, has no problem discounting these metrics; why do you?

In his paper, Probabilities of Year 2000 Damages, he very clearly makes the point that "the year 2000 problem is primarily a code issue and the number of year 2000 defects are smaller than the numbers of all bugs", which causes him to disregard the metrics related to various other stages of a development project, such as Requirements and Design.

8) Anecdotal evidence: This is tricky but still important. Anecdotal evidence goes both ways, of course. There are people on this forum who tell us that their company (not named) is doing great. Good enough. Again, we expect good news and lots of it. The metrics don't say that nobody will finish, just that some significant percentage won't. But there are also people on this forum who tell us of significant shortfalls. And it's not the systems that are fixed that are going to bite us, it's those that aren't fixed. The anecdotal evidence still points to some serious failures.

Yes, recently we've seen a rash of postings regarding failures. How many turn out to be fact is anyone's guess. I know what my estimation is, but it's just my opinion.

9) Miscellaneous, non-Y2K factors exacerbated by Y2K: War, terrorism, loss of civil liberties to federal goonverment grabs, economic instability due to overvalued stock market/personal debt, collective loss of a moral compass. All of these add up to make me much gloomier about the national and world situation than I was a couple of years ago. And certainly Y2K instabilities easily contribute to greater risks in all of those areas.

Again, all part of everyday life. Things to consider, for sure, but some of the conspiracy theories being propagated are, let's just say stretching the envelope.

-- Hoffmeister (hoff_meister@my-dejanews.com), May 17, 1999.

Good points David - thank you.

Everybody: QA me on these notes from the "loud" guy in Dallas - see if you can point out any flaws in these observations:

1) He is using a reference from a programmer doing verification on remediated systems, right? So these things are already "repaired" - that is, they have had the year 2000 problem supposedly already completely eliminated, and the program is released for use, but was sent out for verification. So why is he (the poster) HAPPY to see "100's of (residual) bugs in millions of lines of (remediated) code" - isn't this exactly the 3-7% percent of "new" errors found in projects after a change is made? Sure the fractions aren't 0.03; but this is IV, not alpha or beta release software - there should be NO errors if it were tested thoroughly!

No errors is of course unlikely (impossible) but you get the point - remediated software should not have errors - so why are they present?

2) How many errors were present in the original programs? What will be the impact of using the un-remediated programs on the company? Why has no company - ever - tried blind fix-on-failure to simply set its dates ahead and just fix (now) what breaks?

3) No where in his screams is any comment about what would be the result of errors and y2K failures left in systems - to be blunt, I don't care about y2K failures that don't affect me, my family, my business, my community, or any company in any supply chain I'm involved in.

But my judgement is: there will be many millions of failures, and some of them are going to be fatal (shutting systems and processes down) that will cause widespread economic and personal impacts worldwide.

4) In his rantings, he is willing to make any assumptions about all sorts of things except the probability of failures - that is, he discounts any chance that things will fail, much less how many, how long, where at, and to what magnitude, and then what impact if/when they do fail.

His answer, apparently, is that he just wants to scream "THEY WON"T FAIL" at the top of his lungs, but give no reason for assuming that.

5) Has he run a business or fabrication plant? Has he tried to build and ship prodducts other than real estate? Has he ever relied exclusively on received goods to stay in business? (A gas station, for example, requries two or three deliveries a week - if they don't get gas, or don't have credit card access, or dn't have power, water, and phone service, hey can't stay in business long.

An auto repair shop absolutely needs the same thing to get parts to finish the jobs they have in - house. Sure, they can stay open a while, but they can't make a profit (or even see under the hood) if they have power outages.

-- Robert A. Cook, PE (Kennesaw, GA) (cook.r@csaatl.com), May 17, 1999.

Re: CPR's hysterical histrionics. (I suggest a long vacation for CPR.) I don't know about the rest of you programmer/analysts, but it has not been my experience that maintenance projects come in on schedule any more than new development comes in on schedule. I can only recall TWO pure development projects in my nearly 20-year career (I can recall them because they stand out because they were new development, not maintenance or enhancement; one on a legacy system in VAX BASIC, the other in Visual Basic under Windows 3.1), so I'm talking an awful lot of man-years of maintenance, what with me and my colleagues.

