IT team forced to camp in the office for days after Y2K bug found in boss's side project

December 26th is a holiday across much of the Reg-reading world, but it's also a Friday – the day on which we present a fresh instalment of On Call, the reader-contributed column that recounts your tales of tech support encounters and exasperation. This holiday season we're dipping into tales of tech support in the time of Y2K …

[Gomez Adams]

Working in IT for a bank our small team took it in turns to be on site for the duration but all of us were still expected to be on call as we all had our own sub-areas of experience and expertise including knowing the best team member to contact for any given type of issue (lot of in-house bespoke software in play). Had a decent remuneration in return ramping up to quad time for on site hours and double time for on-call time if called upon, single time if not. But thankfully, due to our due diligence, the event passed us by without a hitch so the only downside was having to moderate our celebratory Y2K drink.

Now, about those neutrino hits on the computer core causing the occasional inexplicable system failure ...

[MiguelC]

I've already told this more than once, but the tale never gets old - well, it does get one older as years go by, but you get the point :)

That fateful night I was on duty, after a year and a half working on that Y2K project for a large bank.

Midnight passed without a hitch, and around 1 AM I went to the nearest ATM and checked my balance and latest account movements (not an account from the bank I was working for). There was an interest credit of around the equivalent of 3000€. Resisting the urge to withdraw it there and then, I went back, showed the slip to my co-workers and wondered what would happen from then on. At 8 AM, after an uneventful night on the job, I went down and checked my balance again. Without a trace of that earlier payment, it now showed the correct and, unfortunately, much smaller interest deposit...

I bet someone's night was a lot more eventful than mine

[Doctor Syntax]

This story has also been told here previously but worth repeating because there's a moral in it.

Client had 2 Sparc based systems running their accounting. One was new enough to have run the Y2K prepared version of the application, the other, a warm standby wasn't. (In the course of testing I discovered that the warm standby not only wasn't new enough, it also wasn't a warm standby because the overnight copy was failing due to lack of time. As this was previously unknown it wasn't a factor in the decision making.)

They decided to replace them with a pair of SCO boxes running the latest version. My job was to set these up and look after user testing. The bean-counters ran their acceptance tests and pronounced them OK. The planned cut-over was between Christmas & New Year when the plant was shut down.

At the last minute, despite dire warnings, bean-counters completely forbade any replacement before they'd completed their year-end bean-juggling on the old system which would take them up to mid-January. If we'd had a bit more warning we could have insisted on their testing the year-end on the new boxes but, of course, they hadn't done that.

Inevitably there were problems which required the A/C package vendors dialling in, sometimes multiple times a day, to sort out the mess until we were allowed to switch over.

The moral of the story: for those who insisted that all preparation was just unnecessary freeloading and panicking, those who chose to deliberately run on a nonY2K-proofed system suffered real on-going consequences.

[steviebuk]

THIS. THIS IS THE STORY THAT NEEDS TO BE TOLD!

The amount of "quiz shows" etc that bring up the Y2K event and the "comedians" always take the piss that "nothing happened" ignore that "nothing happened" because people people spent months BEFORE the date fucking fixing it.

I would say I missed out on all the money earned then as was just finishing off a HND so not in work, but appears so did lots of people when companies were being tight and just gave people the bollocks of Time off in Lieu.

[jake]

"people spent months BEFORE the date fucking fixing it."

Months? Try a couple decades, on and off.

[David Austin]

I wish all of those people a merry 19th January, 2038.

[Anonymous Coward]

"in December 199 he worked in the IT department"

Wow, a true greybeard. I'm remembering a video of a monk teaching someone how to use a book... And it wasn't called the UK back then, but I'm sure they had huge booze purveyors, even multinational ones.

https://en.wikipedia.org/wiki/Pithos

"And when you need a crane to move your booze, it's time to admit you have a problem!" Arthur, Legacy of Time (quoted as close as I can recall)

[mobailey]

Re: "in December 199 he worked in the IT department"

It was the Y2C bug back then.

[Pickle Rick]

Re: "in December 199 he worked in the IT department"

A Meux & Co's Horse Shoe Brewery like event?

INVALID DATE DETECTED. RESETTING SYSTEMS... SYSTEMS RESET... FERMENTATION THERMAL RUNAWAY DETECTED. UNINTENDED UNSCHEDULED DECONTAINMENT IMMINENT. BRACE BRACE BRACE!

*SPLOOOSH*

Operator: I've always dreamed of going out like this! >glug glug glug< >glurp<

[Sorry/thanks Simon!]

