Is this the crash where the pilot failed to recognize the airspeed sensors had frozen up and he stalled the plane? I could see how this was an Air France fault since the pilot was not properly trained or experienced to fly this plane in these conditions. Not sure why Airbus is responsible.
it's the crash where pushing nose of the plane down (correct enough-altitude stall response) caused alarms to activate, while pulling nose up caused alarms to silence
Airbus kind of embodies the "trust the computer" mentality; and if you're going to do that the computer damn hell better be right all the time - it must not have "backwards" failure modes.
Boeing, in similar situations "in the past" would just sound a "computer is giving the fuck up, fly this pig dog" bell and leave it to the pilots to figure it out.
Comparing Boeing's compliance hack and Airbus' system that's pushing 40 years now is very questionable. Airbus planes don't get in the way of flying, and there's extensive procedures and redundancies for everything that could go wrong. It's a proven system, and events like these are the exception proving the rule, especially since there was also a human factor here.
As another computer person, I'd trust aviation more than any other field, especially when it doesn't involve the modern US. Computers can't be perfect, but they can be almost always good at integrating and helping humans that remain in control. Advocates against including any fly-by-wire or computerization in aircraft at all fail to consider all the accidents that said computerization has helped avoid. Putting a billion steam gauges and blinking lights in front of pilots and asking them to correlate and understand everything themselves is actually not simpler, easier or safer.
The same thing happens on the 777 and 787: if too much opposite force is applied on both yokes, they lose their linkage and are averaged. There is no warning or priority button, unlike on Airbus planes.
Older Boeing planes also have a mechanism to unlink the controls if too much opposite force is applied. The left yoke would control the left side of the plane, the right yoke would control the right side.
I don't really understand what's so jaw dropping about input averaging. Let's be clear - this is a fallback state that handles a situation that should never come up. Pilots aren't supposed to try to control the aircraft from both seats at the same time, both fly-by-wire and not. What we're talking about isn't a deficiency that can sporadically cause a dangerous situation, like the MAX, but a situation where the pilots have already made a massive mistake and the automation didn't bail them out. It's not like there's no workaround, either. Making conflicting commands results in the plane blaring a 'dual input' warning at you, and if one of the pilots desires exclusive control, they can press the side stick priority button. A further improvement of the system would be to add force feedback to the side sticks, to simulate the linked yokes of a non-fly-by-wire aircraft, but even without it, I feel like this issue is given way more publicity, and it's used as the scapegoat for the ultimate cause, pilot error. All incidents that involved this were ruled as being caused by pilot error - in the crash this article is about, the PF was literally holding his side stick full back until almost the very end. A force feedback system might've helped them realize it sooner, or it might not have - there's plenty of historical incidents where pilots managed to stall conventional aircraft out of nowhere in a similar fashion, but those were ruled to be their mistake only.
I struggle to think of a situation where "average the pilots' inputs when they disagree enough to sound the alarm" would ever be the expected or correct action to take, especially given the existence of the sidestick priority logic
Really seems to violate the "principle of least astonishment"
The behaviour you describe above only occurred after the pilot flying stalled the plane. There was a procedure for unreliable airspeed indication. Had the pilot flying performed it, the situation would have been resolved without incident.
AF could perhaps be held liable for insufficient training on high-altitude stalls or recognising and responding to reversions to alternate law. But it's hard to see how Airbus can be responsible for a pilot ignoring the most basic first response.
The article from this subthread contradicts this, though. Regarding recoverability of the situation, it says this:
> By now the airspeed indications had returned to normal, but the pilots had already set in motion a sequence of events which could not be undone.
That was before the prolonged stall warnings. But maybe this phrasing is just an embellishment?
But further down, the article is pretty clear that the training was inadequate for this type of unreliable airspeed indication:
> Although procedures for other phases of flight could be found in the manual, the training conditioned pilots to expect unreliable airspeed events during climb, to which they would respond with a steady nose-up pitch and high power setting that would ensure a shallow ascent. Such a response would be completely inappropriate in cruise.
Once the aircraft was stalled there was a narrow window to recover from it, which obviously did not occur. But the stall was entirely caused by pilot input of full nose up! The procedure for unreliable airspeed (which was in both the QRH and the FCOM) was simply to fly a known safe power / pitch from the tables provided in the QRH.
At no time was any of the pilot's Attitude Indicators (Artificial Horizons) inoperative -- all they had to do was maintain straight and level flight at a known power setting and everyone would have come home safely.
While true, pilots aren’t trained to just “respond to the alarm” they are trained to fly the plane.
Once there were multiple alarms that made no sense at all (petty early in the event), the pilots should have ignored them as per the checklist.
