Talk:Air France Flight 447/Archive 2

Portuguese statement translation
Further to the above "Text messages over the Atlantic?" section, it would be greatly appreciated if someone who understands Portuguese would be so kind as to provide a translation of this statement. Google Translate, whilst very useful, has its limitations. A translation would be extremely valuable to the English Wikinews community where an understanding of this statement is of crucial importance in discussion about a news article there.

Whilst acknowledging this isn't directly related to this article, I trust others can appreciate that it is difficult to know where to start in trying to find some assistance with this. Additionally, I guess it could be helpful in the ongoing discussions about the text messages here. Thanks. Adambro (talk) 22:07, 2 June 2009 (UTC)
 * I've posted the translation at Wikinews. The statement is from the Brazilian Airlines Association (SNEA), denying that they were the source of the "text messages" rumour and saying that such messages were "practically impossible unless there were some sort of satellite link". Physchim62 (talk) 22:36, 2 June 2009 (UTC)
 * Thank you for your assistance. Adambro (talk) 22:47, 2 June 2009 (UTC)

Infobox Survivors fatalities
It is obvious there are no survivors and there were 228 fatalities. Air France is preparing the religious services for all of them. Infobox should be updated. To keep having the survivors data as unknown is to deny the fact that they all perished. --Camilo Sanchez (talk) 22:25, 2 June 2009 (UTC)
 * And yet, we must remember that Wikipedia is about verifiability, not truth. What we "know" is not the same as what we can verify. Frank  |  talk  22:38, 2 June 2009 (UTC)
 * Are you deliberately trolling? Physchim62 (talk) 22:50, 2 June 2009 (UTC)
 * Most certainly not. Frank  |  talk  23:42, 2 June 2009 (UTC)
 * What is 'verifiability' in this particular case? Do you want to see 228 dead bodies, in a situation where it is more realistic that only a few (lets hope more) will be found. Mawijk (talk) 22:56, 2 June 2009 (UTC)
 * We can certainly verify that current publications list all passengers as dead. Since that's also the truth, WP:V says we can just list 228 as dead.--Prosfilaes (talk) 23:27, 2 June 2009 (UTC)
 * I disagree. Although it is obvious that there were no survivors, it has not been confirmed yet.  What various news articles report does not equal confirmation - they could very well be wrong.  Until we can an official confirmation from Air France, the Brazilian government or whoever else, there should be no numbers... --BignBad (talk) 23:30, 2 June 2009 (UTC)
 * So a memorial service (tomorrow, 1400 UTC) in Notre Dame de Paris to be attended by the President of France, or a declaration that "we can do nothing but cry in sorrow and support the families" from the President of Brazil  is not good enough for you? Polite words fail me in the face of such ************* Physchim62 (talk) 00:12, 3 June 2009 (UTC)
 * I just wanted to second that to lend my support to the editors working on making this a good article. Argel1200 (talk) 00:30, 3 June 2009 (UTC)
 * BignBad is right. Kslotte (talk) 03:29, 3 June 2009 (UTC)
 * Somebody added "unverified but presumed", that will probably be OK for now. Other who agree? --Kslotte (talk) 03:59, 3 June 2009 (UTC)
 * I added it due to the concerns from the views expressed above. Seems to be the most logical edit until we get some definite official statement (in my opinion anyway)--661 kts  04:04, 3 June 2009 (UTC)
 * Various news articles is verifiability; not that we can verify the truth of what we're saying, but that we can verify our sources. What you're looking for as confirmation is truth, but as you said Wikipedia is about verifiability, not truth.--Prosfilaes (talk) 13:57, 3 June 2009 (UTC)

It has come to my attention that some editors are living in a parallel Wikiverse, and wish to pretend that all the information available to mere humans does not indicate that there are no survivors. This is completely contrary to all that Wikipedia stands for. These editors are trying to create a separate WikiTruth, different from what any normal human being would understand, and different from the standards applied across the rest of the encyclopedia. In the interests of the project, I will not hesitate to request any further measures against those editors, should their delusion persist. Physchim62 (talk) 14:58, 3 June 2009 (UTC)

Turbulence and Aircraft
At a certain portion of the text we can read the following statement: "The aircraft reportedly encountered strong thunderstorms over the Atlantic Ocean, but all modern airliners are designed to withstand the effects of severe thunderstorms, including heavy turbulence and lightning strikes."

The reference provided for such text (www.nytimes.com) links to a completely different statement, as we can read from the site: "Experts were at a loss to explain fatal damage from lightning or a tropical storm, both of which jetliners occasionally encouter, despite efforts to avoid them. Pilots are trained to go over or around thunderstorms rather than through them — as much out of concern for passengers’ nerves as for the planes’ safety."

The text was placed so one can understand that all modern aircraft can go through "strong thunderstorms" (sic) and survive, what I declare, explicitly, it's unreal, at most. Weather radar onboard such "modern" aircraft are designed to provide the only means to enable a safe deviation from Cumulo-Nimbus (CB), the thunderstorm clouds, and when I say deviation, it means a really distant detour around the cloud formation (ice and heavy turbulence can be encountered far from the CB core). If a pilot loses such equipment in similar conditions verified at that part of the ocean, at the time of the accident, a catastrophical structural failure can really happen. The reason for this kind of accident being so rare is for the highly efficent and dependable weather detection equipment on board, not the strenght of the airframe, itself. When a pilot loses a WX Radar in that particular situation, odds are that he simply penetrates a thunderstorm inadvertently. Or worse: CB penetration due to ignorance, to distraction produced by some distress (electrical failure, perhaps) or even virility demonstration (believe me, it exists!), etc. The general rule here is: If you're going to fight with nature, you'll lose...

I think, IMHO, that the aforementioned citation (and the common thought of aircraft weather survivability) must be discussed. RobertoRMola (talk) 00:09, 3 June 2009 (UTC)
 * Well, obviously something went wrong, and probably lots of things went wrong. Bad luck must have played a large part in the disaster. However, this is neither a forum nor a soapbox to discuss the possible causes. The official version for the moment is that "no possible causes are ruled out". Physchim62 (talk) 00:36, 3 June 2009 (UTC)

I'm not discussing the causes (I left this subject to the investigation team), but the reference for that statement. It is wrong. Agree?RobertoRMola (talk) 00:47, 3 June 2009 (UTC)

I agree, the general rule in all aircraft is to stay out of the way of thunderstorms. Also see Southern Airways Flight 242.--661 kts  00:54, 3 June 2009 (UTC)

I agree a better source is now needed for that claim (the NY Times article changed throughout the day) and it may need to be toned down. Here is a possibility from the Salon column by an airline pilot. It has a much more nuanced view but on balance, still seems to come down on the side of "the airplane should have been able to take it". http://www.salon.com/tech/col/smith/2009/06/02/askthepilot322/index.html --75.223.157.75 (talk) 04:58, 3 June 2009 (UTC)


 * The meteorology of the crash is explained here: A detailed meteorological analysis by Tim Vasquez. The explanation is too complicated for the average reader and so the reference lasted barely one minute. Gabriel Kielland (talk) 23:03, 3 June 2009 (UTC)


 * See also http://www.eumetsat.int/groups/public/documents/image/img_af447_crash_weather.png, which shows a Met-9 IR image with "major thunderstorm clusters on extrapolated flight path of AF447." Can we use this image? —Preceding unsigned comment added by Hallucegenia (talk • contribs) 08:51, 4 June 2009 (UTC)

The image that Hallucegenia refers shows a really bad formation. Never experienced something like that before (I think I was lucky!). I believe it would be an interesting add to the article.

I've read the article by Patrick Smith at Salon.com about turbulence. I agreee partially with the phrase "Neither lightning nor turbulence is normally (my italics) harmful to commercial aircraft." A perfect plane entering into a storm is not a "normal" situation. However, as I stated earlier, am not interested in start a "private" accident investigation (I have full confidence on the investigation team already working at the site). And, beside that, the text about turbulence was removed, anyway.

