auraflyer

Not quite. People at Xerox PARC invented the idea - but their windows were not able to overlap. They couldn't figure out how to do overlapping windows. (The issue has to do with the arbitrary shapes that can be formed when one window is overlaid by multiple others, which you need to do clipping of content when drawing, and when refreshing.)

Apple toured the facility (with permission) and their engineers saw the interface. Unaware that PARC engineers hadn't been able to do overlapping windows, Bill Atkinson actually worked it out. In assembler. (Anyone who has programmed 68k Mac should be aware of 'regions', which is how it's done.) That was one of the most important parts of getting the interface into sellable form.

It also helped that some of the Xerox people (e.g. Larry Tesler) were so impressed with the Apple people that they joined the Apple team working on the Lisa.

There's a lot of cool material about this on the web, but Hertzfeld's book "Revolution in the valley" is a really great read on this subject.

Edit: mixed up Bill and Andy as the inventor. See the original patent.

Lyman

Hi wozzo. Here we see a drastic hole in the popular myths. "I've lost my VSI". Erm, "How's yours?" This narrow, myopic tilting at published data driven conclusions is getting bizarre.

We are left to flutter in the winds of our need to know, and the thin soup served us by BEA.

It didn't happen this way....

Four minutes is still four minutes, and each fragile crystal of "conclusion" is less than useless when weighed against the totality as recorded.

There is no conclusion that the Thales were the cause, there is no conclusion that the a/c climbed to STALL with Pilot input only, etc. etc.

In a free and open society, isn't it clear that there are stones in the soup?

Yet every (nearly so) uttering is the whine of Pilot error.

It's in there (PE), oh yeah. So is a great deal more, and EASA is holding back.

BEA are not the problem, they are the messenger. And Mercury is muzzled. One has some trust of the science in BEA, and none in the wags who call their orders.

infrequentflyer789. I isolate the A330; it is in the family, but it is the only TWIN which does what it does.

Haven't the 330's been removed from Rio-Paris? See?

wozzo

Same. And therefore subject to the same misinterpretation.

Also, the info does not come from BEA.

Turbine D

Dozy

I've sat on the sidelines watching and reading lately. Your posts were at one time informative and interesting to read, but have turned into total protection of Airbus recently. Airbus doesn't need protection, they are experts at "protections".

I have got to go with gums on this. The computer system, e.g., Alternate Law 1, Alternate Law 2 and subsets of Alternate laws are very confusing. Normal Law and Direct Law are very clear: Normal Law - the computers fly the plane and the pilots watch, Direct Law - the pilots fly the plane and the computers watch. That is simple. The in-betweeners are a mish mash of: You have this but not that except when this is this or that is that....

Think about it, in the case of AF447, they had at most three minutes to determine what was wrong, what was going on and what to do, with the balance of time being on the express elevator to the sea. From a technical and engineering point of view, here is what I see wrong on Airbus' part in this saga:

1. Airbus failed to handle the pitot tube problems in a manner that they should have.

2. Airbus failed in their risk assessment/risk abatement to adequately cover the total waterfront, e.g., total flight envelope. They were nearly mute on flight problems at high altitudes and speeds. Guess they assumed problems couldn't happen there.

3. Airbus failed to provide complete key memorization items in their flight instructions for the A-330, leaving out for the most part, high altitude and high Mach cruise situations.

Now you can say this is Monday morning quarterbacking, but it isn't. It is good common sense, business best practices - risk and risk abatement processes that should have been in place but were not.

You can also say and you have on many occasions, the plane did exactly what it was supposed to do, but in reality, it crashed. That was not Airbus' intent I am sure, nor was it the pilots flying or Air France's intent either.

Like it or not, Airbus has the leading role, it is their airplane (a good one at that), their design, their computer control flight system. Airbus therefore has or should have the leading role in establishing all of the criteria, instructions, do's and don'ts to pass along to individual airlines who buy their planes and then the pilots who will fly them and even train them. Additionally they have the lead role in determining what should be done when critical monitoring/detection devises develop a history of not performing to expected standards or expectations.

