'Y2K Seems Like a Joke Now, But in 1999 People Were Freaking Out' (npr.org)
- Reference: 0175785059
- News link: https://it.slashdot.org/story/24/12/29/195226/y2k-seems-like-a-joke-now-but-in-1999-people-were-freaking-out
- Source link: https://www.npr.org/2024/12/28/nx-s1-5116271/y2k-year-2000-preparations
> The Clinton administration said that [2]preparing the U.S. for Y2K was probably "the single largest technology management challenge in history." The bug threatened a cascade of potential disruptions — blackouts, medical equipment failures, banks shutting down, travel screeching to a halt — if the systems and software that helped keep society functioning no longer knew what year it was... Computer specialist and grassroots organizer Paloma O'Riley compared the scale and urgency of Y2K prep to telling somebody to change out a rivet on the Golden Gate Bridge. Changing out just one rivet is simple, but "if you suddenly tell this person he now has to change out all the rivets on the bridge and he has only 24 hours to do it in — that's a problem," O'Riley told reporter Jason Beaubien in 1998....
>
> The date switchover rattled a swath of vital tech, including Wall Street trading systems, power plants and tools used in air traffic control. The Federal Aviation Administration put its systems through stress tests and mock scenarios as 2000 drew closer. "Twenty-three million lines of code in the air traffic control system did seem a little more daunting, I will say, than I had probably anticipated," FAA Administrator Jane Garvey told NPR in 1998. Ultimately there were no systemwide aviation breakdowns, but airlines were put on a Y2K alert....
>
> Some financial analysts remained skeptical Y2K would come and go with minimal disruption. But by November 1999 the Federal Reserve said it was confident the U.S. economy would weather the big switch. "Federal banking agencies have been visited and inspected. Every bank in the United States, which includes probably 9,000 to 10,000 institutions, over 99% received a satisfactory rating," Fed Board Governor Edward Kelley said at the time.
The article also remembers a California programmer who bought a mobile home, a propane generator, and a year's supply of dehydrated food. (They were also considering buying a handgun — and converting his bank savings into gold, silver, and cash.) And "Dozens of communities across the U.S. formed Y2K preparedness groups to stave off unnecessary panic..."
But the article concludes that "the aggressive planning and recalibration paid off. Humanity passed into the year 2000 without pandemonium..."
And "People like Jack Pentes of Charlotte, N.C., were left to figure out what to do with their emergency stockpiles."
[1] https://www.npr.org/2024/12/28/nx-s1-5116271/y2k-year-2000-preparations
[2] https://clintonwhitehouse4.archives.gov/WH/Work/121499_1.html
Data point (Score:5, Interesting)
Many freeze dried canned foods of this age have a nominal 25 year shelf life. I lurk a number of "prepper" groups on Facebook, and 2024 has triggered a number of "using up the Y2K cans" discussions. Fun times.
worked that night (Score:4, Interesting)
I got paid $1000 to sit in the call center that night, our main customers were phone companies, with a few large businesses and universities. I got a single customer call, about nothing unusual. Went home at 2am.
Re: (Score:2)
I was onsite, on the phone to a vendor as the clock struck midnight. Sure enough, the error occurred. Paused the app, give error message, got the email, copied the patch to a floppy, ran it, test ok, resumed operations. Downtime, 4 minutes or so. Customer satisfaction, priceless.
It was my only client that needed a stand by. The rest had all settled weeks in advance. One client fired us because we refused to recommend migrating from NetWare to Windows NT. Their new NATS hardware crashed 1/2/2000 when the bac
Re: worked that night (Score:2)
I chose on purpose to work the late shift as a call taker during late night talk radio at KGO radio in San Francisco, because being around newsrooms and talk show hosts with a lot of adrenaline during important events is fun. I assumed it would be crazy that evening.
It was boring as hell. Nothing happened and I also went home early.
Re: (Score:2)
I was working at an electrical utility a few years later and heard about how close the electrical grid had come to imploding. At that time all sales and exchanges of energy were done manually, much of it by emailing around a giant Excel spreadsheet full of macros doing marginally-to-not documented things. That thing would have crashed and burned, as well as their backup system, and left North America without an effective way to trade electricity. The new system was put into production Thanksgiving weeken
Y2K was a joke... (Score:4, Insightful)
... in the same way the pandemic-preventing measures required during the Covid disaster, or Measles and other vaccine programmes were jokes.
i.e NOT
Re: Y2K was a joke... (Score:1)
The US keeps trying to force the "truth".
We already know the truth, and it doesn't matter.