[Doctor Syntax]

Re: "in December 199 he worked in the IT department"

That would be the abacus department worrying about whether they had enough beads for CC.

[C R Mudgeon]

Plus ça change

"in December 199 he worked in the IT department"

In December 199, they didn't know it was the year 199. AD (more modernly, CE) dating wasn't invented until Dionysius Exiguus came up with it in the year he called, and we still call, 525.

I wouldn't nitpick your joke apart if it weren't that this might be surprisingly on the topic of calendar-based hysteria.

Dionysius's main reason for the AD thing is irrelevant here: that he didn't like the epoch in use at the time, which was based on the reign of Diocletian, a notorious persecutor of Christians.

It has been suggested, however, that he had another reason as well. It seems that some believed that (to oversimplify) Christ's Second Coming and the consequent end of the world were due to occur exactly 500 years after his first one -- and that Dionysius hoped to dispel any panic by pointing out that the forecast Doomsday was already 25 years in the past.

[collinsl]

Re: "in December 199 he worked in the IT department"

Medieval Helpdesk

[Pascal Monett]

Ah, the great Y2K scare

I remember spending months on client sites checking any and everything that could possibly go wrong, making some updates here and there to coding commands that had been deprecated and replacing them with the newer "accepted" ones.

And I had it easy. As a Notes developer, in those days Notes belonged to IBM and, say what you want about IBM and how it markets its products - or fails to do so (and I have already vented a lot in these hallowed pages) - one thing you cannot take away is the fact that IBM made Domino/Notes a solid, resilient platform.

So Y2K came and went and I didn't get any panicked calls from customers because of Y2K-related issues.

But the groundwork had to have been done.

[Pickle Rick]

Re: Ah, the great Y2K scare

My largest client at the time was a significnt polymer compound manufacturer. I wrote and maintained their Production System, works orders and the like. When I asked the MD about their Y2K policy, including embedded control gadgets, his response was "We don't give a toss, if it breaks, we'll go manual. The Production System's good, right?" - which it was. Now that's unflustered contingency planning :)

[david 12]

Re: Ah, the great Y2K scare

A company I was contracting with had been told that if anything failed at Y2K, people would be fired.

So they spent 2 years getting ready for Y2K, locked everything down, and had only 6 months to prepare for the introduction of a V.A.T.

They missed the deadline. With more than a million dollars of unpaid invoices after 2 or 3 weeks their supplier refused further supply, and they were using paper ledgers while IT worked on fixing VAT in the accounting system.

[CorwinX]

Re: Ah, the great Y2K scare

I used to specialise in Notes/Domino - mail not so much dev.

It was pretty bulletproof.

Y2K - not an issue - it had always stored full 4-digit years.

It had 2FA before anyone had heard of the term (User ID files).

It mostly disappeared because it wasn't flashy compared to Exchange.

It also introduced web access, which was pretty seamless compared to the full Notes client.

With full public/private key encryption fundamental to how it worked.

There's still some financial companies that use it because they just don't trust Exchange.

[SecretSonOfHG]

Re: Ah, the great Y2K scare

Domino was a rather clever and solid document database replication system with an incredibly bad user interface, to which someone bolted on top a mail system whose UI was even worse.

As it was created on a time when nobody wanted or needed a distributed document database, workflows and email were its only chances of survival. The workflow part survived much longer, but the email part could not compete with even free offerings. And its “webification” was… atrocious, and perhaps sabotaged by fears of losing these juicy client licenses. For a client software that nobody wanted And everyone hated.

Of course, IBM purchasing it was the final nail on the coffin.

[tip pc]

Re: Ah, the great Y2K scare

IBM. sold it to HCL

https://en.wikipedia.org/wiki/HCL_Notes

[Richard 12]

Re: Ah, the great Y2K scare

It also worked just as well over slow and long-latency data links, eg satcom to a ship.

Exchange, not so much - and MS365, not at all.

[Not Yb]

Re: Ah, the great Y2K scare

Notes also attempted to hide the user's password length by using a random length of Xs (I think) for each keypress. Using it was the first time I'd seen software quite that paranoid.

[theblackhand]

Re: Ah, the great Y2K scare

Out of interest, how did you check Notes was working?

Because in my experience of Notes over multiple companies including resellers using Notes/Domino to replicate IBM parts and price lists, bring broken was almost the default state.

[David Hicklin]

Re: Ah, the great Y2K scare

My experience at a small company as the sole IT guy was mainly compliance checking plus a few patches for software to "bodge" the date issue by deciding that anything less than 50 using just 2 digits must be in the 2000's, 50+ was in the 1900's.