Prior to AF447, there was a substantial case history established on the pitot tubes (selected and installed by Airbus) which indicated there was an icing sensitivity problem. A good risk assessment/risk abatement analysis (an industry best practice) would indicate this to be a significant safety item needing rapid attention. Things that happened in flight on other airplanes would be an indicator that sooner or later a flight crew could/would respond incorrectly. As in the role of leader, Airbus must not have/didn't push the issue hard enough with EADS, or so it seems or didn't view correctly the critical nature of pitot icing at high altitude and high Mach.

AF447, three minutes to go (not knowing that at the time), A/P come off, A/T come off, the stall warning sounds, the nose is slightly down and one wing is lower than the other. We are in Alternate, is it 1 or 2? Going too fast? Going too slow? What is the speed? What are the protections? What do we have and what don't we have? What does the memorized list developed by Airbus say? Do we have a memorized list for high altitude/high Mach? That was the situation. Now Airbus did do a good risk analysis/risk abatement for low speed, low altitude situations, landings, T/O's, thus developed the memory lists. But they stopped short. In fact, if you apply the low altitude low speed do's to high altitude/high Mach, it probably makes the problem worse. What happens to THS flight protection in Alternate law? What's the AOA? Confusion? Help? Panic? 3 minutes to sort things out. Cavalry charges, single chimes after single chimes, single chimes every 5 seconds, lots of crickets? Get the gist?

Airbus had the lead on pitot tubes, they selected and installed them, they had the wealth of data regarding icing and what other airline/Airbus aircraft did at cruise when pitots iced. They knew there was a difference between manufacturers and performance/non-performance. What did they do about it? Did they do a risk assessment/risk analysis? Was it deemed to be a hoe-humer or a critical problem when flying at high altitudes and high Mach? What did they advise EADS to do? Airbus has the lead here, not EASA. Did EASA respond appropriately based on Airbus' strong recommendations? Did Airbus have a strong recommendation?

So at cruise in the middle of a moonless night in the middle of the Atlantic Ocean in the active ITCZ at some level of turbulence the A/P quits, and computers go off line, partially but not fully and the plane doesn't even remain in level flight. Now Airbus knew the AOA was a critical key component to assure safe flight, but where was the indicator to know what it was and where it was going? Is there a key memory list, e.g., level wings fly pitch and power? What should be the correct pitch setting when you have no speed indications?

And I could go on in more detail and ask more questions, but I won't. I think there is enough to illustrate that Airbus shares much responsibility as do the flight crew, the airline and even EASA if they relaxed on an Airbus' "Strong" recommendation, if there was one. But Dozy, to continue to protect Airbus is fruitless, it is like pi**ing into the wind, sooner than later you are going to get wet and I think you are getting wet.

Gretchenfrage

What about the digital readout? Should still have been there.
This shows quite clearly that all the advocates about "it's on the screens, just read it; who needs feedback?" are naive:
In stress, humans read badly. We need either tactile feedback or a more clear display than many of the digital displays.
Simply telling the pilots to read the five-color s#!thouse on reflecting and worn-out screens with a multitude of same color digital indicators and FMAs with five windows on three rows with zillions of different abbreviations is, to cite the Borg, futile.....

I do realize that there is no feedback on VSI, but there is a more intelligent display mode than the donkey's dick that just disappears on the lower end.
It's the tape-style color coded indicator on a MD11. It stays filled white and is way more legible, even under stress.

AlphaZuluRomeo

Didn't it?

BEA's interim report #3, CVR transcript

“I’ve lost VSI” is just another translation for “j’ai plus de vario là”.




AF443 : B777-300ER
AF445 : A330-200 (this was AF447 before)

Hardly conclusive, IMO...

DozyWannabe

If that is the case then I suggest you read what I've been posting in isolation from what others (including yourself, it would appear) chose to read into what I was saying.

Then you didn't read what I was saying - which is that at the most basic level, all a pilot has to understand about the control laws is that you have hard protections in Normal Law, but do not have them in any of the other laws. What that means in the case of this accident is that as such, pulling up and continuing to pull up in the face of a Stall Warning is just as dangerous as it would be in any other airliner once Normal Law is lost.