The Brazilian media is broadcasting today that Air France is stating that "wrong" or inappropriate airspeed was held by the plane inside the storm. Don't know what they're insinuating here (high penetration speed can be one of that non-normal situations during a turbulence) and seems like a guessing, to me, as we don't got FDR or CVR data (yet). I wonder where they got such information... Watching for more evidences.RobertoRMola (talk) 14:44, 4 June 2009 (UTC)

21st deadliest
The lead says "would be tied as the 21st deadliest accident in aviation history," but I doubt the accuracy of List of accidents and disasters by death toll. It could easily be wrong or be missing some. That should be changed to "One of the deadliest accidents in aviation history." Reywas92 Talk 00:30, 3 June 2009 (UTC)
 * I don't think there's any need for the qualification at all. Twenty-first in anything doesn't mean that much to most people, while 228 deaths certainly does. Physchim62 (talk) 00:41, 3 June 2009 (UTC)
 * Taking in account that it's tied with Korean Air Flight 801 - 21th is OK to be removed. --TAG (talk) 00:52, 3 June 2009 (UTC)
 * I have removed the claim saying it is the 21st deadliest accident in aviation history since there is currently a debate going on whether that is actually correct or not. --Jesant13 (talk) 15:52, 3 June 2009 (UTC)
 * According to This list, it's the 18th worst, not 21st. Mjroots (talk) 16:12, 3 June 2009 (UTC)

Bomb threat to an Air France flight on May 27, 2009
http://momento24.com/en/2009/05/27/bomb-threat-on-air-france-flight/ —Preceding unsigned comment added by 24.215.231.170 (talk) 00:51, 3 June 2009 (UTC)
 * Unless there is a source that links this together, this cannot be included in the 447 article. WhisperToMe (talk) 01:05, 3 June 2009 (UTC)

Qantas Flight 72
This is not designed to be speculation, a report on PBS tonight mentioned the following: More details ACARS indicating multiple systems failures the first of which was a malfunction of the ADIRU unit. As I mentioned in a previous thread, now deleted, the causes of many in flight hull failures are often do to known history of an aircraft or aircraft model. This particular malfunction is particularly notorious in A330 and A340 aircraft. According the PBS report (News Hour 06-02-2009) the autopilot spontaneously disengaged, followed by reports of malfuction in the ADIRU, followed by electrical failures in a variety of aircraft control and stability systems. This is a quote from the wiki article on Qantas Flight 72 "The accident began at 1240:28 WST. The plane was travelling at 37,000 feet (11,000 m) when pilots received an electronic message of an irregularity with the autopilot and inertial reference system. The autopilot disengaged automatically and the aircraft climbed 200 feet (60 m) under manual control. The autopilot was re-engaged with the aircraft returning to the selected altitude before the autopilot was disengaged for the remainder of the flight. At 1242:27, the aircraft made a sudden uncommanded pitch down manoeuvre, recording -0.8g, reaching 8.4 degrees pitch down and rapidly descending 650 feet (200 m) in about 20 seconds before the pilots were able to return the aircraft to the cruising altitude. At 1245:08, the aircraft then made a second uncommanded manoeuvre of similar nature, this time reaching +0.2g, 3.5 degrees pitch down and descending 400 feet (120 m) in about 16 seconds before being returned to level flight.[13][3][4][14] Unrestrained passengers and crew as well as some restrained passengers on board were flung around the cabin or crashed on rooftop luggage compartments. The pilots stabilised the plane and declared a state of alert (a pan-pan is broadcast), which was later updated to a mayday when the extent of injuries was relayed to the flight crew[8][15]. Forty minutes later, the plane made an emergency landing at Learmonth airport." Accordingly the statement According to Air France CEO Pierre-Henri Gourgeon these faults created a "totally unprecedented situation in the plane".[16] At that time, the probable location of the aircraft was about 100 km (54 nmi) from the waypoint TASIL, assuming that the flight had been proceeding as planned. is not correct, there are precedences involving 3 A330-A340 type aircraft.PB666 yap 01:20, 3 June 2009 (UTC)
 * TIME magazine used that comparison: http://www.time.com/time/world/article/0,8599,1902421,00.html WhisperToMe (talk) 02:22, 3 June 2009 (UTC)
 * PBS later reported on the ADIRU unit, which based on the Qantas page has malfunctioned on several occasions between Perth and NE destinations, some pilots believed the closeness of the affected flights was caused by a Naval radiotelecommunications center and flights were instructed to steer clear of this station. There are no radiotelecommunication centers however in the MidAtlantic.
 * This is a brief synopsis of the events:
 * Airbus A330-303  VH-QPA, [Qantas Flight 68] 12 September 2006. On 12 September 2006, Hong Kong and Perth, Australia. 0452 local time, 41,000 ft, there was a failure of ADIRU 1. The ADIRU was the same unit (serial number 4167) as on the 7 October 2008 flight.  Technical logs noted a NAV ADR 1 FAULT and that they had received numerous ECAM messages.
 * Airbus A330-200 VH-EBC (Jetstar Airways 07) 7 February 2008, Sydney to Saigon. 0604 UTC, ADIRU 1 failed. The ADIRU same model (serial number 5155). The crew logged a NAV IR 1 FAULT message.
 * Airbus A330-303 VH-QPA (cn 553) This is the airframe that was involved in a midair incident on the 7th October 2008 whilst flying from Singapore to Perth as QF72. It suddenly lost height and resulted in 20 serious injuries. The aircraft diverted to Learmonth Airbase, North of Perth. . The ADIRU unit had the same serial number as previous
 * Airbus A330-303 VH-QPG (Qantas Flight 71) 27 December 2008, Perth to Singapore, 1729 local time, 36,000 ft, the autopilot (autopilot 1) disconnected and the crew received an ECAM message (NAV IR 1 FAULT). ADIRU 1 was the same model but a different unit (serial number 4122) relative to flight 68 and 72 incidences.[see above reference]


 * Here is a synopsis: Airbus searched a database for reprosts of similar problems. No matching reports were found for any other airbus aircraft. There are about 900 A330/A340 aircraft in operation, with 397 having the same model of ADIRU as fitted to VH-QPA, VH-EBC and VH-QPG. The problem appears to be with fly by wire aircraft. The following other incident was found.
 * Boeing 777-2H6ER 9M-MRG (Malaysian Airline) 1 August 2005, Perth to Kuala Lumpur, 1703 Western Standard Time. EICAS reported a LOW AIRSPEED advisory during climbout through 38000 ft. The aircraft’s slip/skid indicator also malfunctioned. Airspeed displayed the aircraft as overspeed and stall speed simultaneously. (However common aircraft training indicates stall is not the result of low speed directly but incorrect angle of attack relative to climb vector of aircraft motion). The aircraft climbed 3000 ft decreasing speed from 270 kts to 158 kts (indicated air speed) causing a stall warning. (158 knts is a low speed for a 777 that is full of fuel and flaps retracted, dangerous in level flight or climb). The aircraft’s flight recording systems were removed for examination. The recorders indicated unusual tridimensional acceleration. These values were supplied by ADIRU to the flight computer, etc.209.30.247.12 (talk) 04:39, 3 June 2009 (UTC)

(reply relocated)
 * "It is not yet known whether Air France 447, an A330, carried the troublesome variety of ADIRU." per the article, so it probably does not belong in the Wikipedia article yet, being speculation about what might have happened. Edison (talk) 03:54, 3 June 2009 (UTC)
 * This is in response to the claim that the A330 ACARS report was "totally unprecedented". As a matter of fact part of the report has 4 matching precedences with the same version of Aircraft and with a different ADIRU unit on a different aircraft. IOW that statement of a perfect safety record of A330 equipment is not true, on the Qantas Flight 72 the Aircraft pitched down so rapidly it pitched peoples heads through the interior ceiling of the aircraft forcing the aircraft to redirect to a closer airport to evacuate the injured.209.30.247.12 (talk) 04:55, 3 June 2009 (UTC)

This is pure speculation. None of the previous incidents caused changes in flight that could lead to airframe breakup, especially given the experience of the Air France pilot and the last known height of the plane which gave a good safety margin for recovery, even from a stall or a steep dive. Unless an investigative body declares this as a possible cause we cannot include it in the article. --Ferengi (talk) 06:24, 3 June 2009 (UTC)


 * On the topic of the Naval Communication Station, keep in mind that while its influence is being seriously investigated by the ATSB, reports so far have stated that those investigations have not turned up anything substantive to date. On the topic of inclusion, while it is simply unsubstantive speculation it has no place in the article.  If the reports regarding ACARS messages about ADIRU faults actually do satisfy WP:V then in my view it justifies inclusion as long as it keeps to the fact and doesn't cross over into speculation. -- Rob.au (talk) 11:39, 3 June 2009 (UTC)