[Herby]

Re: Ah, the great Y2K scare

Choosing an arbitrary date (1950 in this case) is putting off the inevitable. There will come a time when the year 2050 rolls around, and someone will unexpectedly will discover the "but" that was designed in. All you can hope for in that case is that the guilty are way into retirement (or ...).

[Richard 12]

Re: Ah, the great Y2K scare

It's very likely that those companies don't exist anymore.

However, they'll have been bought out and the purchaser won't have any idea...

[Anonymous Coward]

Re: Ah, the great Y2K scare

Indeed, and it's a problem that we have to consider - working on something that's intended to be in use well after I've shuffled off my mortal coil. We have a rather long list of "interesting" dates. These include the obvious (e.g. the Unix 2037 rollover, the earlier NTP rollover, and dates like 2030 that were used as a bodge to put off the problem, and ...) It is a very interesting problem given the number of systems involved, and the fact that many vendors are "reluctant" to discuss the internals of their products.

[catprog]

Re: Ah, the great Y2K scare

Sounds like the issue going around where people can't use credit cards with an expiry date in 2030

[Robert Carnegie]

Re: Ah, the great Y2K scare

Interesting. And not a problem that I have (not yet on my personal card), but I can see in Google a couple of reports of it - but one with no answers, and another on Reddit where I can't read it for some reason.

There's a lot of coverage of Mastercard planning not to use 16-digit card numbers by 2030, so those results come up, instead.

[Yorick Hunt]

POS

Back then I was writing point-of-sale software, and though everything I wrote handled years as four digits, there was plenty of legacy code that needed to be tweaked (not least of which the data being sent to upstream AS/400s).

The process was simple enough - after agreeing on a format change with the AS/400 developer, I simply added the appropriate century depending on the year - 19xx for years greater than 30, and 20xx for years less than or equal to 30. The oldest date record at the time was 1939 (the DoB of the oldest staff member at any of our customers), so no issues there.

If any of that stuff is still running in five years' time from now, well, fugg'em for not having upgraded in the past quarter-century :-P

[Giles C]

Re: POS

All I will say to that is there is nothing so permanent as a temporary fix.

So they are probably still running some form of that code, or it is sitting idling until it detects the “wrong” value.

Who knows?

[Richard 12]

Re: POS

Absolutely.

About 20 years ago a colleague of mine visited a site and found that an important Cat5 run had been introduced to a rapidly spinning toothed metal disc. Unsurprisingly, neither cable worked.

As the place was supposed to go live the next day, they stretched one of their test patch leads across the rack room at about waist level and was promised that the electricians would run a new pair of cables.

I went back to the site a few years later, and had to limbo under it...

[Pomgolian]

Y2KY Jelly...

...making 4 digits fit into the space of 2..

[Bebu sa Ware]

Re: Y2KY Jelly...

Took a while for the penny (1d not 1p) to drop.

Works on whole lot of levels.

The "4 digits fit…" …… well never mind … what's the internet for if not to provide graphic and salacious visual illustrations.:)

I am grateful my coffee cup was already empty…

[phuzz]

Re: Y2KY Jelly...

Took a while for the penny (1d not 1p) to drop.

If it takes more than one hour to drop, you should consult a doctor.

[Jamesit]

At least Cane was Able to to figure out it was a screensaver problem that caused the PCs to crash.

[Pickle Rick]

You actually posted that! I don't Adam and Eve it!

[21st Century Peon]

That an Eden better pun than the OP's!

[Robert Carnegie]

Not evidently proven - perhaps verified later - but it does sound like where a "divide by zero error" could occur.

I suppose you test it by setting the time on a PC to 23:59 and then letting the screen saver run, and if it goes bang at 00:00 then there you are.

[J.G.Harston]

That doesn't seem to be "not testing against the Y2K bug" but "not testing against an end condition". If it was Y2K'd-bugged it would have gone from "1 second to..." to "36524 days to...." It was actually *CORRECTLY* functioning, showing a negative amount of unelapsed time to a past event. The 1st January *IS* minus one days before 31st December the previous year.

Regominised as Cane because he wasn't Abel?

[Pickle Rick]

To paraphrase Michelle (talking to Fry): Why must you analyze everything with your relentless logic? Nothing matters but our love of a good story.

+1

[Doctor Syntax]

That doesn't prevent something unprepared for and hence unpredictable happening at zero. The clue in the story that this happened was that the PCs running crashed.

