Live data from Hacker News

Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

flightglobal.com

61–70 of 311 posts

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#61
post #37

Earlier quoted context omitted.

The question isn't really how many would think it meant nanonmeters, it's how many would recognise it as nautical miles.

In aviation? Literally everyone. Context clues should make it obvious.

And how many would assume that the incorrect unit abbreviation would mean something else? If I said Kbps and KBps, it's an entirely different unit of measure. NM and nm are VASTLY different, and unless you are already familiar with measurements in Aviation and know it's Nautical Miles, your first instinct is gonna be to read it as the unit it's supposed to be.

The article title is talking about location data and computers, I've seen many people forget floating point precision when comparing and getting bit by tiny differences at the 10^-9 or smaller. That seems just as obvious on the outset as non-unique location designations in what the average person would assume to be a dataset that's intentionally unique and unambiguous.

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#62

Earlier quoted context omitted.

[flagged]

In aviation? You know how big an aircraft is right?

> In aviation? You know how big an aircraft is right?

Exactly, which makes the headline particularly intriguing—how could a 3600 nanometer difference matter? The standard resolution, which I pursued, is to read the article to find out, but it doesn't mention the distance at all.

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#64

Earlier quoted context omitted.

> he system can't recover from an error in an individual flight plan, bringing the whole system down with it From the system's POV maybe this is the right way to resolve the problem. Could masking the failure by obscuring this flight's waypoint problem have resulted in a potentially conflicting flight not being tracked among other flights? If so, maybe it's truly urgent enough to bring down the system and force the h…

Yes I agree. The reason the system crashed from what I understand wasn't because of the duplicate code, it was because it had the plane time travelling, which suggests very serious corruption.

Waves hand... This is not the SQL injection you're looking for. It's just a serious corruption.

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#65

You know there's a software engineer somewhere that saw this as a potential problem, brought up a solution, and had that solution rejected because handling it would add 40 hours of work to a project.

C dev: "You are telling me that the three digit codes are not globally unique??? And now we have to add more bits to the struct?? That's going to kill our perfectly optimized bit layout in memory! F***! This whole app is going to sh**"

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#66
post #39

I've posted this here before, but they really need globally unique codes for all the airports, waypoints, etc, it's crazy there are collisions. People always balk at this for some reason but look at the edge cases that can occur, it's crazy CRAZY

Coming up with a globally unique waypoint system is trivial. Convincing the aviation industry to spend many hundreds of millions of dollars to change a core data type used in just about every single aviation-related system, in order to avoid triggering rare once-a-decade bugs? That's a lot harder.

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#69

I don't know how long that failure mode has been in place or if this is relevant, but it makes me think of analogous times I've encountered similar: When automated systems are first put in place, for something high risk, "just shut down if you see something that may be an error" is a totally reasonable plan. After all, literally yesterday they were all functioning without the automated system, if it doesn't seem to b…

Dig into the OpenZFS 2.2.0 data loss bug story. There was at least one ticket (in FreeBSD) where it cropped up almost a year prior and got labeled "look into layer," but it got closed.

I'm aware closing tickets of "future investigation" tasks when it seems to not be an issue any longer is common. But, it shouldnt be.

Re: Air traffic failure caused by two locations 3600nm apart sharing 3-letter code

#70
post #18

Earlier quoted context omitted.

Given the context, I'd say NM actually https://en.wikipedia.org/wiki/Nautical_mile

I was clarifying the post title, which uses "nm".

Yes, it looks like they should have written "NM" instead of "nm".
Post reply on HN