 * Mention of QF72 and its ADRU smacks of original research, even if casual observers think this might be an obvious link. Let's quit the speculation please and wait for the investigation - there's no hurray to jump to any conclusions.   Socrates2008 ( Talk )   12:56, 3 June 2009 (UTC)


 * I repeat myself once again. A speculative statement "According to Air France CEO Pierre-Henri Gourgeon these faults created a 'totally unprecedented situation in the plane'.[16] At that time, the probable location of the aircraft was about 100 km (54 nmi) from the waypoint TASIL, assuming that the flight had been proceeding as planned." claims that this situation is "totally unprecedent". It is neither unprecedented in the Airline industry China Airlines Flight 611 nor is the first line of events, the cessation of the Autopilot initiated by an ADIRU fault unprecedented. Either remove the speculative comments or I will add comments concerning Qantas 72,71,68, and etc. Time magazine and PBS NewsHour are as good of sources as Air France's CEO (who has a vested interest to protect the image of his airline WP:COI) Wikipedia ask for balance does it not? The mechanical mid-air failure of most aircraft can be attributed to known defects in the aircraft or line of aircraft. For CAL 611, and JAL 123 it was a known requirement for repair of the tailcone bulkhead that the doubler plate extend from the upper to lower plate so that two rows of rivets on each plate held the double plate to the bulkhead. This is science, not Devil's Tiangle or Displaced Dipoles. It is now known that certain ADIRU units can malfunction in midair, causing near catastrophic damage, it is known that there is a pattern, and statistics tells us that with a small sample size the outer bounds of the distribution, in this case-structural damage, cannot often be predicted based on the known sample. It is known that Airbus is aware of the problem, however no change in the model and replacements made so that this is an outstanding defect in the Aircraft. There is another problem, however, the A330 carries 3 inertial reference units, and yet the ADIRU 1 is recognized by the flight computer and 2 others are ignored. It would seem that there is a flaw in the Flight computer software based on the 4 known events. There is now grounds to believe the aircraft had electrical problems in India, and electrical faults (brown-downs and overvoltage) can trigger malfunctions of CPU governed machinery (I.e. flight computers in fly-by-wire aircraft).PB666 yap 14:07, 3 June 2009 (UTC)


 * PB666 - presently there is no valid connection between AF447 and QF72. It has indeed been established that one of the ACARS messages mentioned an ADIRU fault so this is fine to say in the article, but we must not move beyond this into speculation, as this does not meet with Wikipedia's policies.  Here in your own comment you have speculated that in this instance that cessation of the autopilot was initiated by an ADIRU fault.  The established facts to date do not allow us to reach this conclusion.  The ADIRU was one of a significant number of systems which appear to have failed according to the ACARS messages.  The message about the ADIRU fault came after the message about the autopilot disconnection, but at this stage drawing any conclusions from this is pure speculation... it doesn't even confirm anything about the real sequence of events (as the ACARS reporting system may have its own logic for sequencing its brief reports).  So it could be that the ADIRU failed first.  But it may not be.  We don't know.  Perhaps later on in the investigation process valid connections may be made with respect to the ADIRU and at such time we can include them in Wikipedia.  Obviously your other point of contention appears to be along the lines of "If we can't add this speculation, why has the article included the quote from the Air France CEO?", per WP:NPOV.  My feeling is that the comments by the Air France CEO are notable and the do not preclude the theory that the ADIRU could have been the root cause.  He appears to be commenting that the combination of factors is unprecedented, which at this stage seems to be a fair comment.  If we compare to QF72 - even if the root cause turns out to be the same in due course - clearly the situation for AF447 was very very different.  The pilots of QF72 were able to remain in radio contact with ATC and their company and in due course were able to land their aircraft safely.  The Qantas crew also did not have a cabin depressurisation event.  Clearly - at the very least - something more happened to AF447, assuming it was not in fact something completely different (which again, at this stage, we do not know).  The term "unprecedented situation in the plane" seems to hold just fine, based on the verifiable information we do have so far. -- Rob.au (talk) 17:24, 3 June 2009 (UTC)
 * Very well summarised, Rob.  Socrates2008 ( Talk )   10:00, 4 June 2009 (UTC)
 * This is to say, politely as one can say it, that any mid-air airframe failure is "totally unprecedented" for that airframe, by definition an airframe can never completely fail twice (other than thorugh metal recycling) and is therefore a pedantic statement. And yet with TSB investigations it is generally found that most mechanical airframe failures have lessor or other catastrophic precedences (either human error or mechanical problems). If you put a million sensors on an aircraft, any and every pressure hull failure will be unprecedented, but as long as there are engineers to investigate these things they will find precedences. For this reason the industry has managed over decades to reduce the fatality per mile travel by almost a magnitude for large commercial aircraft, as there are generally few exceptions to the rule (remember the [De_Havilland_Comet|Comet]]!). Therefore precedences do speak against the claim.
 * The principle danger posed by ADIRU failure is a pitch-up stall in which the FCS disallows the pilot to manually correct the stall by lowering the angle of attack and increasing, if possible, turbine thrust before the plane begins a dead mans spiral. Whereas the other risk is a large negative angel of attack with turbine throttle up resulting in engine cavitation and failure, at which point one has a forced water landing on questionable recontrol of aircraft. The other possibility is that the pilots could not wrench control from ADIRU (i.e. two units simultaneoulsy failed because of a systemic electrical problem, software or hardware problem with the FC computer or there is human error in the response to the ADIRU) and basically the FCS prevented pilots regaining control, this was one of the predicted risk with fly-by-wire aircraft. In line with this risk is that the pilots are not buckled in their seats in which there could be a loss of personnel capable of overriding the FCS before a "no-going back point". In the simulator one can basically create a no velocity stall, which most aircraft can correct from, but in most instance the aircraft, AOA -90 quickly accelerates to the overspeed limit, before the controller can bring the aircraft into the safe equilibrium, and older aircraft may have stress failure of the hull. Absence of evidence for system generated pressure hull failure follows no Black Swan logic and is to be discouraged.
 * I have looked at the BBC-News projection of 447s final path between the tropical convergent associated precipitation, other than spurious turbulence, storm cell activity does not appear to be a major factor or unavoidable on the flight path, ergo, despite all the hype this looks more and more like a electro-mechanical irregularity. Which is to my point, lets get the hypey CYA statements off the main page. Turbulence may be a factor (a cause of ADIRU failure), but weather appears to be no more of a factor than ADIRU. If one wishes a wiki-balanced approach, the fat needs to be trimmed.209.30.247.12 (talk) 18:51, 3 June 2009 (UTC)

Guys, it is clear to me that the ADIRU failure caused the plane to go into a steep dive from which they could not recover and it eventually broke apart. However, we have to wait for the investigation to confirm this before putting it in the article. Until the investigation comes up with some conclusions (at least preliminary) it will remain speculative. 77.28.2.118 (talk) 09:49, 4 June 2009 (UTC)

If "this particular malfunction is particularly notorious in A330 and A340 aircraft," then why do they use them? Grundle2600 (talk) 11:55, 4 June 2009 (UTC)


 * They are used in many different aircraft, not only Airbus. You're drawing premature conclusions about the cause.   Socrates2008 ( Talk )   12:04, 4 June 2009 (UTC)

Correct statement ?
I read the statement "France dispatched the research ship Pourquoi Pas?, equipped with two mini-submarines capable of performing search operations at the 4.7 km (2.9 mi) depth of the seabed in the area where the aircraft went down. The ship confirmed the wreckage of Flight 447 and found a 3-mile long trail of debris across the ocean floor, confirming that all lives were lost.[4]" But the Brazilian press is reporting that the mentioned ship is at Azores Island, and by Jun, 02 it will take 30 hours to arrive at scene. The [4] goes to the BBC site reporting nothing about it. I could not verify if the information is correct or not. What is confirmed was that debris were found floating. —Preceding unsigned comment added by Asilva11 (talk • contribs) 01:27, 3 June 2009 (UTC)

Miles O'Brien article link
I had added the following to the external links, but 661kts (talk) keeps on deleting it without giving any indication why:
 * A Dark, Stormy Night over the Atlantic, article by aviation journalist Miles O'Brien

In my opinion it's a well written and profound article on the subject which definitely adds good information. Should it be kept or go? What is your opinion? --Berny68 (talk) 03:39, 3 June 2009 (UTC)

I actually think Miles O'Brien is probably one of the best journalists out there when it comes to aviation. But what he provides is basically speculation so why should it be included in external links? How would you argue if somebody else came along and added another link to a journalist that was speculating, would you argue that Miles O'Brien is a better speculator?--661 kts  03:53, 3 June 2009 (UTC)

