Flint,

Can we do "the code is broken" next?

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 27, 1999.

Decker,

Yes, please. Lets. :)

-- Stephen M. Poole, CET (smpoole7@bellsouth.net), May 27, 1999.

-- The pollys (Decker@nd.Poole), May 28, 1999

Answers

The code is buried and very broken. Don't stir confusion. Many people are still on the fence and in denial. Were in a potential wwwIII situation. It doesnt matter. People should prepare for the consequenses of a Govt asleep at the big wheel. Nuclear weapons dont determine who is right only what is left. We need to get in there face and peacefully protest the horror show in kosovo and presist until they stop it. Y2k really sucks and it wont be going away real soon get use to it. Grand canyon full of bad marbles is an understatement

-- y2k aware mike (y2k aware mike@ conservation.com), May 28, 1999.

prove that "the code" is broken

-- Super Polly (get@grip.sheesh), May 28, 1999.

The regulars here know my story. The date code at my job, for our #1 system, is so badly broken, that we are throwing out 300+ Cobol and Macro Assembly programs on our mainframe. We looked at the then 17 year old system in 1996, and because of it's extensive use of dates, and because the system had extensive annual maintenance, worked on by who knows how many programmers, that all think differently, well, we figured it would be impossible to "patch" the system in time. We now have 14 people developing a system from scratch, using all state of the art stuff, SQL, Visual Studio, etc. Here's the best part. We will not be done by SUMMER 2000. We will have enough working to get basic annual production out the door, but the products will be missing quite a few details that are on the current mainframe system.

So, from my tunnel vision point of view, yes, the code is broken. But what do I know. I've only been working and playing with computers for 31 years. <:)=

-- Sysman (y2kboard@yahoo.com), May 28, 1999.

Why all this stuff about "code"? Computers are programmed. Ham operators use "code", moorse code.

To me "the code is broken" means a secret code has been cracked so the information can now be understood. Now if the software has errors, that makes more sense. If it is broken, it does not work at all. sheesh

-- Cherri (sams@brigadoon.com), May 28, 1999.

Yes, I rather think it is.

-- Andy (2000EOD@prodigy.net), May 28, 1999.

Cherri,

"If it is broken, it does not work at all. sheesh"

correctimundo - A+!

well done,

bravo!

amazing perspicacity - full marks, top ho,

excellent, wonderful,

by jove I think she's got it

(Pssst. What happens at rollover Cherri-baby? Uh, the code is broken...)

-- Andy (2000EOD@prodigy.net), May 28, 1999.

Been there... Done that... (25+ years)

The "code" is broken. Yes it is. Oh it may WORK, but I wouldn't trust MY life to what comes OUT...

Oh, and Cherri, please don't try to tell us what jargon to use. With all respect, it's obvious that you're one of the "masses". There's WAY too much techie stuff in this business for you to learn and NO TIME left to learn it in...

Just go turn on the t.v. and enjoy your evening...

-- Dennis (djolson@pressenter.com), May 28, 1999.

Sysman,

So, to use an analogy, you have been a mechanic for 30+ years and your personal car is broken. From this, you want to make a general statement about the functionality of all cars everywhere? (And I respectfully suggest there is much more diversity in software coding than in automobiles.)

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

So, to use an analogy, you have been a mechanic for 30+ years and your personal car is broken. From this, you want to make a general statement about the functionality of all cars everywhere?

Bad analogy. (Nice try, though.) Way, way too general.

Better analogy, more specific: "You have been a mechanic for 30+ years and the door handle on your personal car breaks and -- hey! -- you can't get into the car as usual. From that, you quite reasonably conclude that if the door handle breaks on a 'cars everywhere', then people everywhere won't be able to get into cars as usual."

So, do you get paid to think badly on this forum, or do you just do it for fun?

-- Lane Core Jr. (elcore@sgi.net), May 28, 1999.

I suppose Cherri also thinks we shouldn't say "the code is broken" until we find out "what the definition of 'is' is"!!!!!!!!!!!!!!!!!!!!!!

-- Scarlett (creolady@aol.com), May 28, 1999.

Sysman, then Y2K is a blessing for your system. Great! You finally got rid of that outdated stuff. But you didn't have to start from scratch. Unlike most software development efforts, you already knew the requirements, the functionality, you were updating your system quite different from starting from scratch.

BTW, on the average only 10% of the code is broken. As a rule of thumb, 10% of software is infected with dates and not all dates go through calculation (that is they are only displayed - not a biggie).

-- Maria (anon@ymous.com), May 28, 1999.