And...
We failed.
Re: (Score:1)
No one gives a shit. The origin makes no difference in how the pandemic was handled. It just helps satisfy you that you can blame your own country's incompetence on someone specific.
Re: (Score:3)
Very true. A lot of hard work was done by many thankless programmers during the years leading up to 2000. Thanks to their hard work, y2k was just another new years celebration, instead of a disaster. Funny how people look back now and think there really wasn't any problem.
Re: (Score:1)
Indeed.
We had some servers disconnected from the intranet with their clocks settable to Y2K for testing executables, and did much searching thorugh code.
We had one actual Y2K bug after Y2K, luckily it was in some non-critical software and easily fixed when triggered.
We, like many other organisations, did the work to prevent any big problems.
And we succeeded.
Our Y2K handling was a success.
It was not a joke. Or a disaster.
Re: (Score:2)
I worked at an insurance company in the late '90s, they were running their core process on the obsolete and no longer supported Empress database on an ancient VAX. Their Y2K testing came through fine, and Andrew was bored a few days later and decided to test the change from February to March since there was no February 29 that year. To his shock the DB crashed hard when he changed the date to February 28! Working backwards he eventually found that for no known reason Empress puked on February 11, 2000.
Re: (Score:2)
I can tell by your user id you were around in 1998 too...
I had stopped coding in 1992-ish and in 1999 I got a call from a buddy saying " hey you got any VAX-cobol skills?" I said "yes, why?" He said " I got 1-3 million lines of code that need your help. pay you a lot for it."
that was the largest check outside of commissions earned at that time. it was larger than my wall street bonus at the time.
I jury rigged a windows 95 computer, a terminal program, MS access and I think wordstar ( not sure ) to do massiv
People were freaking out, but techs weren't (Score:5, Interesting)
We all knew nothing substantial would happen in 2000. Why? Because a lot of smart people had already worked hard to alleviate the problem years prior.
But the media frenzy was relentless: stories of Y2K apocalypse went around all newsrooms, and it was a competition between them to capture as much audience as possible.
Me, I was working for a small company doing embedded MS-DOS. We made a fortune selling a "Y2K cure" floppy disk: all it did was set the clock to 2000, check whether DOS returned 2000 or 1900, and in the latter case, install a completely trivial TSR to add 100 years. We sold that disk $60 a pop and it sold like hotcakes.
And you know who bought it most? Companies who knew perfectly well their computers were fine (because their IT guys had tested them using the exact same trivial method) but needed to be "Y2K-compliant" for insurance purposes.
Y2K was a great scam. A lot of people made off like bandits.
Re: People were freaking out, but techs weren't (Score:3, Insightful)
100% spot on. I was in so Cal at the time, a member of a club with many programmers. Their view of the tech industry was realistic. FUD - fear, uncertainty, despair - was used to force upgrades, hire more IT, get updated hardware. This snowball just kept rolling downhill, increasing in size. Programming once was parsimonious, very fast and efficient, but then, we got layers and layers of middlemen surveillance, ad tech, and now, so called AI ... and of course if we continue on the current AI trajectory
Re: (Score:3)
I was at my desk at midnight 31/12/1999. I watched the fireworks out my office window. One of my colleagues and I ran through our test plan that we had repeated several times before in controlled environments. We monitored and measured, and by about 1:15am, we left the office confident that we had completely resolved all the issues in the mission critical systems.
In the following days, only one incident was reported across 1245 different software (and hardware) systems. It was in a third party product
Re: (Score:2)
The potential problem wasn't whether DOS, or other operating systems (Solaris, etc) would handle the date, but rather all the legacy applications which potentially might have assumed the century as 1900 and only handled the year as last two digits (year % 100). This was compounded by the fact that some of the really old systems, most likely to have assumed they'd not still be operating in the year 2000, were written in COBOL or ancient versions of FORTRAN, so it was a payday the retired COBOL programmers ta
Re: (Score:3)
Agreed. I was in SF and surrounded by nascent SiliValleyites. I can't remember any of them considering Y2K anything but a nuisance. We all spent a lot of time going over high-level sources hunting down 2-char year fields, and then it was over. A true nothingburger.
The Y2K panic was one of if not the first of the pure taken-out-of-context, cable-media-created apocalypses. They were astounded by the audience shares Y2K stories were getting, and learned quickly how to flog a BS issue til it dies - all for
Re: (Score:2)
For me it was noting on that particular date, but that's because I had altered the code so that the tw digits were expanded to fit in the range 1903 - 2002 (updated yearly with the current year as the center).