Yes, I would argue that Miles O'Brien is about one of most technically profound journalists on the subject matter. Since currently there's nothing else we can do regarding the causes of the incident than speculate, at least we should provide a link with the most sound and substantive speculation there is available. If there's other speculation on a similarly good level, it should be added too. However as long as causes are unknown, and they might be unknown forever, we should provide at least the best speculations there are here. --Berny68 (talk) 04:17, 3 June 2009 (UTC)

If you want to speculate you can go to countless forums like pprune.com, forums.jetcarrers.com, airliners.net, etc... This is supposed to be an encyclopedia so we don't try to include speculation especially when the incident happened recently.--661 kts  04:25, 3 June 2009 (UTC)
 * I have no idea who Miles O'Brien is; I know nothing about aviation. However, I think 661kts's arguments for deletion are weak - though stronger arguments may exist. If O'Brien is respected and his speculations notable, I think berny68's arguments for inclusion are perfectly valid - in fact, including a sourced summary of the speculations (clearly indicated to be speculative) would make perfect sense, and would be perfectly encyclopaedic.--Noe (talk) 15:07, 3 June 2009 (UTC)

Infobox image

 * Air France Airbus A330-200 (CC3.0 image)
 * Air France Airbus A330-200 (CC2.0 image)
 * F-GZCP, the actual plane involved (fair use image)

Well, the image has been changing some more...

File:F-GZCP.jpg has also been nominated for deletion.

70.29.208.129 (talk) 05:12, 3 June 2009 (UTC)


 * The FFD discussion for F-GZCP.jpg is at Files_for_deletion/2009_June_3 &mdash; 70.29.208.129 (talk) 08:27, 3 June 2009 (UTC)

See Talk:Air France Flight 447/Image discussions for infobox image discussions. /wangi (talk) 08:30, 3 June 2009 (UTC)


 * Those are old archived discussions. The image in the infobox is shifting now. A new discussion is possibly necessary. 70.29.208.129 (talk) 11:34, 3 June 2009 (UTC)
 * Those are not archived discussions, but rather centralised to keep it in one place. /wangi (talk) 11:50, 3 June 2009 (UTC)
 * Those are dated, and there is no banner saying a centralized discussion area has been created. So how is it not an archive? 70.29.208.129 (talk) 12:57, 3 June 2009 (UTC)

Sorry, but I've got neither the time nor inclination to get into a meta discussion. If you want to discuss the infobox image then go ahead, but it's really all been said already and until such point we have a free image of the actual plane the current image really should be left as-is. Certainly it would be wrong to try and claim fair-use on a non-free image. Thanks/wangi (talk) 13:19, 3 June 2009 (UTC)
 * If no one can find the "centralized discussion" it's not much use. People editing the article and looking at this talk page certainly won't see it. It's a walled garden. 70.29.208.129 (talk) 13:50, 3 June 2009 (UTC)
 * Wangi - please stop misrepresenting Wikipedia's policy on fair use images. Wikipedia does not prohibit fair use of images - we have a detailed policy on doing so.  Another image of the crash aircraft which was fully free would justify replacement of the fair use one we have now (which is CC-BY-NC-ND anyways - and therefore nearly completely Wikipedia "free" anyways, only the NC being an issue).  But other images of other A330 aircraft miss the historical significance of the particular F-GZCP.jpg image being of the actual crashed aircraft.  Photos of historical events or things which are now not freely recreateable as the thing happened, or went away, are one of the clearly allowed fair use categories.  This plane is now gone, unfortunately.  We can't go take new photos.  If you want to convince the Flickr photographer to CC-BY or CC-BY-ND or GFDL release it, fine.  If you can find another, fine.  This is historically relevant here and now.  Georgewilliamherbert (talk) 19:18, 3 June 2009 (UTC)
 * We allow fair use images if there are no reasonable free versions/replacements. George William Herbert, I argued your point regarding the Comair image, but the community consensus was that a photo of a similar aircraft was a reasonable representation of the aircraft while intact. WhisperToMe (talk) 19:20, 3 June 2009 (UTC)

I resent the comment that i am misrepresenting Wikipedia's policy on fair use images. Clearly our views differ, but i have backed mine up considerably at Files_for_deletion/2009_June_3 and Talk:Air France Flight 447/Image discussions. A particular note on the former - this was a discussion on the issue, a discussion you should have participated in before recently changing the infobox image back to a non-free image. You cannot have a "nearly free" image - the photographer does not want us to use the photograph commercially, we should respect that. In anycase there are more appropriate non-free images of the actual plane - ones taken on its final take off. /wangi (talk) 19:42, 3 June 2009 (UTC)
 * Comment: As an observer I can't help but to think that this is incredibly lame and just works to ensure that our readers receive sub-standard coverage. An image of the actual plane is incredibly important in coverage of its' demise, even if just for sentimental reasons (hence real images being used in grown-up coverage of the incident). I don't plan to discuss it here as I've already wasted plenty of time trimming, reducing and uploading the original image, but a little more objectivity in place of the rampant wikilawyering wouldn't go astray. -- samj in out 09:08, 4 June 2009 (UTC)

Southern Airways Flight 242
Isn't it a little early to have a link to Southern Airways Flight 242 in the "See Also" section? While weather may have played a role, this isn't confirmed yet, so the link seems premature.75.164.208.19 (talk) 08:36, 3 June 2009 (UTC)

Bomb threat on Air France flight days before crash of 447
Just a note. Maybe worth a mention: Bomb threat delayed an Air France flight to Paris days before Flight 447 crashed. DragonFire1024 (talk) 11:12, 3 June 2009 (UTC)
 * Unless you can show a direct link between the two, it's off topic for this article (and already mentioned above on this talk page). &mdash;  The Hand That Feeds You :Bite 13:33, 3 June 2009 (UTC)

Plane that crashed may have been the same one that was grounded for electrical trouble
This one seems worthy of investigation.I have found only these references, conflicting so far, which point to the possibility that the crashed plane had been experiencing technical difficultly a week earlier in India. If true, it is curious that it hasn't been picked up by other media. if false, it would be good to verify it. Accordingly, I have not tried to modify the article itself since i am not sure if it is true or not.Here is the info, contained in 3 articles.

ARTICLE ONE: IMPLIES NOT THE SAME PLANES, THOUGH IT WAS IN INDIA WITH NO PROBELMS This article mentions that two planes (at least one the same Air France plan that crashed later) stopped over in Bangalore India in late May):

Crashed plane was in Bangalore twice last week http://www.indianexpress.com/news/Crashed-plane-was-in-Bangalore-twice-last-week/470314

"The Air France aircraft that was lost above the Atlantic on a flight from Rio de Janeiro to Paris on June 1 had been in Bangalore twice last week.

The four-year-old A330-203, bearing the registration number F-GZCP, flew into Bangalore on May 26 and May 28, leaving for Charles de Gaulle, Paris, the same day on both occasions. The aircraft did not report any technical problems, air traffic controllers who cleared it for take-off said......

....[A N Vishwanathan, General Manager, Aerodromes, BIA] clarified that this aircraft was different from another that landed in Bangalore at 11.46 pm on May 30, and left for Paris just over two hours later. “The registration number of the aircraft that left Bengaluru airport on May 31 was F-GZCI, as per our records,” Vishwanathan said. "

ARTICLE TWO: SORT OF CONFIRMS THE CONCLUSION OF THE FIRST This article from the website Bangalore Aviation also suggests the crashed plane was one of the two that the above article mentions stopped in Bangalore, and also the one that did not have any technical problems:

Air France Airbus A330 flight AF447 Rio de Janeiro to Paris missing; 228 feared lost - Update 3 http://www.bangaloreaviation.com/2009/06/air-france-airbis-a330-flight-af447-rio.html

"An Air France Airbus A330-200, registration F-GZCP performing flight AF447 which departed May 31, 2009, from Rio de Janeiro Galeão - Antônio Carlos Jobim International Airport, Brazil ( IATA: GIG ICAO: SBGL) to Paris Charles de Gaulle, France ( IATA: CDG ICAO: LFPG) with 216 passengers and 12 crew, is missing and overdue at Paris for more than two hours. The airplane had departed Brazil's radar coverage normally."

