Why were people so scared of Y2K? How could dates changing to 2000 affect computers?
 Thalidomide Vintage Ad Shirt $22.14 |
 DMT Has Friends For Me Shirt $21.68 |
Thalidomide Vintage Ad Shirt $22.14 |
Why were people so scared of Y2K? How could dates changing to 2000 affect computers?
back in the day it was common practice everywhere to refer to the year using only two digits (like say, '85, '92, etc), and there was some cases of programs being implemented internally like this as well, only expecting two digits for the year, so when 2000 came long, some programs may have rolled over to "00", meaning the year apparently went backwards, which can confuse some programming, since it will have been designed for the year to only move forward
there was also some cases of the year becoming "100", which can break things in different ways, like i remember seeing pictures of websites where the date was listed as "January 1, 19100", as it was programmed such that it just appended the two-digit year to an assumed "19"
also, i know that sounds like a dumb mistake to make, but remember it was the 20'th century for almost 100 years at that point, and for the entirety of electrical computing, so it's not surprising that the everyday habit of two-digit years would seep into some programs
it seems silly now only because we've now lived through a millenium change, and the scare that was y2k
The dangerous part comes from if an important piece of infrastructure or something similar is set on a clock/timer that wasn't Y2K compliant.
>bridge is supposed to go up automatically at 1/1/(20)00 oops its 1/1/00, bridge didn't go up
similary, we saw this same issue — albeit with a significantly less degree — when browsers got to version 100 and some websites that waren't prepared for a three digit UA
first reply, best reply
And nothing happened.
Year 2038 will be a fun one too (32bits int overflow)
now this one is more fundamental, since you don't even need to have made any kind of oversight or lazy shortcut for that to frick you over
that one is a fundamental component of a very widespread time counting format
people have already exploited that, too, like with the iphone 1970 hoax
How are we gonna solve this one, bros?
change all the code to 64 bit and dont worry about it for some millennia
upgrade to software using 64bit time instead
>(2^31) seconds = 68.0511039 years
>(2^63) seconds = 292,277,266 millenia
This. And we better start soon, because everyone knows some Unix(like) systems out there, even connected to the Internet, are running very outdated OS versions. If we keep rolling out systems using the original 32bit Unix time, they might actually still be in service in 2038.
Upgrade from debian stable to testing.
Year 2038 is to the computer what year 2000 is to the human being.
Pretty sure some heavy shit's going to go down in 2038.
Goddamn those homosexuals for not using an unsigned 32-bit integer, which would've pushed this problem back to 2106 or something.
using an unsigned int would have been moronic, since it would have meant they couldn't record dates of past events, since the epoch was just around the time the format was made
also, i really doubt anybody there really expected people to still be using it 68 years later
>also, i really doubt
Dude this computer thing just started and everyone was punching these cards LOL. I do not think they where thinking anything or understand that this computer thing will be big YO.
If DOS was the dinosaur age then these times where after the big bang before any heavy particles started to form. Only hydrogen atoms everywhere.
>also, i really doubt
Dude this computer thing just started and everyone was punching these cards LOL. I do not think they where thinking anything or understand that this computer thing will be big YO.
If DOS was the dinosaur age then these times where after the big bang before any heavy particles started to form. Only hydrogen atoms everywhere.
>this was implemented before home computers were a thing
See above.
Extra byte for past dates, most use of time functions are actually for logging things.
Not unlikely, outside of consumer products, which are the critical ones, which you'd know, if you weren't a homosexual.
computers weren't a consumer product in the early 1970's
Making it even more unforgivable, if they didn't expect a standard to outlast its runtime.
Remember, a standard, not hardware itself.
There are very likely ISO and DIN standards in use that are at least a century old.
Nope, just a bit somewhere that instructs code to interpret it negatively, easy.
68 years is not that long, they could've anticipated this.
But 292 million millenia is fine, go ahead and take the bit.
was it really a standard at that point though?
it's easy to consider "unix" a (set of) standard(s) now, but in the early 70's it was just one of many operating systems, the time format was probably already in stone before they even considered porting it to something besides a PDP
>68 years is not that long, they could've anticipated this.
See
My theory is that they where thinking something like this
>And then we make Unix time 2.0 who uses the year 2000 as the year zero
I think during that time when most shit was printed out of the computer and there where what 30 computers in the world simply changing the format was considered normal.
So a full stop and start under this new system of all computers would have bean acceptable.
Punch cards, paper print outs.Also memory is expensive YO !