[doublelayer]

It would be correctly functioning if it just counted down through zero into negatives, but that's not what happened. I'm not sure what happened, but it evidently crashed at zero, which is not correct behavior or necessary. I keep assuming that there was a divide by zero in it somewhere except I can't find a reason why it would need to divide anything by the remaining time and that shouldn't have crashed the system, just the application. So while you're right that it wasn't a y2K bug, it was definitely not correctly functioning.

[C R Mudgeon]

"that shouldn't have crashed the system, just the application."

I don't think we know that it's the screen saver that crashed the systems. Perhaps its Y2K bug was innocuous, though of course embarrassing, but perhaps there was a second Y2K bug, in some more critical subsystem, which took the systems down.

In fact, the directive to shut down the PCs over that critical night suggests that inability to handle the century rollover was a known issue.

[doublelayer]

Anything is possible if we just assume that, but if there was a bug in something else, that should have been in the article. Explaining its absence from the article requires a lot more assumptions including that this was known by somebody not including the people employed to work on Y2K problems in advance; not fixed, mitigated (terminated the affected application before midnight), or hidden; only had a problem at midnight and did not continue in crash mode afterward; and got eventually fixed by someone who didn't tell the IT people about it. That's a lot of things to need to believe and no evidence to use for them.

I think Occam's razor applies here. I can also paint a picture of how this could have been caused by a screensaver. I can't prove this any more than you can prove your story, but since even the reporter doesn't have all the information, we can go with it. The screensaver wasn't intended to count through zero and into the negatives. It was planning to do something special when the timer reached zero, such as displaying an image telling IT staff to start looking for problems, but when trying to do this special case, it did something with RAM it shouldn't have in what was likely a Windows 9x desktop so therefore the screensaver was running with enough permissions to take down everything. The check was only looking for remainingTime == 0, not <= 0, so it didn't trigger again.

[Manolo]

"Regominised as Cane because he wasn't Abel?"

I was thinking Cane because his employer distilled rum?

[An_Old_Dog]

Definition of "Correctly Functioning"

... in my book, does not include a side-effect of crashing the system.

[seldom]

Re: Definition of "Correctly Functioning"

You've never worked at Microsoft I assume.

[FifeM]

I discovered that my client at the time had a Y10K problem - years were 4 digits. I offered them a fix, complete with a money back guarantee if it failed, but they declined.

[Neil Barnes]

I note in parentheses that the MS FAT32 file system will fall over in 2107... It's been deprecated for years but it's still one file system that _everything_ understands, so I'm sure it will still be around.

[An_Old_Dog]

Y2K - Learning to be Paranoid

We tested/remediated everything for Y2K, pre-Y2K.

We tested systems which the vendors said they had already tested.

We found one system, claimed by the vendor to have been Y2K-tested, either had not been tested, or had been sloppily-tested.

In our tests, we found that on March 1, 2000, the system deleted the previous month's patient data. This system tracked experimental-drug results.

This failed result was kicked up the food chain. I believe boiling oil and solicitors were readied. We received a patch a week later.

Which we tested. (It worked properly.)

Myself and another software engineer were on-site that fateful eve, and nothing bad happened.

Because we had tested.

[Anonymous Coward]

Re: Y2K - Learning to be Paranoid

As Bebu sa Ware posted below, 2000 was a leap year, one of the weird ones (years divisible by 400 are, divisible by 100 but not 400 are not). Given your March 1 issue would have been the day after February 29, it might have been a leap year problem rather than a Y2K one. (Yes, it was in 2000, but the Y2K bug is specific to the 2-digit year problem.)

[Will Godfrey]

Re: Y2K - Learning to be Paranoid

I remember working though the logic to implement a full calendar program in C on a very primitive machine, with just a basic CLI only interface. It worked and managed Y2K, but I never want to do that again!

P.S. It dove a factory throughput per shift display system... and was deeply unpopular with the worker bees.

[Old Used Programmer]

Re: Y2K - Learning to be Paranoid

For part of the 1990s, I worked at a company that had an in-house set of date handling subroutines that we were all supposed to use. At one point, a new version of them was released, so I tested a couple of critical dates. On the first pass I discovered that it showed 1900 as a leap year, so I set the programmer who'd done the work a note about the century year rule. One the second pass, 1900 wasn't a leap year and neither was 2000. Another note was sent. On the third pass he got it right.