SHOUTING that Y2k remediation is MAINTENANCE not DEVELOPMENT strikes me as just plain ignorant. But maybe my own experience has been atypical. (I think not.) Where is the evidence that maintenance projects, by and large, come in on time?

-- Lane Core Jr. (elcore@sgi.net), May 17, 1999.

Thanks Hoff for a reasoned reply. I have started a point-by-point response but thought I'd post this as a pretty good answer to your central thesis. From the NY Times (also posted on another thread):

<< The largest companies in the nation continue to fall behind their schedules for Year 2000 repairs, and most suspect that their budget estimates for the remaining work are too low, according to a survey in April that was the latest in a closely watched series that began in 1994.

About 22 percent say they do not expect to have all of their critical systems tested and ready to adjust when the clock ticks over to Jan. 1, 2000. That is up from 16 percent in November and 12 percent last August....

About 85 percent of those polled said that spending would have to rise beyond current estimates, mostly to support more testing and the creation of emergency command centers and other contingency plans aimed at keeping business running despite any Year 2000 glitches. Projects are slipping past their expected deadlines at 92 percent of the companies. >>

Whoops!

-- David Palm (djpalm64@yahoo.com), May 17, 1999.

Yes, David, I saw the article.

More interesting, perhaps was the final paragraph, where 75% felt that Y2k was going to give them a competitive advantage. Says something for how they think Y2k will turn out, wouldn't you say?

-- Hoffmeister (hoff_meister@my-dejanews.com), May 17, 1999.

How they HOPE Y2K will turn out, Hoff.

I'd hope that too. Just not very likely, from a global perspective.

However, entrepreneurs will always find opportunities, and companies will always fail. Just how many and for how long ... this time around ... and at what cost ... is the question?

Diane

-- Diane J. Squire (sacredspaces@yahoo.com), May 18, 1999.

See, what they (the one's who are actually completing remediation at or near their schedules) are hoping for is that the other companies; (and the smaller competition) who aren't getting ready in time; who aren't going to finish at all, or who are doing a shoddy job of remediation and testing will crash and burn.

Thus, the ones left over have a stronger, more update IS suite, more reliable programs, and dead competition.

-- Robert A. Cook, PE (Kennesaw, GA) (cook.r@csaatl.com), May 18, 1999.

Of course, David, I never thought I could reduce your gloom. At this point, and after reading some of your other articles, that was a foregone conclusion.

You overstate the point when saying 22% admit they will have mission critical failures. This view stems in part from considering a companies Y2k project an "all or nothing" affair. It is hard to evaluate without more specifics, but we have no way of knowing just how many systems will not be ready. Some of these may be systems that will not require processing until after the rollover. And, a very good educated guess is that it is exactly those systems that will be the focus of contingency plans. No, for the processes those systems encompass, the efficiency will not be the same. But it does not follow that mission-critical failures will occur by necessity.

As for citations that companies are being overcharged, my usual answer would be any that have contracted with the Big Five (4? 3?) to do the work. I could dig up some links to the Florida case from a couple of months back, as well.

I suspect expanding budgets are caused by many reasons. Some portion due to underestimating work. Other factors are probably expanding contingency plans, and part may be pure money grabs by IT departments, given a basically "free reign" on spending for Y2k.

-- Hoffmeister (hoff_meister@my-dejanews.com), May 19, 1999.

<< Of course, David, I never thought I could reduce your gloom. At this point, and after reading some of your other articles, that was a foregone conclusion. >>

Yeah, sure Hoff. I just have my Y2K blinders on. Never give any rational reasons for my positions. That's the trouble with us gloomers, isn't it? ;-D

-- David Palm (djpalm64@yahoo.com), May 19, 1999.

CPR, the real estate agent that thinks he knows more about software projects than Ed Yourdon, said "Y2K is only a maintenance project".

Y2K is no ordinary maintenance project. It has already been cited by Capers Jones that the metrics for y2k work may turn out worse than new development. Also keep in mind that programmers will sometimes completely rewrite unfamiliar code rather than try and figure out how it works. It can be quicker in the long run.

-- a (a@a.a), May 19, 1999.

