Crashing iPad App Grounds Dozens of American Airline Flights
infolation writes: American Airlines was forced to delay multiple flights on Tuesday night after the iPad app used by pilots crashed. Introduced in 2013, the cockpit iPads are used as an "electronic flight bag," replacing 16kg (35lb) of paper manuals which pilots are typically required to carry on flights. In some cases, the flights had to return to the gate to access Wi-Fi to fix the issue.
Electronic Flight Bag (EFB) software is an essential tools for aviation. One iPad can handle multiple charts, maps, and devices which would can weight of more than 20 lbs. Jeppesen software is the American Airlines is the corporate EFB software. A recent update crashed. The Jeppesen tool is a well known company and has Aerospace level of testing. It still failed. There are other EFB tools out there. This has nothing to do with WiFi and everything to do with software development.
The US naval ship Windows NT crash meme is somewhat of a myth - there was a testbed ship (USS Yorktown CG-48) running an experimental ship management and integration system. The crash did indeed occur, but it had nothing to do with Windows NT and everything to do with invalid data being entered into the apps management system causing all linked systems to stop working. While everyone jumps on the "Windows NT" aspect of this, it would have happened under Unix as well.
Software glitches leave Navy Smart Ship dead in the water
a database overfow error (resulting from a divide by zero operation) caused the ship's propulsion system to fail.
The Yorktown Affair
It seems reasonable to have three tablets on the flight deck, running iOS, Android, and Windows 8 for Atom.
The app crashed, not the OS. So having multiple OSes may help in some situations, but not in this one. Some mission critical applications are implemented by two teams working independently. Since this app is basically just a PDF reader with a customized menu, that should not be difficult.
No can do.
The problem isn't the iPad. Or the application. It's that one particular updated doc caused a problem.
And by flight regulations, EVERYONE has to carry the latest revision of the document. And every document is on a different update schedule.
Some documents are changed only when there are updates. Other documents have fixed expiry dates and must be updated to the latest version before that.
And at all times you must have the latest available updates - sure there's maybe a week of grace when the new edition comes out before the old edition expires, but that's about it.
In the paper world, people were actually employed to go through all 35lbs of documents ensuring the latest versions of every page were present (pages are usually supplied as differences in binders, so you remove the old page and stick in the new page. Pages were versioned (typically by date) and there's often a cover sheet saying what's the latest version of each page (updated every time there's an update).
Of course, if you have hundreds of pilots each having to do this, eventually the human version of patch(1) will screw up, so you need to double check for this.
It's why EFBs have been so widely embraced - not having to have someone check 35lbs of documents practically daily, not having to have a whole infrastructure set up to distribute updates, not having to spend time updating documents, etc, it's a terrible chore.
In fact, given the number of updates and how long it's been going on, it's surprising it's only happened once that an update screws up - I'm sure in the past with paper it happened dozens or hundreds of times a day because updates happen that often, usually to different subsets of the pilots.