[Anonymous Coward]

Re: Y2K - Learning to be Paranoid

Unisys 1100 systems got bit by the century leap year fun. They were redoing the system log format. Totally changed so put in a double word (72 bit) time format that counted in nanoseconds since January 1st 1900. Yes, the coder didn't know 1900 wasn't a leap year.

The system log changes hadn't been release yet, but one of the big customers was doing testing and wrote their own date print routine and noticed it was off a day from what it should have been.

So things had been in in testing for too long, so they didn't "fix" it in the code, they just documented that the time started at December 31st, 1899.

[C R Mudgeon]

Re: Y2K - Learning to be Paranoid

It's not a bug; it's a feature.

[I could be a dog really]

Re: Y2K - Learning to be Paranoid

Ah, the Microsoft approach - don't fix the problem, document it as a feature. Look at the specs for the Office Open Document format for spreadsheets - there's a bit in there that codifies an Excel date bug.

[Robert Carnegie]

Re: Y2K - Learning to be Paranoid

I think I found Forbase+ on SCO UNIX was Y2K patched but didn't know that 2000 is a leap year. Fortunately it wasn't quite a critical tool for us at the time.

It's an odd mistake, since if you just take every fourth year as a leap year, then you're OK from 1980 to 2099. Maybe it had some kind of lookup table of leap years and it stopped at 1999 because "in 2000 we're ####ed anyway". In which case it wasn't actually Y2K patched. But this is me guessing shortly before 29/02/2026. :-)

[Robert Carnegie]

Re: Y2K - Learning to be Paranoid

Foxbase+ - I thought I corrected that. But then, I thought I got it right in the first place.

[Bebu sa Ware]

Fizzer

After a lot of very expensive ffaffing around and putting stickers on PCs not much happened on 1.1.2000 and wouldn't have mattered much as the whole show was on vacation until the second week of January. It was the sort of place that was pretty certain it was still the Century of the Fruitbat and likely consumed the entire annual production of Dried Frog Pills.

The wall controls for the aircon was showing 19990 but the day·of·week [Mon·Fri] and time·of·day were correct (or not as it never grokked daylight saving) so I don't imagine it was ever fixed.

The only Y2K problem I encountered in March was a *really old* SGI workstation whose libc didn't think 2000 was a leap year.

I hope the Regcommentariat had a satisfactory Hogfathers with little sticking your craw other than an aberrant fragment of turkey or sprout^† and that the clouds were spared an earful. But never mind 2026 is shaping up to be even more dismal than 2025. :)

_{† Brussels sprouts: Brassica oleracea—grown for its [debatably] edible buds.}

[EngineerAl]

Y1999 Problem

I worked for a municipality in the traffic (roads) department, and colleagues working on Y2K found a problem in certain traffic signal controllers that would fail* on the roll over from December 31, 1998 to January 1, 1999. Happily this was found in plenty of time before year end 1998.

* I don't know now what the failure would have been, but it needed fixing.

[Phil O'Sophical]

Re: Y1999 Problem

Just a standard off-by-one problem...

[Diogenes]

Re: Y1999 Problem

Or using 99 as a guard condition

[PB90210]

Re: Y1999 Problem

I vaguely remember there was a problem with some credit card swipe terminals.

They had passed the usual y2k tests but crashed the week before y2k because they calculated a 'settlement date' a few days ahead but that hadn't been thought of... and that happened to coincide with everybody's last minute shopping

(y2k rollover was fine because the settlement date was long passed and no longer a problem)

[Antony Shepherd]

At the company I used to work for I found a bunch of Y2K problems which were my job to fix.

The problem was that to fix these problems properly a whole bunch of other stuff would need to be changed right up to the software system we sold to our customers (major banks and financial trading companies around the world), which also meant a big change to a piece of code the boss had declared sacrosanct and Must Never Be Changed.

So instead a bodge was applied which would fix it for now but the code would Go Horribly And Terminally Wrong at some time in the future.

"By then", I hoped, "It'll be someone else's problem not mine."

Luckily, the company went into liquidation in the mid 20teens and the bodge was nobody's problem any more.

[Little Mouse]

As a CS student in 1992 we learned all about the impending Y2K issue. A year in industry was also a part of the course.

I did my stint at ICI, and inherited a bunch of home-brew software modules from the guy before me. The year field had still only been allocated two digits, but the Y2K problem had been simply offset with some basic arithmetic hand-wavey programming.

IIRC, it should all collapse in a heap in 2075...

[Peter Christy]

Conned!