ARTICLE THREE: SUGGESTS THE PLANE THAT CRASHED WAS THE ONE THAT HAD DIFFICULTY IN INDIA Yet this one from the same website as the second suggests that the plane that crashed IS the same one that had problems at Bangalore, highlighting the contradictory information in the 3 articles.

Air France Airbus A330 grounded at Bengaluru International Airport - Update 3 http://www.bangaloreaviation.com/2009/03/air-france-airbus-a330-grounded-at.html "Something is in the air at the Bengaluru International Airport. For the second time in less than a week, an Airbus A330 is getting grounded at the airport with a technical fault.

Following the five day grounding of a Dragonair A330, it is the turn of Air France. F-GZCB which was to perform flight AF191 early this morning, has been grounded with a technical flaw in the avionics bay....

Update 2 - 06:00GMT 11:30 local March 22

Poor F-GZCB is still sitting on the ground. Very forlorn. The major electrical problem has been addressed. The parking bill from the Airport is going to be big.

Update 3 - 18:00GMT 23:30 local March 22

F-GZCB finally flew out as a ferry flight at 19:00 local time (14:00 GMT). Even the departure was not without drama as the aircraft was originally scheduled to leave at 15:45 local to avoid the night curfew....

Since F-GZCB was in "AOG" status (Aircraft On Ground) for such a long period, she will undergo a thorough check today before being pressed back in to service."

Paul in Tokyo


 * We're not here to speculate - let's wait for the investigation to draw any conclusions.  Socrates2008 ( Talk )   12:58, 3 June 2009 (UTC)

Hi TALK

I agree. But I was simply wondering aloud why this has only been mentioned in a handful of places. One would think other media would have picked it up or it would have been either confirmed or denied by the authorities. So I guess, more than waiting, I would hope we could keep an eye out for it. I suppose it is still in the category of rumors since no authoritative sourcvehas directly said the same, right?

paul —Preceding unsigned comment added by Tokyoprogressive (talk • contribs) 23:19, 3 June 2009 (UTC)

These appear to have been different planes... F-GZCP vs F-GZCB. -- samj in out 09:13, 4 June 2009 (UTC)

Crash site
There are at least two and possibly three separate debris locations. At about 4:00 UTC (local night time) on June 2 an R-99 radar aircraft received multiple returns indicating metallic and non-metallic objects floating at the sea. Shortly before 9:00 UTC on the same day (at roughly local daybreak) and also almost one hour later a C-130 visually confirmed debris at two locations separated some 60 km (40 mi) from each other: among those, an airplane seat, an orange life jacket, small white debris, a barrel, and oil and kerosene spots. At that point, Col. Jorge Amaral, the Brazilian Air Force spokesperson, did not want to confirm that flight AF 447 has been located, "given the small amount of debris compared to the large size of that aircraft". Later on the same day, at 15:20 UTC, a KC-130 discovered a band of debris 5-km long, consisting of "white pieces and electrical conductors" and more oil patches. Then the Minister of Defense talked to the grieving relatives and confirmed that the A-330 has been destroyed. Sources:

BAF Bulletin Number 5

http://www.fab.mil.br/portal/capa/index.php?mostra=3095

Please see the accompanying map

http://www.fab.mil.br/portal/imagens/avista_voo447.php

BAF Bulletin Number 6

http://www.fab.mil.br/portal/capa/index.php?mostra=3103

Aldo L (talk) 13:36, 3 June 2009 (UTC)

Debris not retrieved yet
That claim by CNN is not substantiated. Neither the Brazilian Air Force nor the French Bureau of Inquiry and Analysis for Civil Aviation Safety confirm the retrieving of any material belonging to the A-330. Aldo L (talk) 15:16, 3 June 2009 (UTC)
 * The Brazilian Minister of Defense says that "he's sure" that the debris found is from AF447. There's no confirmation of an being retrieved yet. Physchim62 (talk) 16:36, 3 June 2009 (UTC)


 * The Brazilian Navy just made it clear that it is coordinating the merchant vessels' search efforts and none have found anything. Source: https://www.mar.mil.br/menu_h/noticias/busca_air_france/air_france-desaparecido.htm Aldo L (talk) 21:18, 3 June 2009 (UTC)


 * It seems that the oil spit is not from airbus 330 but is from a ship. The amount of oil found is to big for a airplane. Voo 447: Destroços e óleo recolhidos do mar não são do avião, in portuguese. He said: "We discard this possibility in function of the amount of oil found. In an airplane, this amount does not pass 50 liters in each engine. The oil spit found is so much bigger than this quantity would make".--Eduardoferreira (talk) 23:45, 4 June 2009 (UTC)

ADIRU in the ACARS report
This is not a request to add information of ADIRU to the main page, but that speculation about precedences be removed from the main page.
 * 1) The claim that the ACARS report is "totally unprecedented" is not factually correct.
 * 2) 4 instances of autopilot disengagements on A330 have been previously reported and at least one instances is certainly due to ADIRU malfunction since it occur 3 times in two incidences on the same airframe, the last time injuring passagers and damaging the aircrafts interior.
 * 3) The reliance on a single ADIRU unit (generally ADIRU 1) when two other units per aircraft are functioning normally indicates that the evaluation by the flight control computer may have hardware or software weaknesses.
 * 4) The probability that this would occur on 4 of 397 units of the same model but 0 of 526 units of different models is significant, p=0.029 but only indicates that problems may occur at a much lower rate, failure of other ADIRU may create other different systemic responses (more or less severe?). This is noted by the ADIRU malfunction on a single 777. There is a report that the AIR FRANCE ADIRU are not made by Northrop Grumman (As Qantas) but by Honeywell.
 * 5) It appears though I cannot verify that within a small time frame several messages were sent, the first of which appears to involve autopilot disengagement and a ADIRU failure. One message involves the change of cabin air velocity, which may indicate decompression but might also indicate a rapid change in the aircrafts angle of attack and airspeed. PB666 yap 20:06, 3 June 2009 (UTC)

This is an update on the ACARS message from
 * 1) ACARS messages of system failures began to arrive at 02:10Z. Autopilot had disengaged and the fly-by-wire-system had changed to alternate law Alternative Law, however does not disengage ADIRU if it senses any of the ADIRU are still active.
 * 2) Between 02:11Z and 02:13Z messages regarding ADIRU and ISIS (Integrated Standby Instrument System) faults arrived. Note:I found this is the most pertinent EASA notification on the type of ADIRU error that appears in the ACARS report. Note applies to  aircraft with Nortrup Grummen  But parts may apply to any ADIRU 1 and 2 (1 and 3 or 2 and 3) failure:  Flight controls are in alternate law. Faulty ADIRU are to be turned off specifically the IR are to be manually turned off. Use AIR DATA switching as appropriate. Use ATT HDG switching as appropriate. Do not use speed brakes • If CG  above 32%, Manually perform a forward fuel transfer from the trim tank. Note: If trim tank pump is not available, do not perform manual forward fuel transfer while speed is at or below 270 kt or while in climb. This indicates that a flight system failure of this magnitude can happen and requires a number of manual corrective actions.
 * 3) At 02:13Z PRIM 1 (The primary flight control computers) and SEC 1 faults (Sec 1 - Secondary flight control computers) Both receive inputs from the 3 ADIRUs.
 * 4) At 02:14Z the advisory regarding cabin vertical speed was sent. Changes in cabin pressure that are comparable to climbs of more than 1000 vertical feet per minute result in automated warnings.