>Extra byte for past dates, most use of time functions are actually for logging things.
handling past dates separately sounds like it would be a pita, especially for things that need to support past and future dates (which is a lot of things, including databases, which was an extremely popular task for computers early on)
reserving a bit for it was a perfectly sensible thing to do, and i'm sure they were certain it would be long replaced well before 68 years had passed, this was implemented before home computers were a thing
>unsigned 32-bit integer
Are you fricken insane ?
Signed integers can let you detect overflows since if the year is negative this means something fricked up.
>which would've pushed this problem back to 2106 or something.
Or change to 64 bit that lets you push the problem to the year
>Using a signed 64-bit value introduces a new wraparound date that is over twenty times greater than the estimated age of the universe: approximately 292 billion years from now
Yea that one.
Move from 32 bit 64 bit unix time.
I expect more hardware to fail in 2038 however.
>Signed integers can let you detect overflows
Wow.
Unlike signed, unsigned is actually fully defined by the standard, by the way.
Learn C and talk to my gayot.
It helps.
You suffer from Dunning-Kruger.
>The problem with this is that according to the C standard, signed integer overflow is undefined behavior. In other words, according to the standard, as soon as you even cause a signed overflow, your program is just as invalid as if you dereferenced a null pointer. So you can't cause undefined behavior, and then try to detect the overflow after the fact, as in the above post-condition check example.
https://stackoverflow.com/questions/3944505/detecting-signed-overflow-in-c-c
>And nothing happened.
people worked around the clock to fix it in time you moron
>If only we could change the date on these devices and see in advance what will happen.
You can't do that to a live banking mainframe numbnuts
>what is a development environment
>And nothing happened.
There where incidents.
https://en.wikipedia.org/wiki/Year_2000_problem#Documented_errors
>On 1 January 1999, taxi meters in Singapore stopped working, while in Sweden, incorrect taxi fares were given.[47]
>On 28 December 1999, 10,000 card swipe machines issued by HSBC and manufactured by Racal stopped processing credit and debit card transactions.[16] The stores relied on paper transactions until the machines started working again on 1 January.[48]
>In Japan, NTT Mobile Communications Network (NTT Docomo), Japan's largest cellular operator, reported that some models of mobile telephones were deleting new messages received, rather than the older messages, as the memory filled up.[51]
>Why were people so scared of Y2K?
There were many bad programmers also back then.
>How could dates changing to 2000 affect computers?
Rollover would landed u sin the year 19101 or worse. Back then I was in a team to audit out old products and there were components that just refused to boot up past year 2000. We fixed the problem by setting the local clock to 1970.
>And nothing happened.
That is wrong.
>Year 2038 will be a fun one too (32bits int overflow)
We also had a week 1024 rollover that also caused havoc. It could happen again.
If only we could change the date on these devices and see in advance what will happen.
They wanted to practice scare tactics on the population at large, then they did 9/11, now we have even more sophisticated scare tactics like masks and vaccines.
congratulations, zoomer. you just discovered Y2K bug. you have been browsing the aesthetics wikia arent you?
In the years leading up to it, manufacturers put Y2K compatible stickers on all kinds of tech, even shit that had no chance of being affected by it, like 56k modems.
Same reason people were afraid of 5G radiation.
Tech illiterate boomers and normie morons.
>How could dates changing to 2000 affect computers?
ZOOM ZOOM does not understand integer overflow.
>How
Order these numbers from biggest to lowest
01
15
08
25
39
80
99
00
Now only use 2 digits for the year
Tell em what is bigger
98 or 99
What is bigger
01 or 00
What is bigger
99 or 00
1999 + 1 = 2000
or
99 +1 = 00
Fun fact it messed with systems who did not upgrade you will have the same thing in
The end of unix time is coming.
>How
Zoom zooms do not understand technology.
I'm not a zoomer, I'm 43
God I wish Y2k would have ruined everything.
It was a psyop by tech companies to sell support contracts to the computer illiterate.
If you have an old Mac and set the year to 22, it thinks it's 1922. I think DOS 6.22 was Y2K compliant but older versions (not sure which ones) weren't.
Dates are hilariously mishandled in CS in general. Take all the bad habits of lazy programmers you know now and set them back a few decades as all this is newly discovered. Keep in mind that modern computing still mostly relies on counting seconds from the Java epoch in 1970.
Examples:
>Print year? Easy, just concatenate strings 19 and variable limited to ranges 00-99
>1999 to 1900
>Print year? Easy, just concatenate strings 19 and variable that increments from current year (99)
>1999 to 19100
Space was a commodity back in the days when floppy disks were more prevalent than CDs. We've always had the habit of carrying forward or porting legacy systems instead of outright replacing them. Look at point of sale systems if you want an example, that shit's written in COBOL or PASCAL.