A close relative of mine was working for a small manufacturer and im/exporter of hobby goods. They didn't have any real internal IT bods, having bought an off-the shelf accounts/goods tracking system. All they needed to do was the occasional reboot, and if an update was needed, call in the original supplier. I visited the place once or twice and noticed that their system bore quite a resemblance to the Newsroom system that was employed where I worked in broadcast TV. One day my relative informed me that the company had got some "experts" in to Y2K-proof their system. "But isn't it Unix based?", I asked. "Er, yes!", replied my computer literate relative, "but these guys turned up, bamboozled the "Powers-That-Be" with all kinds of bovine excrement, and landed an expensive contract to "fix" the problem!" "But didn't you tell your boss?", I asked. "Yes, but who are they going to believe - a teenage intern, or a fancy sounding IT specialist?" And the rest is history.......

[C R Mudgeon]

Another Y2K screen saver

I thought I'd told this, but a search doesn't turn it up, so here goes...

I had a couple-of-months' gig at a major stock exchange in 1999 (not on Y2K work). The screen saver on all the desktops was counting down, but not to New Years Second as one might expect. If I reconstruct correctly, it was to 10:00 AM on Tue, Jan 4. At any rate, their zero hour was the start of trading on the first business day of the new millennium.

[RobDog]

Banking, your early morning call sir

Yep I supported the dealing platforms for a small but high-profile bank for this event. Months and months of prep, upgrading the many bespoke applications that had been running for years. They put me up in a hotel up the road at considerable expense as you’d imagine. I was a little keyed up about things, not the technical side but the reporting upwards side if anything went wrong…5am on site start… I went to bed nice and early and didn’t touch the mini-bar…and at 1am some dickhead pulled the fire alarm…we got evacuated and had to mill about outside for an hour. Obviously no chance of any more sleep so I showered and went in at 4am to witness - nothing. They noted my attendance and thanked me for my service with a generous OT rate. I’m still working for the same company.

[CorwinX]

On the night when the clock ticked over

I was put up in a very nice hotel very near to the office of the company I worked for.

With remote access to all the systems from my laptop.

All expenses paid.

Spent most of the time raiding the mini-bar, ordering room service and reading a book.

And fielding phone calls asking me if everything was still OK.

They almost seemed disappointed that nothing whatsoever happened.

Still, a nice break for me.

[Marty McFly]

Bastards

I had to work on New Year's day 2000, just in case the world ended. This was after two years of making sure customer x286 systems from a decade earlier were properly upgraded. I got promised an 'exchange' holiday for working. Okay, I am a team player and all, so I reported as ordered. Got RIF'd a few months later... "Your job is going to Atlanta, you are not going with it."

And do you think they remembered the extra holiday I had earned in my severance package? FSCK'ers.

Yup, totally trivial and has no bearing on where my life is today. Yet it still taints my memory of Y2K, enough that I post it here for the world to see.

[Yes Me]

Re: Bastards

What did the bastards expect you to actually do if the world ended? (But at least in that case they wouldn't have owed you a day off.)

[Hawkuletz]

Surprised that no one mentioned jwz's prank (that seems quite close to the spirit of the story if not the fact)

https://www.jwz.org/blog/2012/02/pre-millennium-tension-the-dali-clock-y2k-easter-egg/

[Pirate Peter]

2038 is the next fun date

https://en.wikipedia.org/wiki/Year_2038_problem

2038 is a time computing problem that leaves some computer systems unable to represent times after 03:14:07 UTC on 19 January 2038.

The problem is systems which measure Unix time—the number of seconds elapsed since the Unix epoch (00:00:00 UTC on 1 January 1970)—and store it in a signed 32-bit integer. The data type is only capable of representing integers between −(231) and 231 − 1, meaning the latest time that can be properly encoded is 231 − 1 seconds after epoch (03:14:07 UTC on 19 January 2038)

so that will be the next big party time for IT

[catprog]

Re: 2038 is the next fun date

2030 will be fun.

So many systems were written expecting 29 to be 2029 and 30 to be 1930.

And I don't think many people are aware of it.

[dmesg]

After Y2K

Unlike many commenters here, I was not gainfully employed prepping for cyber doomsday in the BeforeY2K run-up. But during that time I did faithfully follow the web comic AfterY2K, by Nitrozac.

Anybody else remember that?

Icon for what *didn't* happen.

[Anonymous John]

"That became an issue as the year 2000 approached, because as the clock ticked over into the new year some programs would assume it was the year 1900 and malfunction … perhaps catastrophically."

And as computers didn't exist in 1900, they would simply disappear.