'a', Capers Jones discounts the existing metrics, to allow for the fact that Y2k is in essence maintenance work.

David, mine was just a general comment. People who post thoughts, information and analysis, are generally looking to engage in a discussion. Your post claiming victory, because no one apparently addressed your post, makes it clear you were merely looking to win an argument.

-- Hoffmeister (hoff_meister@my-dejanews.com), May 19, 1999.

Hoff: the argument goes (IIRC & paraphrasing):

Since companies that have allowed their software infrastructure to fall victim to y2k in the first place are generally incompetent, their remediation metrics (technicallly classified as s/w maint) will be poorer than new developement work performed by the average programming shop.

-- a (a@a.a), May 19, 1999.

I like some of your arguments, David, but I like some of Hoffmeister's responses more. I particularly agree with the "completion of remediation work in December of 1998 with a full year to test."

Here's why I agree with it: As Hoff pointed out, companies do NOT say anything when their remediation is complete without having first tested and moved into production. Therefore, we have NO CLUE if the companies that stated they'd be done with remediation at the end of 1998 were actually 100% done, had two programs left, 100 programs, left, etc. What I HAVE heard however, (on this forum) is the ASSUMPTION that the projected dates of total completion (including testing) slipping beyond original estimates indicate that what's actually slipped beyond the estimates is the remediation itself rather than the testing. Posters then go on to state things like "leaving only 3 months to test, leaving only 6 months to test", etc. In actuality, the remediation may very well have been completed at the end of 1998 and testing has ALREADY been done for 3 months, 6 months, etc. Do you see the difference?

I must also agree with Hoff on another point: Once remediation is complete, the system will be frozen. We're seeing that already, Hoff. I have yet to E-mail the Cobol Dinosaur about it, but we've heard from contractors nation-wide that the work has dried up. A recruiter friend states the same thing. We believe it to be simply BECAUSE the systems are frozen right now. Of course permanent positions are available on non-Y2k work, and some we know have decided to accept positions simply because they need the money. Others are in a position to hold out longer and see if firms finally say, "I can't simply wait on these other 2000 maintenance requests until after the Year 2000."

I had to laugh at a FoxPro contractor friend who recently told me that 90% of their work of late was Y2k. She said, "Have you been sucked into Y2k work yet?" How do I gracefully say, "Yes...5 years ago I was, but it seems to be either done or already contracted out at THIS point." Of course Cory has that project of 400 contractors just contracted in April in D.C., but we're not seeing that nation- wide.

This was a great thread. I enjoyed the read.

Anita

-- Anita Spooner (spoonera@msn.com), May 19, 1999.

I gotcha, Hoff. Actually, I was being facetious with the "victory" thing; I should have put it in quotes. I really don't want to be right on this. But I think I am right and I'm the kinda guy who argues his position vigorously with the intent to persuade.

Anita,

I just don't think your position squares with the facts, most recently expressed in the NYT article. And you, like Hoff, are really only addressing a small segment of my entire case so you folks just don't sway me very much.

I'm enjoying the thread too.

-- David Palm (djpalm64@yahoo.com), May 19, 1999.

Thanks Dave, a real pleasure to read all your comments, not plesant, but very effective. As a senior systems analyst, catholic and son of parents who suffered greatly through the great depression, and grand parents who suffered through the wars I agree. Your honestly will be blessed with your comments becoming truthfull. Justthinkin

-- justthink com (y2kaok@justthink.com), May 19, 1999.

Anita: Cap Gemini just released figures saying 22% of large firms in this country will not complete even mission critical systems by 31 Dec 99.

- This is the "big boys" not small business, who are supposedly much further behind.

- This is the US, not the rest of the world, who is supposedly much further behind.

- This is mission critical, not all other systems, which are supposedly much further behind.

What part of this FACT is not clear to you? Or as your buddy the real estate tycon CPR would say,

WHAT PART OF THIS FACT IS NOT CLEAR TO YOU?

Regarding the lack of y2k work, IMHO the work is not getting done. I know for a fact that testing is not even being considered in many cases, as the actual conversion work has just started and is hitting major snags. Unless of course they switch to SAP, then it's just a piece of cake right Hoff? :)

-- a (a@a.a), May 19, 1999.