 * There is a meterologist who did a detailed weather study for ~200Z June 1 2009 indicating the following which I will detail; however here is one of the finds It yields a worst case instability of 1048 J/kg of CAPE, which is moderately strong but considered borderline for typical severe weather. Vertical velocity can be obtained by w=2*CAPE^0.5 yielding a maximum possible updraft speed contribution of 45.8 m/s or 102 mph, though in reality this is usually much less (on the order of half or less) due to precipitation loading and other factors. Therefore strong updrafts as 447 left the active area may have driven the airplane upward 100s of feet per minute. The models differ from those presented by BBC-News and were the debris was found, to the east of INTOL and, presumbably the prevailing currents would have carried surface debris to the West North West, indicating that 447 chose a vector east and then northeast into Senegal Airspace.
 * Having said that the time profile developed indicates that there were scattered developing updrafts to the east that could be flown around.PB666 yap 02:07, 4 June 2009 (UTC)


 * PB666, there's too much WP:OR presented in your argument here - but critically, I don't share your interpretation of the "totally unprecedented" quote. You appear to be suggesting that the quote only refers to the ADIRU failure as totally unprecedented.  My interpretation of the quote is that it refers to the combination of factors - including, but not limited to the ADIRU failure - as totally unprecedented.  I am not aware of - and therefore have not seen any evidence of - any other incident that has involved the same combination of factors that we know AF447 suffered.  While it can sometimes be frustrating, at Wikipedia there are policies that need to be followed which include not jumping the gun into WP:OR and speculation.  I understand you have concerns regarding the speculation that the same root cause my exist between the Qantas incident and this one - we all do - but we have to stick to the facts as we know them.  Wikipedia is dynamic - when more facts come out, they can be added.   There's nothing about the quote from the Air France CEO that precludes the possibility that the events are related, so I'm struggling to understand what is concerning you about it. -- Rob.au (talk) 10:56, 4 June 2009 (UTC)


 * It is interesting that when one brings in facts WP:OR is thrown out, but when one brings in things like Bermuda Triangle, these lovely WP are not applied. WP:OR is frequently applied here when people disagree, its a political tool. Let's get this strait once more, I did not make the claim that the inflight emergency was "totally unprecedented", and frankly it is very pedantic to argue in defense of the claim, since by its very nature any mid-air frame-hull loss without a substantive amount of background information is unprecedented, that's why investigations are conducted. The only passenger jet mid-air failure that was totally unprecedented is BOAC_Flight_781 here is a synopsis from wikipedia: "Engineers subjected an identical airframe, G-ALYU ('Yoke Uncle'), to repeated re-pressurisation and over-pressurisation and after 3,057 flight cycles (1,221 actual and 1,836 simulated), Yoke Uncle failed due to metal fatigue near the front port-side escape hatch.[13] Investigators began considering fatigue as the most likely cause of both accidents and initiated further research into measurable strain on the skin. Stress around the window corners was found to be much higher than expected, 'probably over 40,000 psi,' and stresses on the skin were generally more than previously expected or tested. This was due to stress concentration, a consequence of the window's square shape."
 * In the main for BOAC 781, it is clarified that the engineering specs called for glueing of the window in the frame no riveting, and the riveting allowed tears to form, that when stress widened to the point of catastrophic failure. It is precisely these types of weaknesses and errors, most often known in advance, that are the mechanical failures that bring down aircraft. Even this first disintegration was not totally unpredictable. As we see below AF 447 was involved in a previous collision in which its apparent wingtip collided with the tail of a different aircraft. The introduction of tears in rivoted structures as in CAL 611 and JAL 123 in the BOAC Flight 781 are frequently precursors to catastrophic failures. But more importantly than that, before the engineering testing of the comet was done this is what was concluded: There was no obvious reason for the crash, and the fleet was grounded while the Abell Committee met to determine potential causes of the crash. The committee focused on six potential problems: control flutter (which had led to the loss of the de Havilland Swallow), structural failure due to high loads or metal fatigue of the wing structure, failure of the powered flight controls, failure of the window panels leading to explosive decompression, or fire and other engine problems. The committee concluded fire was the most likely cause of the problem, and a number of changes were made to the aircraft to protect the engines and wings from damage which might lead to another fire.
 * Once again, speculation based on a lack of information is generally wrong, what is or is unprecedented can only be determined after rigorous testing of equipment comparable to the failed equipment or simulated failures are conducted, and investigations of Apriori damage and apriori weaknesses in the aircraft equipement. For example CAL 611 and JAL 123 both had takeoff associated tail strike incidences prior to mid-air and ground crash events. Hence, I am striking the comment from the mainpage. Enough of this.PB666 yap 12:50, 4 June 2009 (UTC)

You raised an interesting point about the message indicating a change in cabin vertical speed. Currently the article says this indicates a "possible cabin depressurization". That's probably true but not the whole truth in the sense that it could also indicate a rapid climb or dive, right? --70.198.105.32 (talk) 14:58, 4 June 2009 (UTC)


 * On ABC News Nightline last night, that is what of the aviation experts also said. However both conclusions are speculative, it could be a pressure hull failure, the consequence of a rapid dive, or the result of sensor malfunction as a consequences of an electical fault initiated cascade (Systemic over, undervoltage, or voltage fluctuation).PB666 yap 17:01, 4 June 2009 (UTC)


 * PB666 - I don't see a single element of your argument that justifies removal of the quote by the Air France CEO.


 * This is troubling in itself.PB666 yap 17:01, 4 June 2009 (UTC)


 * You have failed to address the point I have made that the Air France CEO's comment doesn't actually form speculation at all, it simply is an apt summary of the facts to date and does not in any way, shape or form preclude an eventual finding in keeping with the speculation you are making. I honestly do not understand at all what you have against it. -- Rob.au (talk) 15:34, 4 June 2009 (UTC)


 * One does not know whether the statement is speculative or not until all the facts are in, which makes the statement speculative. The claim that the circumstances of failure are unprecedented cannot be made until the cause of the airframe failure is studied. He, nor you, are qualified at this point to draw such conclusions. The word is "unknown", IOW that cause of failure is unknown.

What is speculative
These issues have come up. However, it is possible that any, (more than one and other) are contributing factors, based on passed precedences.
 * 1) Weather caused or was a major contributing factor.
 * 2) Lightning strikes were a contributing factor to electrical system malfunction.
 * 3) ADIRU malfunction initiated a cascade of events that was a major contributing factor.
 * 4) Flight control computer malfuctioned or software has bugs.
 * 5) Previous damage was a contributing factor.
 * 6) Previous electrical failures or recent major overhauls.
 * 7) Attempted explosion or sabatoge.
 * 8) ? others. The more we have looked, the more we have found.

Other forms of speculation
 * 1) That this is the first time ACARS reported the failure of the aircraft in the process of destruction. (We don't know if the airframe was failing or failed after the reports)
 * 2) That the pilots did not have time to communicate or were unable to communicate because required activity and attention to immediate tasks.
 * 3) Flying over magical earth events (Geographic triangles, anomolous magnetic dipoles, Equator)
 * 4) Which direction AF 447 was flying prior to its disintegration or why.
 * 5) Reasons as to why debris is spread widely.

It is clear that there can be multiple factors, as there often are in the failure of an airframe midflight, therefore any conclusion regarding past precedences are unjustified at this point, simply they are unknowns. Unknowns are not precedences lack of precendences.

There is a succession of events that are unprecedented for expected reasons.
 * 1) The inflight reporting of faults prior to a mid-air disintegration. There have been no previous mid-air disentegration of A330 airframe. However, all aircraft are vulnerable to similar stresses, for example overspeed stress. The overspeed limit on the A330 is similar to other aircraft of its size and function. What seems likely at the moment, but speculative is that this airframe was placed, either by turbulent air patters or electronics failure or a combination of both in a circumstance in the A330 design tolerance. Obviously as new technologies are incorporated into new aircraft that allow the ability, in realtime, to observe failures that eventually result in airframe disintegration, however at this point we do not know whether these failures preceded structural failure, were simultaneous with structural failure or occurred after structural failure.
 * 2) While the pilots were not bound by the directive, if two faults of the ADIRU had occurred they might have been busy trying to shut down or others electronic equipment they believed to have failed. It is not uncommon in these types of disasters for pilots to make very brief explicatives while trying to 'right' the aircraft. Airbus clearly states that ISIS should not be shutdown during flight, if ISIS failed and reinitiated then this could have rather strange consequences. Some of these faults these pilots were aware, however there may have been system faults we are unaware, therefore the entire 4+ minutes of aircraft malfunction/disintegration may have been spent trying to keep the aircraft aloft.
 * 3) If we remove the ACARS message, the over-water disappearance of an aircraft has many precendences, in fact the first bonafida cruise level flight disentregrated was the Comet in the early 1950s. For every aircraft type that has wide circulation there is going to be first mid-flight disintegration and for every disintegration there will be an investigation that attempts to formulate conclusions based on passed precedences, and when those fail, new precedences will be created. For example, fly-by-wire aircraft are well past critical mass in the global fleet that we can expect that combinations of stress, pilot errors, and unknown weaknesses will create new precedences for incidences. We don't get mid-air failures of technology unless we put technologies in mid-air.
 * 4) The failure of aircraft around weather phenomena is also precedented, the failure last winter of a commuter jet is an example. Weather is frequently a factor, although less frequently at altitudes over 30,000 feet (Hail risk is low, water ingestion into the turbines is lower, ice accumulation on control surfaces is lower).
 * 5) Widely spread debris has precedence, JAL 123 for instance lost part of it tail section over Oshima Island and later crashed into a mountain well into the Japanese mainland. Specialized loss of aircraft components prior to impact is often an indicator of a stress related equipment malfunction on the aircraft, that does not mean weather was not a factor. In the case of JAL 123, ascending past a certain altitude was a factor, environment can add critical stress.