A four digit year would have been nicer, but the storage requirements wouldn't fit on our machine. (I forget how many dates were involved, but it wasn't trivial.) This only worked if the assumption that nobody over 97 or under 3 was legitimately using transit was correct, but I didn't find any proble
Re: (Score:2)
> Because a lot of smart people had already worked hard to alleviate the problem years prior.
No. Not years prior. The craze of Y2K kept going up until December 1999 precisely because people weren't ready. 1999 was a great year for people on contracts and people with overtime pay working in IT. It was a very real issue up to the end with in some cases major institutions still testing code in the last month.
Practical Joke (Score:5, Funny)
A friend of mine went to the bank in December 1999 and bought a stack of $2 bills. She went shopping at various stores just before the New Year and paid with $2s. Clerks asked, "What's up with the $2 bills?" She said, "I went to the bank for some money and these are all they had left."
The look of terror on a few clerk's faces was priceless. Here I am, stuck at work and the bank run has started.
Re: (Score:2)
There were 279 million people in America at the time. You feel really confident that no one pulled precisely this prank? I mean even if you are right and this person just made it up, it means that at least one person has thought of it already which means that the odds are someone else thought of it too.
I would wager it happened on more than one occasion.
Calling out liars is best done for bullshit claims rather than things that people would absolutely do for fun.
Everyone was sure (Score:2)
The power was going to go out on NYE
Re: (Score:2)
It would have if the operators of the North American power grid hadn't completely revamped their electrical purchasing system, IIRC the new system went into production Thanksgiving weekend of 1999.
Still a lot of legacy tech out there (Score:4, Interesting)
I keep hearing stories about how companies are caught with legacy tech without replacements, as hardware breaks and employees leave and then suddenly getting in trouble. Windows 10 going out of support will be a huge issue because it will mean the huge replacement of non Windows 11 compatible computers and software having to be migrated to TPM and MS accounts. The end of 32-bit support is slowly winding through the industry as well as well as a lot of legacy software had to be retired because it wasn't TLS 1.2 complaint. We will see a lot of software breakages that are more than just time based, as so much modern software is tied to remote servers that can shutdown when the corporate credit card expires.
There was little freaking out (Score:5, Interesting)
A few people were nuts, just like today. We spent a lot of time auditing code to avoid disruptions to financial systems. Anybody with half a brain was laughing at the idea your car would stop running or anything of that nature.
A lot of people stayed home on New Year's not because of Y2K fears, but because venues got it in their heads they could up their rates by 1000x for a millennium party and people noped out of the rip off.
Re: (Score:2)
This, exactly. SOME people were freaking out. Most of us were aware enough of all the previous several years' hard work that'd gone into addressing the known problems that we expected, at most, some minor inconveniences.
And that's exactly what we got - relatively minor inconveniences.
Re: There was little freaking out (Score:1)
zeitgeist We were all on the same page
Re: (Score:2)
It was a bit nervous, but a LOT of work was done. Some idiots think it was a scam, or a big panic over nothing, but if none of that work had been done it would have been highly disruptive. We were seeing weird issues show up a couple years early at times, credit cards with weird expiration dates and stuff like that.
What's surprising to me is that so many people still can't figure out how to program with time. Most think that problem was just dumb database guys trying to save two bytes, but you see other
Re: (Score:2)
> because they're still naive enough to be using local time for all base time calculations
Yeah, I don't care what you show to the end user, but EVERYTHING should be stored (and processed) in UTC. No daylight saving adjustment, no time zone adjustment. All UTC. Behold, all conversion errors go away and mistakes are never made again - because the only possible mistake is on the final display and you're going to catch that on day one.
Then again, I think the only logical date/time format is YYYY-MM-DD HH:
Re: (Score:2)
FWIW, MSWind95b didn't roll handle the rollover correctly. Files sorted by date started appearing in the wrong order.
Re: (Score:2)
Univision and the rest of the Latino press had scared the crap out of their viewers, our niece was so frightened that she cried when she heard that we were planning on going to the Space Needle for the celebration. We ended up staying home drinking and screwing like bunnies, so still a good time.
Some Lessons Learned (Score:2)
I have an APC rackmount UPS that gets ignored if it doesn't alert.
Well the battery got old and alerted so I logged in and noticed that the date was off.
Turns out the year can't be set past 2019. This was bought in 2015.
And I guess Eaton owns it now and maybe if I install windows on a laptop I can piecemeal upgrade the firmware to something without the problem?
Or just turn off NTP and ignore it again.