The sub-modes are there purely for engineering and systems management purposes - as I said, think of them as the different systems configurations that would have been handled by the Flight Engineer on aircraft of earlier generations in the case of damage or failure. Gums knows his stuff, there's no doubting that - but he has got it into his head that the Normal Law protections are akin to autopilot limitations, when in fact they give the pilot considerably more leeway than that (I sent him along some of the Flight articles that covered the A320's development as evidence).

They'd have had a lot longer if the PF had not started pulling back on the sidestick and continued to do so for the majority of those three minutes. If this was any other aircraft this would have been an open-and-shut case, but because there are those on here that will continue to insist that the computers must be to blame in any Airbus accident, we're still going over the same stuff months later.


They told the operators to fix it and sent out bulletins to pilots explaining workarounds until such time as the fixes were completed - what more could they have done without grounding the entire fleet (which prior to this accident would have looked like overkill)?

Them and the rest of the industry.

That's a little bit of editorialising from you. I think it's fairer to say that - like the rest of the industry - they assumed that pilots would be sufficiently qualified and knowledgable to recover from most high-altitude problems.

Could you be more specific?

It wasn't Bombardier's or ColganAir/Continental's either, but the sad truth is that if you shave the training budgets to the extent where basic airmanship is deprecated in favour of checking only up to the onset of warnings then you will start to get problems.

Not wanting to nitpick, but do you mean EASA (European safety authority) rather than EADS (Airbus's parent company)? I'm sure that Airbus went into consultation with EASA and the airlines and together they determined what the response should be. For better or worse, it seems to have been assumed by the industry in general that problems at high-altitude by their nature should buy the pilots enough time and altitude to remedy the situation. Even the most extreme examples of UAS upset prior to AF447 led to relatively minor level busts rather than significant control difficulties - so even if Airbus came to the discussion with a serious AD in mind, it is likely that the airlines would have pushed back at the prospect of large chunks of their fleets spending weeks in MX.

OK -so we're not in any danger yet. Let's not touch the controls until we're sure what's happening.

[Actual response : Immediate grip on the PF's sidestick including a significant nose-up command, far in excess of what would be required to bring the nose back up to S&L. ]

Doesn't matter. Anything other than Normal => No hard protections. We need to be more careful with the inputs (though no more careful than in any airliner without protections)

We don't know, so fly pitch and power until things are stable.

See above - there aren't any (or to be more precise there aren't any that will counteract pilot demands).

Pitch and power. That's just airmanship - no memory list should be necessary (although admittedly it would likely have helped).

As did every other manufacturer.

Well yes - the procedure the PF appeared to be closest to following was in fact the low-altitude Wind Shear/Microburst escape - although no procedure was in fact followed to the letter.

Same as the others - it goes away - so we need to be careful.

We're getting Stall Warnings - we've had them for a minute. Why then is our nose at 15 degrees up?

[Answer : Because the PF was holding it there.]

Yes - replace them as quickly as [the airlines] were able (see above).

Same place it was on every other modern airliner - nowhere.

The assumption was probably that basic airmanship would preclude a 15 degree nose-up pitch at high-altitude - unfortunately relying on that does not appear to have been enough.

Which I've never argued with, if you go back over my posts on the subject - my contention has been with those that have argued that there is something fundamentally wrong with the aircraft design and that as such Airbus must carry the whole can.

OK - let me make this as clear as I can - I'm not "protecting Airbus". I'm simply taking issue with some posters on here that are not discussing the factors of the case honestly, but instead are coming to the table with their pre-conceived notions about why they dislike the Airbus FBW design and are trying to hammer the facts of the case to fit the mould of their prejudice.

At least one poster continues to come up with ever more lurid theories about how it must have been the aircraft's fault. Indeed, insisted until *very* recently that the VS must have broken off in flight, then made excuse after excuse including how there must have been a software bug that popped the spoilers on one side, that the THS control software suffered a glitch and ordered a runaway trim without the pilot's knowledge...