IOW there are no totally new precedences for aircraft failure to be found in what we now know. As I see you have replaced the quotation, in doing so it means that the quotation is now a target for critique on the Main page, is this what you want?PB666 yap 17:01, 4 June 2009 (UTC)


 * PB666, please read carefully WP:OR. Understand that people have nothing against your comments, but since you insist on making edits based on your own 'conclusions' without providing verifiable sources for any of your analysis and then go on to blame others for censorship when they point out WP:OR, you are going close to the edge of NPOV here. You clearly have a point to make and want to make it forcefully.


 * Further more, I appreciate your analysis, but it belongs to an aviation enthusiast site, which there are many on the internet. Not a Wikipedia article on an ongoing event. Notice that I don't comment here, nor analyze or speculate; I'm certified avionics technician with a degree and several certificates for various levels of maintenance as well as specific systems and devices. I could point out several flaws in your speculation, but we can discuss that on another site.


 * You said We are not here to investigate the fate of this flight - there are experts who do that for a living (have you ever read a full accident investigation report?, it runs in 1000s of pages). Is the point I was trying to make. And secondarily relevance of an ad-hoc statement to the topic Accident is not for me to prove or disprove, it is for its proponents. So far the supporting evidence is:
 * He's a really big guy at Air France.
 * The media quoted him.
 * You've just stated that the relevance or accuracy of his statement is irrelevant.


 * That does not sound like it qualifies for wikipedia, except maybe as a judges comment at a beauty contest or a qoute from Donald Trump about an Apprentice. No evidence was presented that links this quotation to the accident, since he was in Paris, and the pilots were in the mid-Atlantic trying to keep the aircraft aloft. One final point, you can get a freeware version of Firefox, I have the same problem as yourself with spelling. Firefox is wonderful at finding and correcting spelling errors and because firefox pointed these out, I corrected them for you, this may better help you to see the inconsistencies in your argumentPB666 yap 00:58, 5 June 2009 (UTC)


 * We are here to compile an article based on notable information from credible, vefiable sources. Not make up our own stories, however expert and exiting it might be. We might have to wait a year, but eventually an accident analysis by a group in experts will be released, who have looked at all the information gathered eventually and reported exactly the technical details and the most likely scenario of that happened to this flight - what you are trying to do now on incomplete information, expertise, alone, on a forum which is not meant for it.


 * Your criticize people's sources as speculation - yes, there are alot of speculation on the media right now - but that speculation, when made by notable credible sources, is still notable and belongs to this article - your analysis of what is and what is not speculation is not relevant - wikipedia does not try to establish facts, and certainly not through OR - it relies on credible sources to do that for us - having an opinion of those facts being speculation is irrelevant --Miikka Raninen (talk) 19:34, 4 June 2009 (UTC)


 * There are 10700 news article at present on AF 447 many with notable expert quotations, therefore it is unrealistic to believe that every notable quotation published belongs in wikipedia, and BTW that is what the talk page is for.PB666 yap 00:58, 5 June 2009 (UTC)


 * At this point it seems that *everything* about this accident is speculative. An air accident in mid-flight where no collision happened is very unusual. Happening at the center of an ocean, where helicopters cannot reach without refueling, and none of the neighboring countries have helicopter refuelling capability. No communication of any emergency by the crew. The only data is a few minutes of telemetry, lasting approximately the time it would take for the airplane to fall. Therefore, the only sure conclusion is that a major failure happened very fast, and there is no known mode for such a failure. It could well be that, in a few years, the only appropriate encyclopedia article for this accident will be "Flight 447 was an accident whose cause was never determined". 200.152.98.65 (talk) 02:00, 5 June 2009 (UTC)

Request for definition of alternate law
Could I ask that the useful reference to alternate law is put in the main article? Before tempers get too frayed, the article is far mor informative than anything I've seen from any single news media. JRPG (talk) 20:54, 4 June 2009 (UTC)
 * After googling for a while, I could not find a credible source which would provide the definition for 'alternate law'. There were a number of flight simulator sites which had information about Airbus fly-by-wire, but I don't consider them as appropriate sources for wikipedia since they rarely have references to the actual manuals where they get their data from.


 * There is no article on control laws yet. Someone needs to write that. But an article on Flight_envelope_protection does brush the subject, although not directly giving the definition.


 * Without a link to 'alternate law' we would have to explain what 'alternate law' means, which is problematic - not easy to say with a few words. I think the current text "the fly-by-wire computers had switched to an alternate program used in the event of multiple system failures is a very good explanation, and the ("Alternate Law") isn't required since its a technical term for what the sentense just said. --Miikka Raninen (talk) 21:41, 4 June 2009 (UTC)


 * "the airplane went into 'alternate law' flight control mode. This happens when multiple failures of redundant systems occur." Source: Aviation Safety Network --Miikka Raninen (talk) 21:59, 4 June 2009 (UTC)


 * All done --Miikka Raninen (talk) 22:10, 4 June 2009 (UTC)

Memorial
Is it encyclopedic to discuss the memorial in the Notre Dame Cathedral in Paris? WhisperToMe (talk) 21:44, 3 June 2009 (UTC)
 * Yes, perhaps in a small "Memorial" section. Argel1200 (talk) 22:43, 3 June 2009 (UTC)
 * Thank you - I created one, explained who attended, and added the poem used in the final reading. WhisperToMe (talk) 23:17, 3 June 2009 (UTC)

Verifying Takeoff Time
Ladies, gents - there is a conflict in the takeoff time. It was originally reported at 7:03 pm local time. It was being reported as 7:19 pm local time on CNN today. I just wanted to make a note of this potentially significant discrepancy. I will look around and see if anyone echoes 7:19 pm in print. (by dioxinfreak, not logged in) —Preceding unsigned comment added by 71.169.46.161 (talk) 21:29, 3 June 2009 (UTC)
 * Some sources have cited "takeoff time" as the moment the aircraft pushed back from its gate, which is one of the important times that the ACARS system records. Others have given the "wheels off the ground" time as the takeoff time. Keep that in mind as you search around. AniRaptor2001 01:47, 4 June 2009 (UTC)
 * 7:03 pm is "gate departure" -time, "airbourne" or "runway departure" -time was 7.29 pm. These times have been given since take off. But "estimated gate arriving" -time 11.10 (am) was given by the Airline as late as 10.56 (am), over six houres after the disaster! Heikkiwp (talk) 06:44, 4 June 2009 (UTC)
 * That's not unusual. They didn't have any "real updates", so until they were sure the plane was lost (i.e. it was beyond its fuel reserves and still didn't show up in any monitored airspace) they had no reason to update it and make people worry. Just think of the alternate scenario, you arrive at the airport and find "lost at sea" on the remarks of the arrivals monitor. Then it turns out that there was "just" a communications failure and the airplane actually lands somewhere safely. Not a good idea, right? --Raistlin (talk) 13:40, 4 June 2009 (UTC)

No original research
Please note that the usual WP policies around original research and reliable referencing apply to this article, particularly with respect to speculation about the accident cause. Thank you.  Socrates2008 ( Talk )   23:34, 3 June 2009 (UTC)

High-octane jet fuel?
Jet fuel has generaly very low octane numbers, because it is not supposed to fuel a piston motor. Normally, jet fuel has similar octane ratings as diesel, which would be 10 - 30. I know this is just a detail and a quote from the interview, but couldn't it be refined any further so that it doesn't sound that odd? Thanks.--84.163.114.249 (talk) 00:06, 4 June 2009 (UTC)


 * Refined? "high octane" is nonsense when speaking of jet fuel (as you point out,) so those words should just be removed as it is not a direct quote from a source.  Jz78817 (talk) 00:26, 4 June 2009 (UTC)


 * Replaced the citation with one having a (probably translated) direct quote. Looks better.Kxx (talk) 08:13, 4 June 2009 (UTC)

notable passengers
ALL passengers are notable. This section is unethical !! --boarders paradise (talk) 00:32, 4 June 2009 (UTC)