Re: (Score:2)
Under the heading of 'unnecessary check', maybe it's just that they did not expect the unit to last 5 years.
Me? I know how to swap out batteries in most anything.
Re: (Score:2)
Ah, I have to deal with 20 year support contracts. And sometimes 20 year battery life requirements. That's a long time in the technical world, but they're devices in the field that are expensive to go out and visit in a truck. But like all software or firmware, there's a lot of creeping features. So eventually the new stuff will not fit in the old devices in the field. Then someone starts to panic. Why isn't the customer buying new hardware (because we support it for 20 years); why can't we make the co
Re: (Score:2)
2010 Spamassassin marked e-mails as spam because of a rule that defined 201x as "Date far in the future".
No, the media was (Score:2)
The rest of us knew not to worry or panic. But you can always rely on the media to blow things out of proportion, so that's precisely what they did.
The real Y2K scam... (Score:3, Insightful)
As someone who was working in I.T. well before Y2K (note the 4-digit Slashdot ID) and who worked on actual code and system remediation projects for Y2K, I can assure you that it was NOT a technical scam. We had old billing systems (think: AOS/VS II hosted COBOL, other dusty old crap like that), along with untold Oracle back-ended applications, which were both critical for a multi-million dollar company's billing cycle yet also poorly written enough that they absolutely 100% would have produced incorrect output starting in 2000. All of that had to be gone through with a fine tooth comb and fixed, which we did, and as a result my own Y2K on-call was as boring as hell....precisely because we'd done all that work. Other companies had similar Ancient Horrors lurking on their data center floors, and all those had to be fixed or replaced too. Never doubt the need, or underestimate the work required to get it all done.
Now, all that being said, there WAS an actual Y2K scam, but it wasn't a technical one: Rather, "Y2K Remediation" budgets became the dumping ground for all the bad ideas, lost projects, failed acquisitions, questionable purchases, mistress payments, and other C-level executive idiotic decision making. No shareholder would question the need to fix the actual Y2K problems, especially given the overwrought news cycle around that messaging. As such, C-types took the opportunity to bury all their mistakes in that budget, never to be seen again. I saw it happen with my own eyes, but nobody outside really grasped what all they were doing.
THAT was the real scam of Y2K, not the actual technical issues that needed to be fixed.
Morons are always "freaking out" (Score:2)
Non-morons look at things and use rational though (theoretically available to any human, but most seem to not use it much or at all) to understand the threat. Then they prepare. And that typically works out just fine.
Godfellas (Score:3)
In the Futurama episode [1]Godfellas [imdb.com], Bender encounters what he believes to be God. The most memorable quote from the episode is, When you do things right, people won't be sure you've done anything at all.
That applies to Y2K. Like the story says, people were working on this long before the cutoff date so when things didn't go wrong the public was left to wonder what the big deal was all about, that it was overhyped. That's because those working on it did things right so the transiton was seamless. A few archived stories from back then:
[2]FAA pre-Y2K testing [cbsnews.com] [3]Flights were unaffected [wired.com].
A tidbit about U.S. and Russian military [4]sitting side-by-side [nytimes.com] during Y2K.
[5]Stories from around the nation [scientificamerican.com] from people at that time.
[1] https://www.imdb.com/title/tt0756880/
[2] https://www.cbsnews.com/news/faa-we-passed-a-y2k-test/
[3] https://www.wired.com/2000/01/no-planes-didnt-fall-from-sky/
[4] https://archive.nytimes.com/bits.blogs.nytimes.com/2009/12/30/the-y2k-that-thankfully-never-happened/
[5] https://www.scientificamerican.com/article/the-glitch-that-didnt-ste/
1/4 century later - still the same issue (Score:1)
DEBUNKING BS:
There were no 23 million lines of code (lmfao) to comb through but there were enough to make it a part time job for many consultants -- who would be released after the new year. You can tell this is a made up number because lines of code is not a measurement used by programmers, and differs greatly based on chosen language. So, for example, C-code is more efficient but uses a lot of short lines. ADA or COBOL are more complex, so less LOC for the same function.
THE REAL PROBLEM:
The underlying
Re: (Score:2)
With "we have DD/MM/YY (Eu), MM/DD/YY (US) YYYYMMDD (programmer standard for sorting purposes) etc." you forgot YY/MM/DD for China and Japan which, as a programmer, is my preference. I'm in New Zealand so I live with DD/MM/YY. So Eu and NZ are little-endian and Cn and Jp are big-endian, WTF do you call the US order and who thought that was a good idea?