Others are trying to turn this into an argument about a 23-year-old accident in which the pilot thought he was better than he was, crashed the aircraft and killed three people (two of them children) in the process - and consistently refused to take responsibility for his actions since, making up excuse after excuse as every one was disproven by the investigation.

All of this comes from the patently false idea that Airbus and airline management were in cahoots with each other to develop an aircraft that would de-skill the job of airline pilots and reduce them to "systems operators" akin to any other office monkey. Even gums, who clearly deserves to be taken seriously in most respects, harbours this idea in the background of his thinking - which is a shame because it clouds the rest of what he's saying.

I have never said that Airbus should not shoulder some responsibility - in fact I've always been at pains to point out that because of the pitot tube issues there is no way they can avoid it. However, I do take issue with some posters going on the "same old sh*t" tour when it comes to incidents involving the FBW Airbus family and bringing up factors that have no bearing on the case (sidestick vs. yoke) or could be applied to the whole industry (lack of preparation for high-speed stalls, recurrent training cut to the bone).

So please - if you're going to level this kind of stuff at me, please have a look at what I'm actually saying - in isolation from what others claim I am saying.

[EDIT : I hope IGh won't mind me borrowing this image from his post on the ANA 737 upset thread in R&N, but it sums up more succinctly than I could what I believe to be the crux of the issue regarding the crew response (mitigated in this case by a poor attitude to training on the part of the industry):


]

wozzo

Ah, OK. So the new report indicates that investigators (BEA or others) seem to be sure that VSI indeed was functioning properly and the PF was misreading. Small step, I know.

averow

Very fine and succinct summary. The problem going forward is to reinforce that there is no substitute for CRM. Rare events can and do happen if one gets lulled into a sense of complacency in ones profession. Trainers, designers and airline companies need to acknowledge this paradox: technology and design may be lulling the next generation into thinking that the machines are foolproof.

lomapaseo

These kind of technical discussions will get out of hand if we decide to express legal arguments about fault in this forum.

It's bad enough about expressing opinions about how to redesign a complex system, although soothing to the soul. Expressing desires for this and that seems to be quite adequate.

Clandestino

Questions:
1. Who wrote that and when?
2. What was the warning about technology and nature, made by the famous writer, who met his doom while flying an F-5?
3. What is written in Ecc. 1:9?

RetiredF4

1. Stick and rudder, 1944, Wolfgang Langewiesche also he didn´t die in an F5, but at the age of 95 in 2002.

I try a potshot at questions two and three:

2. The theory of flight often fails to show the pilot the most important fact in the art of piloting, -the angle of attack- and how it changes in flight.

3. The working speeds of an airplane


Very neat telling from Page 75, i like that one:


No big changes happened since 1944, me think!

Clandestino

That was pretty quick, RetiredF4. 1 out of 3 is not bad.

You're correct that Wolfgang Langewiesche did not die in aeroplane crash. When we mentioned him, his son, William, is pretty good aviation writer too. However, the fellow I was referring to in the second question definitively went out alone in F-5, never to return. His worldwide fame is mostly based on the single shortish book, that only slightly relates to flying. His warning was very general, simple, true and mightily applicable to both AF447 and FBW Airbi.

Answer no2 is way off mark. No3 is too but with ...you got pretty close and I'd certainly agree with that.

OK465

de Saint-Exupéry and there ain't nothin' new under the sun.

Better example: forgive them for they know not what they do...

Back to the Cricetinae Circumvolution

VGCM66

The noose is tightening:

Air France 447: the facts and what's behind them - Learmount

From:

AF447 investigation

Good read even if it is still unofficially.

On top:

AF447 victim families hear damning evidence against airline | Plane Talking

CONF iture

That’s erroneous and you have been answered on that point already here and here.

DozyWannabe

So where would they have fixed them CONF - on the apron? Even if the whole fleet didn't have to be grounded, you'd still have large chunks of it on the ground having repairs and tests done.

Bill G Kerr

3 ...nothing new under the sun.

jcjeant

Hi,

Ask Air Caraibe .. they know how it was before Air France ....

DozyWannabe

Sorry jc, I'm too dumb to "get" vague generalisations - knew how *what* was before Air France?