[Kosh]

consider avionics. its something that absolutely has to work. otherwise you get planes falling out of the sky. in addition to atc systems, some of which are really antiquated.

But are any of the onboard avionics of a plane dependant on the year being read correctly?

[NGTM-1R]

But are any of the onboard avionics of a plane dependant on the year being read correctly?

Possibly. At the time I doubt GPS was thoroughly integrated though. The air traffic control system, however, was known to be a significant Y2K risk.

[haloboy100]

I'm still surprised that this was something to be afraid of, seeing as how easily it seems to have been fixed before it happened.

[The E]

Thousands of coders going through tons of legacy code verifying that nothing unexpected will happen is easy?

Sidenote: If nothing else, this sparked major code reviews for a few important systems, and infrastructure audits that resulted in systems that were ultimately even more stable. So even if the fear behind it may have been unfounded, the effort to fix the issue did produce a few good results.

[Nuke]

updated software is always worth the effort.

i just wish they weren't used as an excuse to release a product early.

[Kosh]

But are any of the onboard avionics of a plane dependant on the year being read correctly?

Possibly. At the time I doubt GPS was thoroughly integrated though. The air traffic control system, however, was known to be a significant Y2K risk.

So if air traffic control thought the year was 1900 then the planes would have fallen out of the sky?

[haloboy100]

no, but plane arrival dates and the like would be completely off.

[Kosh]

Off by 100 years maybe.

[karajorma]

And how do you think any system using time to calculate distance is going to be affected if the time is suddenly very very wrong?

[Kosh]

And how do you think any system using time to calculate distance is going to be affected if the time is suddenly very very wrong?

Not much if it was all affected by exactly the same amount. If every system thought the year was 1900 then it wouldn't matter. Either that or you subtract 100 years worth of hours.

[karajorma]

So you're assuming that software that has been maintained for decades will always have been programmed by coders who used the same method for representing the date and that therefore every clock will be off by the same amount?

Yeah, I think we're done here.

[NGTM-1R]

Not much if it was all affected by exactly the same amount. If every system thought the year was 1900 then it wouldn't matter. Either that or you subtract 100 years worth of hours.

Take a couple of navigation classes.

Or alternately, imagine what apparently going backwards in time does to your apparent speed of advance.

[S-99]

:snipe:IMA****UP

I was thinking it through, but your small mindedness had to come in here and not think about the big picture i was alluding to. If my **** lick mac from long ago could survive it, then that makes you wonder what other 2 year digit date supporting machines would be able to do (like survive).
The y2k problem with people worrying about it back then is a little funny. So a computer goes to year 00. That's about it. There were some minor problems with y2k, but that was it. It's not like a clock reaching 00 on an airplane is going to make it fall out of the sky (well, not unless the clock is the fuel cut off switch...rofl). When instead there were some intermittent problems. The few that weren't intermittent, didn't take much time at all to fix, and also were nothing major like planes falling out the skies or all the nukes in the world launching. For god's sake, the US atomic clock thought the year was 19100.

People were all getting stressed out, and the preparation helped, but really just by having people aware and ready for when the clock hit 00/2000. The main thing the y2k wikipedia entry points out is yet again, people that were aware and ready on standby, and some intermittent problems. Certainly planes falling out of the sky? All of the US's nukes launching at once? Yeah right. Concerning y2k, the only reason such destruction would happen when the clock hits 00 is if the vast majority of the worlds programmers were in the same end of the world cult.

[NGTM-1R]

Read the thread plzkthnx.

[The E]

*sound of S-99 not getting it*

You ... really aren't getting it, are you? You are taking your own personal experience with your home computer (which at the time probably didn't do anything critical for your home or business), and apply it to a wide variety of systems, from mainframes running accounting systems designed in the 60s to embedded systems designed for very, very constrained hardware specs, all of which were deemed so critical for operation that people were actually afraid of upgrading them. Systems where noone knew how they were going to react on January 1st, 2000. Systems where you couldn't just run a quick test to see what would happen then.

In other words, you're still saying "I don't see what the big deal was, nothing happened", when I'm saying "No wonder nothing happened, there was a giant effort to make absolutely sure of that".
Now tell me again who is not thinking of the big picture?

[Kosh]

So you're assuming that software that has been maintained for decades will always have been programmed by coders who used the same method for representing the date and that therefore every clock will be off by the same amount?

Yeah, I think we're done here.

Ok, I have a question. Let's just say that we all know flying from point A to point B should take, lets say, 7 hours. Now, if the plane left the airport at 9:00 pm December 1999, and if the computer system thought the next year was going to be 1900 (as per the Y2K bug), what would it say the ETA was?

I understand that you need to know the time, as in hours, minutes, seconds, days, and months accurately, but I don't understand why needing to have the year be spot on is mission critical.

[Bobboau]

because a year == 365 days a day == 24 hours an hour == 60 minutes, you cant have one unit inaccurate without all the others being effected because they are basically the same unit.

[Nuke]

time is very important in any system that does physics calculations, and atc systems are required to triangulate aircraft positions with various radar hardware and compute speed, altitude and and course vector. all of which needs to be rendered on an atc scope. what they didnt want is blips flickering all across the scopes or appearing in the wrong location. and considering the age of the systems and what you stand to loose if they fail, id say it was well worth the effort to go through and fix all the timing bugs. "planes falling out of the sky" was really just a worst case scenario, and even without atc the planes would continue to function (in theory ), and if atc becomes unreliable they could have reverted everyone to vfr flight while they rebooted the systems and landed planes as necessary. plus you got everyone on alert so the likelihood of something bad happening was slim.

[Kosh]

because a year == 365 days a day == 24 hours an hour == 60 minutes, you cant have one unit inaccurate without all the others being effected because they are basically the same unit.

Ok, so in my above example it would read a huge negative number of hours, correct?

[karajorma]

Or it would crash the ATC systems. Or make them give poor readings which resulted in the pilots getting the wrong landing information.

I really can't see why anyone would have so much trouble accepting that an aeroplane or ATC system having bad data isn't a serious potential risk.