In particular, for "my" 17 million lines of code, only 2200 were date impacted.

-- Maria (anon@ymous.com), May 28, 1999.

Maria,

But you had to look at all 17 million lines to figure this out, didn't you?

That's the whole point.

R.

-- Roland (nottelling@nowhere.com), May 28, 1999.

Touche Roland :)

ayeayeaye

-- Andy (2000EOD@prodigy.net), May 28, 1999.

Doggone it, Roland, don't bring up stuff like that, it just gets Maria annoyed and then she just spouts off more silly stuff.

Now, come on folks, lets get with the Polly Program: It does not MATTER if the code is broken, because its only a matter of mankind pulling together and BELIEVING that it will not matter. Banks WILL be fine. The power grid WILL be fine. Well, OK, except for foreign nations, but who the heck cares about them, and they surely don't have anything to do with us. YOU GOT TO B-E-L-I-E-V-E.

-- King of Spain (madrid@aol.com), May 28, 1999.

Sorry, Roland the point is that's what was broke, not all the code was broke. And "looking" at 17 million LOC comes easy with automated tools. For networks and embeddeds the % impacted drops to even less than 10%.

-- Maria (anon@ymous.com), May 28, 1999.

I've waded through my share of spaghetti code, but never thought of bad coding being "broken code"...I just shake my head in disbelief sometimes and say, "This code SUCKS!"

-- Tim (pixmo@pixelquest.com), May 28, 1999.

Dear Maria,

If you hadn't fixed the broken code, I seriously doubt the unbroken code would have done you much good.

A program is a SYSTEM. If part of it is broken, the whole thing is at risk, hence "broken".

Quit quibbling.

R.

-- Roland (nottelling@nowhere.com), May 28, 1999.

Roland- See what I mean? Hoo-boy, she's even gone into one of her nonsensical statistical tirades now....

-- King of Spain (madrid@aol.com), May 28, 1999.

Lane,

Perfect. I can see you standing beside your car with door handle in hand... unwilling to consider the other doors available to you. Wait, you leave the car in the freeway and walk home... dedicating your life to saving all those who are going to DIE because (this is too rich) "IT'S THE DOOR HANDLES, STUPID."

You write a book, launch a web page. It's systemic. The breaking of door handles will lead to a collapse of society. Trucks drivers will stand outside their rigs, and 'just-in-time' manufacturing will sieze. Police will not be able to enforce the law, home will burn to the ground, emergency services will stop... DOOR HANDLES! I tried to tell them to leave the doors off, but they wouldn't listen.

Nothing can be done because there are too many door handles and not enough time. We started to late. The door handles are broken. Forget rolling down the window, forget manual workaround, repairs, forget the new cars... if the old car doors don't work YOU BETTER RUN FAR FROM THE LOCAL 7-11!

We have to learn to live without door handles. After the collapse, we may have the technology to build new handles, but we won't repeat that mistake again. Only the prepared will survive. Be a GI. Buy a bike. Store several year's of food... and guns. When your neighbors can't get into their car, baby, they'll come looking for yours.

Thanks, Lane, you made my weekend.

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

Thanks Mr. Decker you made mine.

Your highness, you have to know a little about Y2K (and some facts) before you conclude the end.

-- Maria (anon@ymous.com), May 28, 1999.

Go plant some pansies on your "retreat" (window box?), see you back trolling on Monday...

-- Andy (2000EOD@prodigy.net), May 28, 1999.

Peonies, Andy. ("Pansies" may have just been a Freudian slip.) Enjoy your weekend.

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

Mr Decker,

Since you like analogies so much, here's mine.

Once there was a young professional figure skater. He was by all appearances the picute of health ad vigor. His eyes, bones, muscles, internal organs, and mental functioning were perfect. He was the epitomy of youthful good health.

One day, when practicing vigorously, he collapsed on the ice, dead of a massive heart attack. All because of a blockage to a major artery to his heart.

Everything about his physical makeup had been perfect, except for that one small blockage.

Guess what? He was still dead.

R.

-- Roland (nottelling@nowhere.com), May 28, 1999.

Maria divulged:

Sorry, Roland the point is that's what was broke, not all the code was broke. And "looking" at 17 million LOC comes easy with automated tools.

Just a question Maria, have you totally relied on automated tools or have you employed additional methods for remediation for the 17 million LOC? If you have employed additional methods what were they?

This is not a bait question, I am sincerely interested.

Regards, Simon Richards

-- Simon Richards (simon@wair.com.au), May 28, 1999.