Re: (Score:2)
While for human eyes and not programming, for decades I have used DD / MMM / YYYY where MMM is alpha, so 31DEC2024. No matter where you are from it's unambiguous. It might be that the alpha for the month varies by language, but it can be decoded without asking me what comes first
Re: (Score:2)
8, 16, and 32 bit computers will still be around in 2038. You think they're sticking a 64-bit processor in your pace maker? These small chips are everywhere. And to replace them all is too expensive. Thus, different time methods - why should that pace maker use Unix time that was invented for file times? Just pick a different epoch that's not 1/1/1970.
Re: (Score:2)
Lenel, the largest security software company on the planet, uses hardware from Mercury with a 20 year expected life that runs on an 8-bit CPU (and there's a LOT of it out there that is even older).
Exactly two failures reported (Score:2)
I was working for a certain large IT company at the time, and there were two failures that I know of on the night.
First was the local snack machine decided to shut down at midnight, due to its countdown for scheduled maintenance suddenly tripping. Shame, since we were relying on that machine to keep us at our desks and able to pick up the phone should the need arise (it didn't).
The second failure was with the badges at one of our facilities in the UK. We ran out of temporary swipe badges, since a larg
Re: (Score:2)
I've seen weird stuff. Unix time for medical machines always seemed weird. It was invented for file times. Not a universal time method for all objects everywhere. So it had difficulty dealing with birthdates and ages at times. Yes, you needed to deal with someone being born in 1899 and having surgery in 2001, and both dates have to exist on the same screen at the same time.
Then there's the GPS headache, with a time format that wraps around in roughly 19.6 years. Sucks if you want a device that lasts 20
Success (Score:2)
Y2K was a success. People act like they were fooled, rather than saved. Sure there was a lot of hype but that hype was needed to get some people to take action, and it was that action that save a lot of drama on the day. The fact that nothing much happened if proof that the critical systems that need updates got them.
We Worked Hard and Everything Was All Right (TM) (Score:2)
I was working with a large manufacturing company at the turn of the millennium. They had invested in IT early and broadly. By the late 90's, we had a couple hundred VAX/PDP/Alpha clusters and a couple dozen Windows NT-based systems, with 30 years worth of custom code from assembler to FORTRAN and Top-down to OOP. We had FMS Forms, Visual Basic using ODBC, Pathworks, and custom network stacks. COBOL and QNX, RSX with overlays, and DEC everything. We had DCS and other PLC-based systems and interfaces. Automat
Firsthand Experience (Score:2)
I've been a Systems Administrator since the 1990's. I was relatively well informed leading up to Y2K due largely to my brother who was handling it for a government agency. We had a Unix system running the main financials and such at my job. It was Data General gear, if I recall correctly. In the months leading up to Y2K I performed a test where the hardware clock was rolled forward to January 1, 2000. The system was not even able to boot! If we had not tested in advance and hurriedly transitioned to a diffe
I'd love to hear is tales of what did not happen (Score:1)
One thing I'm not sure I ever saw, was an accounting of developers that fixed Y2K issues beforehand of various systems that would have resulted in significant problems if they had not been fixed.
I'm curious what the worst possible case would have been had programmers not bee as diligent at fixing systems... I figure there have to be at least one or two really hair-raising things that got fixed that could have been really bad.
Manufactured non-issue (Score:2)
Many systems had genuine issues but they were recognized, fixed, tested, ready to go when the time came. The panic was totally unnecessary.
I observed 0000 1 January 2000 three times: New Zealand UTC+13, first place west of the International Date Line with lots of computers. 0000 UTC (of course...), then 0000 local (Toronto, UTC-5). I was doing my Master's research in satellite data communications at the time and noted in mid-December that one of the satellites I was working with would be in the sky at 000
Who knows? (Score:2)
There are two parts to Y2K, the real problems that needed to be fixed and the catastrophic results that would happen if they weren't. Plenty of people who worked on real problems during the Y2K hysteria are convinced their hard work was part of preventing a catastrophe. There are others who think any major problems were easily identified and the consequences of the remaining problems manageable. We will never know which is correct.
2038 (Score:2)
We're closer to 2038 now than to 2000. And that bug will not be a "To few digits to display" bug.
First issues already occurred, for example, in systems that calculate when people will retire.
2038 is comeing! (Score:3)
2038 is comeing!
Re: (Score:2)
So switch to a computer with 64 bit integers.
Re: (Score:2)
is that really the solution, back in 1999 when people joked about it, there was no code upgrade at the time.