But the most damning thing is the one pilot pulling the stick back and holding it back for almost the entire event. There aren’t any flying conditions where that’s an appropriate input. Not to mention being told to give up control and ignoring that request.
I agree Airbus has some blame in terms of the computer system not adequately communicating when it drops out of normal mode.
I stand corrected. But suffice to say it’s not an appropriate input when you lose airspeed at 35,000 ft.
I read the Admiral Cloudberg article again and saw that it was procedure for other scenarios as well.
It seem like the normal mode (protected flight envelope) is just encouraging bad habits? “Just go full stick back and hold it, don’t worry the computer won’t let you stall the plane…most of the time”
> It seem like the normal mode (protected flight envelope) is just encouraging bad habits?
Maybe, but at the same time it helps avoiding crashes like Sriwijaya 182 or Flydubai 981. Airbus has shown that planes with fly-by-wire and any kind of flight envelope protection (A320 and newer, A220, B777 and 787, etc.) experience less fatal accidents and less hull losses than planes with traditional controls (A300, A310, B737, etc.), even today: https://accidentstats.airbus.com/fatal-accidents/
Unfortunately, these safety improvements mean that we only hear about cases where automation fail to help, like in the case of AF447, but not cases where it prevented an accident.
Yeah the computer is never flying the plane it is always the pilots who have final decision. Which is ofcourse also why the computer will let you fly into a mountain if you want.
Convenience. Jellyfin is becoming better and better but still lacks proper support for certain (Apple) devices, while Plex mostly just works.
That being said, Plex apps also lack proper support for HW decoding of some 4K codecs and their AppleTV situation is just plain sad (2 years without an update, no proper HW decoding).
With the Pass being $750, I’m expecting Jellyfin to become a more popular alternative than ever. I’d personally never spend this kind of money on Plex.
> That being said, Plex apps also lack proper support for HW decoding of some 4K codecs and their AppleTV situation is just plain sad (2 years without an update, no proper HW decoding).
What does hardware decoding get me on an Apple TV? It seems plenty powerful enough to decode anything I’ve ever thrown at it.
The official Plex app hasn't been updated to take advantage of all the hardware decoding support available on the Apple TV. I moved to Infuse, a paid third-app, which I use to stream my Plex media content.
Jellyfin official apps are sporadic though, so Infuse is popular for Jellyfin users because it is well maintained and allows you to direct stream content without transcoding as often as possible.
--
I'm also the developer of Aviato Media Server (https://aviato.media), which I just released publicly last week and is available only as a beta, and I am an ex-Plex employee.
That doesn’t answer my question. What does it get me? Why do I care if it uses hardware decoding or not if I can stream pretty much any content without any issues? It’s not like it causes stuttering or wastes a bunch of power because it’s using software decoding.
Depends on the server you are using to host your media sever from. If you're using a Raspberry Pi, then you might not be able to support transcoding at all, definitely not for a 4K AV1 video and a lot of modern formats. So even if Apple TV can play it, the Plex app won't attempt it and will force transcoding from the server.
Even if your server can support transcoding, all transcoding results in quality loss, so if you want to enjoy your media in the best formats then you want to direct play. Just depends on the media you are playing and how much you care about the quality.
This has nothing to do with hardware decoding though, no? Hardware decoding generally is more restrictive in supported formats, not less restrictive. Unless you mean AV1 decoding is supported by Apple TV in hardware but not the Plex app itself…
Also, what modern formats for media typically used in Plex servers isn’t supported? Nobody uses AV1 for plex media, no offense. It’s universally h264 or h265. Or perhaps mpeg2 for super old stuff.
Weight is only one dimension. What matters is weight and mileage. Our family has two cars but we drive less combined than a typical single person in our area.
Where that gets problematic is you can't easily validate mileage for something like an online renewal.
I will say you did give me a fun idea however; you could in theory set up a kiosk system where scanning renewal notice + license[0] and get some sort of OBD2 plug-in device, that could record the mileage+vin+other data (to harden against faking attacks[1]) and then you take it back to the kiosk to confirm mileage.
IDK, probably not the best way to do it but maybe there's a good way to handle this sort of scenario.
[0] - In case you abscond with the dispensed device for some reason...
[1] - But this is possibly where it falls apart...
When I bought my 2024 Lexus, there was a sticker on the headlamp saying to push the support call button to talk to a rep if I don’t want any vehicle data collected . So I did and the rep told me they can disable it but it will also disable the SOS/911 calls and crash report if I do that. Choosing my own battles, I begrudgingly told them to leave it enabled.
As a consumer, I’d like to see the end of “universal” builds for various apps. It made sense for a while but downloading and installing ~60% larger bins just doesn’t make sense 6 years later.
reply