Simon, we completed that some time ago but I seem to recall that about 90% used automated tools; some we did by hand because the obscure language didn't have any tools and some source code we couldn't find, so that was a re-write.

-- Maria (anon@ymous.com), May 28, 1999.

>And "looking" at 17 million LOC comes easy with automated tools.

Ahh the eternal optimism of the young... Kinda brings a tear to your eye, doesn't it? I remember (more than a generation ago) when I shared that youthful exuberance...

Oh, sorry, back to reality. Maria, how many lines (of your "17 million LOC") have "non-standard" date-naming conventions that couldn't be caught by the "automated tool" or by an initial human review? Do you know?

You still have not said what language(s)/platform(s) this "code" was in/on. (Windoze doesn't count).

Personally, I think you've got a number of errors that won't show up until the code is placed into production, and rollover arrives. Of course, these are my opinions only. (I just base them on 25+ years in front of a hot teletype... er, uh crt.)

-- Dennis (djolson@pressenter.com), May 28, 1999.

Dennis, don't hold your breath for my languages and platforms. If you read my post to Simon, you would see that some did get done by hand. Of course, some got missed. Oh how sarcastic can you get Dennis. Yeah some people do find themselves very amusing while other just shake their hands. Oh back to the point.

Let me continue, yes, the assessment phase of our Y2K project was the easiest. This confirms my idea that Y2K is not your typical development project, in these cases, the initial phases take longer and are not very simple. The hardest part of Y2K was the testing, not the actual tests (they went smooth as silk) but the test preparation. It's not easy duplicating (as closely as possible) the enterprise, connecting to various midranges across various networks and ensuring that all pieces get into the environment. Test set up was a bitch. But... the test results went great.

I, too, expect some problems on 1/1/00 but they will be manageable.

-- Maria (anon@ymous.com), May 28, 1999.

I guess I *WAS* being a *LITTLE* sarcastic

I am happy that things went well for you. One small factoid if I may: testing NEVER covers all "real-world" conditions. I don't care HOW much time was spent on setup. It just won't happen.

By definition, you WILL experience failures.

An dat's da name 'a dat tune...

-- Dennis (djolson@pressenter.com), May 28, 1999.

Decker, you know as much about gardening as you do about computer code. The proper time to plant peonies is in the fall.

-- Flower Child (h@ashbury.hip), May 28, 1999.

Yup, in the fall...

-- Andy (2000EOD@prodigy.net), May 28, 1999.

"The code" does not exist. This is probably the linchpin of the doomer argument. There is no "the code" that can be broken. Saying "the code" makes it sound like the entire world is run by one computer program. Talk about an analogy being too general!

-- Buddy (buddydc@go.com), May 28, 1999.

Buddy, Ever read "The Machine Stops" ??? Joe-bob says check it out

by E. M. Foster

-- Andy (2000EOD@prodigy.net), May 28, 1999.

Depends on whether or not you consider "the code" to be singular or plural.

Most people say "the data *IS* relevent", but statisticians would say "the data *ARE* relevent". It's all in your definition of the term.

"now go away son, ya bodder me..."

-- Dennis (djolson@pressenter.com), May 28, 1999.

Fall is the best time to plant peonies (in most areas), but not the ONLY time. They are hardy perennials and should do well. I will freely admit that I am not an expert in horticulture or computer programming. But expert status is hardly a requirement for participation in this forum....

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

as you have repeatedly proven dd

-- Andy (2000EOD@prodigy.net), May 28, 1999.

It does not depend on whether "the code" is singular or plural. "The code" is tossed around by many in the doomer camp, especially Hamasaki. It has developed into a legendary saying. Many of the doom & gloom arguments depend on such generalizations.

Sorry I bother you Dennis, but I doubt you're old enough to try calling me "son" to my face.

-- Buddy (buddydc@go.com), May 28, 1999.

Well, whaddya rather hear, buddy? That the compiled source which comprises the application is broken?

-- Lisa (nit@pickin.everywhere), May 28, 1999.

Different words, same meaning, Lisa. Same shit, different day, as they say.

-- Buddy (.@...), May 28, 1999.

Peter de Jager used the phrase "the code is broken" in a speech that can be seen at this link:

http://www.bog.frb.fed.us/y2k/CSPAN/19980602/deJager.ram

The phrase "the code is broken" usually refers to the idea that Y2K would have caused significant disruptions if there had been no attempt to fix it. It also refers to the idea that significant disruptions are possible if enough systems that represent years as two digits are not fixed in time.

Peter de Jager is not the only Y2K commentator who has discussed this point. So did Senator Bennett in the speech at this link:

http://home.swbell.net/adheath/bennett.htm

[snip]

I believe we're going to win; that is I think that civilization as we know it is not going to come to an end. It's a possibility. Possibility, if Y2K were this weekend instead of 76 weekends from now, it would.

[snip]

-- Kevin (mixesmusic@worldnet.att.net), May 28, 1999.

You can see it on almost every thread now: The retreat of the pollyannas into a bizarre defense of shrugging the entire Y2K problem off as either trivial, or one that will not actually affect "us". Code, what code? Oh, that code -- automated tools can do it, its easy. (Forget that such tools are only for a handful of languages on a few platforms, and often do not check for all Y2K problems, such as leap year determination, use of dates to signal special procedural actions. Oh, and also forget the code in embedded chips, if you can't see it it really isn't there.) And its not like its one big program, but lots if itty-bitty programs, so that makes it REAL easy to find and check. Yada-Yada-Yada.

Meanwhile: the code is still broken.

-- King of Spain (madrid@aol.com), May 28, 1999.

Spain,

Let me know when you're on the cover of PC Week or any IT mag. Look forward to your expert opinion shattering the complacent IT community.

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

Okay Mr. Decker, pretty amusing and encouraging analogy, I've saved it to cheer me up when I'm depressed...but the argument is not that there are no workarounds. We all know that systems fail all the time. The problem is, it looks like the rate of failure is about to spike sharply upward. We've never experienced that before. We don't know at what point we run out of workarounds. This point was made by the Defense Department back in 1996--that they are used to falling back on contingencies, but they have no contingency for a large number of failures all at once. Just because our society is robust enough to handle 1% of it's systems being down at any given time, does not mean we can handle, say, 10%.

There's a concept in epidemiology called the "tipping point." The idea is, if 50,000 people a year die of AIDS, and 49,000 people a year contract AIDS, then AIDS will go away. But if 51,000 a year contract it, you've got an epidemic. The difference is only 2,000 people, but those particular 2,000 make all the difference. I tend to think there's a tipping point with Y2K failures, at which point failures cascade out of control and the system crashes. We don't necessarily have to fix everything; if we fix enough so we fall short of the tipping point, Y2K will be a major annoyance and no more. If we cross over, we've got major problems.

-- Shimrod (shimrod@lycosmail.com), May 28, 1999.

Shimrod,

Excellent point. I use the phrase "critical mass." I'm not sure there is a critical mass of systems failures because outside of the iron triangle, I think there is a high degree of independence. It's a great subject for discussion.

Critical mass also interests me in the human factor. How many people have to panic before the human element is more significant than the computer element. My original post on this forum worried about panic. To me, this seems a real danger.

Regards,

-- Mr. Decker (kcdecker@worldnet.att.net), May 28, 1999.

Mr Decker:

I think what you meant to say was that outside the iron triangle, the critical mass is beyond what you consider reasonably reachable. Surely *in theory* enough simultaneous breakdowns by enough economic players, serious enough to render them inoperative for some period of time (weeks, perhaps) would be a critical mass. A 2-3 week period during which *nothing* was produced by anybody would have real economic impact. I agree this is unlikely.

From here, the issue looks fractal. At the macro level, we're looking at nonfunctional countries ("Italy is noncompliant", whatever that might mean). A level or two down, we're looking at major infrastructure players. Below that, we look at activities that don't require such a huge, expensive, slow-to-replace physical plant. Below that, we're looking at critical and non-critical systems within organizations. Below that, we're looking at critical and non-critical bug impacts, etc.

Somehow I doubt that the problem is similar at all scales.

-- Flint (flintc@mindspring.com), May 28, 1999.

Mr. Decker,

It's not my car that is a problem. It's like GM having to recall every car they ever built. Oh, Ford and Honda and all the rest may be fine, but the amount of work needed on each GM car varies widely. Some have been in the shop for years already. Some ignored the early recall, being too busy to bring it into the shop. Some, who knows...

Maria,

Yes, it will be nice having a "new technology" system, but except for the date problems, there was nothing wrong with the old one. We could have kept it running for another 20 years. It did the job just fine. The problem now is we are having to turn away new business, because all of our resources are tied-up in the conversion. We're an ESOP company, and it would have been much nicer if we could have kept the old production dollars, and use our development talent to bring in new bucks.

More later, I haven't been on and have to catch-up. <:)=

-- Sysman (y2kboard@yahoo.com), May 28, 1999.

This thread sorta continues here... <:)=

The "Code" is NOT broken!

-- Sysman (y2kboard@yahoo.com), May 29, 1999.