 * This section is present on many prior articles and noted in WikiProject Aviation/Aviation accident task force/page content. --TAG (talk) 00:45, 4 June 2009 (UTC)
 * While I don't disagree with Boardersparadise, the people mentioned there, with perhaps the exception of perhaps Prince Pedro Luís of Orléans-Bragança, no one there is exceptionally notable. The former conductor MIGHT be worth mentioning if he was currently employed by a major company. Just my two cents.  Hopefully we can come to a consensus with this.  --Hourick (talk) 02:17, 4 June 2009 (UTC)
 * Of course all of the victims are notable in the sense of Wiktionary "notable" entry. That, we sure all agree, is out of the question. Since Wikipedia is an encyclopedia, however we have to consider here in the article, whether the victims meet the Wikipedia notablility criteria for people and therefore justify adding them with their name in the article. Just my humble opionion. --Berny68 (talk) 04:57, 4 June 2009 (UTC)
 * I agree and well said. As soon as it's feasible, then I believe we should edit that particular section appropriately. --Hourick (talk) 07:30, 4 June 2009 (UTC)
 * According to the Wikipedia notablility criteria for people, one possible candidate would be Erich Heine, who was CEO for the Latin American division of Thyssen Krupp. Prince Pedro Luis de Orleans e Bragança would be a very marginal candidate, his only claim is being descendant of an emperor deposed 120 years ago. --vhilden  —Preceding unsigned comment added by 200.152.98.65 (talk) 01:33, 5 June 2009 (UTC)

Death Count
Is it really prudent to already pencil the death count in? Although there is very little chance of any survivors, I think it is more prudent to leave that undetermined until it is officially confirmed (77.102.184.25 (talk) 00:46, 4 June 2009 (UTC))
 * With all due respect, unless they were beamed up with UFO's or landed on an island out there, they have to be assumed lost. --Hourick (talk) 02:20, 4 June 2009 (UTC)
 * Official mourning in France and Brazil, message of condolences from the Pope, how official do you want? We have far more verifiable information that these people are dead than for pretty much any other death reported on Wikipedia. Physchim62 (talk) 08:26, 4 June 2009 (UTC)

Passenger List
The list of passengers looks awkward within the article. Perhaps making it collapsible would be a good idea? AniRaptor2001 01:47, 4 June 2009 (UTC)
 * yes, I agree, or could it perhaps be moved to one side with the writing going on the other side of it? Or is this not allowed? BananaNoodle (talk) 11:38, 4 June 2009 (UTC)

Previous Collision for F-GZCP
Synopsis - F-GZCP was involved in a ground collision that substantially damaged the tail of another aircraft, F-GTAM Airbus A321-211 on a Paris-CDG to Ouagadougou Flight, August 17th 2006. Damage to the F-GZCP was considered as minor. This could be a factor in aircraft stress resulting in failure..

In addition it underwent a major overhaul on 16th April 2009 PB666 yap 02:01, 4 June 2009 (UTC)
 * Good find on the previous accident. This might come in handy in a reference once the investigation is over, but I'm sure inquiries will look into the maintenance logs and find this.--Hourick (talk) 08:27, 4 June 2009 (UTC)

RE Previous Collision for F-GZCP and my post above on trouble in India
See please: Plane that crashed may have been the same one that was grounded for electrical trouble ABOVE

I am not sure what is appropriate. I posted some speculation based on 3 published reports, but here, in TALK, not the actual article. Is this not ok. Seems the same as you have done here, but I was sort of criticized for it and told to wait.

Should we not be looking for sources that, although under-reported, may point to a cause, even if the official announcement or investigation is not complete? Even if only here in TALK? Or should we look for more official sources such as, if they exist, FAA or equivalent accident reports. The stuff quoted above neither confirms or rules out a connection, so I understand why it is not in the articles. But should we not be looking for what happened, whether or not official journalists have reported it? Please correct any misconceptions I have as I am not very clear about the rules. At least I wonder why these reports have not been picked up. (Although I think mine above was mentioned in the Daily Mail or Mirror in the UK)

I wonder, it should be possible to locate official aviation source first hand reports on these. I can look or others can. Would these not be a worthy thing to do? It could either lay to rest any mistaken claims in the sources found so far, or serve to prod professional journalists to look deeper.

〜〜〜〜paul
 * I don't see any objection to recording verifiable factual information about the aircraft. We already have its manufacturer's serial number for example, which is pretty useless information but we know it so we include it just in case someone is interested. You can also be pretty sure that it didn't crash just because it first flew on a 25 February… What we mustn't do in the article is speculate that one fact or another might be important for the crash – leave that to the journalists who are backed up by proper legal teams! The WP rule is let the reader decide. Physchim62 (talk) 09:19, 4 June 2009 (UTC)


 * Well spotted but looks to me like they're different planes (F-GZCP vs F-GZCB). -- samj in out 09:20, 4 June 2009 (UTC)

Related incidents
OK, at this stage this is probably premature, but eventually related incidents should be added to the "See also" section. At the moment, the only one coming to mind is South African Airways Flight 295, which could be related due to black box recovery at high ocean depths. However, while the cause of the AF 447 crash remains unknown, I can't argue for the reference to SA 295 or any other crash. Ron2K (talk) 09:08, 4 June 2009 (UTC)
 * SA295 is already mentioned in the article, for exactly the reason you give! We can have a "See also" section as soon as someone thinks of a link which would go in it…

Categories
I've readded the Category:Aviation accidents and incidents in Brazil. Although it was not in Brazil, the aircraft departed from Brazil and was nearer to Brazil than Senegal. Mjroots (talk) 10:38, 4 June 2009 (UTC)
 * Erm, the category is for "accidents and incidents in Brazil", not near Brazil or that departed from Brazil. The category description also states specifically "Aviation accidents and incidents that took place in Brazil".  Socrates2008 ( Talk )   10:41, 4 June 2009 (UTC)


 * Well, it was the best fitting cat until you created the international one. I suppose that will get an underpopulated tag added at some point. Mjroots (talk) 10:56, 4 June 2009 (UTC)
 * It's no more underpopulated than Category:Aviation accidents and incidents in Togo! There's a guideline somewhere that says that this sort of category is OK even if it's sparsely populated: I'll find it if anyone is really worried. Physchim62 (talk) 11:01, 4 June 2009 (UTC)
 * I've added a few articles to the category. Si bon? <font style="font-family:Monotype Corsiva"> Socrates2008 ( Talk )   11:27, 4 June 2009 (UTC)


 * The accident/crash site is within Brazilian territorial waters (near the Brazilian Saint Peter and Saint Paul Archipelago). "Defence Minister Nelson Jobim confirmed that the spot in Brazilian waters was the crash site of the Air France Airbus A330." (Refs:  ). Therefore, the category Category:Aviation accidents and incidents in Brazil is appropriate.


 * Crash is within Brazil EEZ - it was my reasoning for initial inclusion. As well - in addition to EEZ - aircraft was still in control of CINDACTA-3 ATC, departed from Brasil, involved brasilian citizens and Brasil has declared official mourning. It's definitely related to Brazil and reason to omit it from in category are not clear. --TAG (talk) 22:20, 4 June 2009 (UTC)

Weather circumstances
A reference or two is needed for this section if anyone can find one, if not it should really be deleted. BananaNoodle (talk) 11:42, 4 June 2009 (UTC)


 * I am surprised that the article currently discusses at length a bomb threat from a week earlier in a different country but says nothing about the weather circumstances, since the latter seems to be the current primary focus of speculation about the cause (granted that little more than speculation is possible at this time). --128.30.92.150 (talk) 16:20, 4 June 2009 (UTC)

Wikinews
Should Bomb threat delayed an Air France flight to Paris days before Flight 447 crashed be added to the Wikinews box? 70.29.208.129 (talk) 12:06, 4 June 2009 (UTC)
 * IMHO, no. Physchim62 (talk) 15:41, 4 June 2009 (UTC)

Laws of Airbussing
One can find the very lengthy "Normal Law" here: Since this concerns cruise level control the Ground Mode and Flare Mode are eliminated.

Normal Law

 * High AOA Protection (alpha prot is the normal limit, further stick back increase tension, and stalls are not allowed)
 * Load Factor Limitation (prevents pilot sidestick deflections in excess of stress limits)
 * Pitch Attitude Protection (Range -15 to 30 degrees)
 * High Speed Protection (pitch load up on VelocityMax or MachMax and cannot be overriden)
 * Flight Augmentation (Roll rate is independing of airspeed, per stick deflection the rate response is always the same, bank angle remains static for turns less than 33 degrees without stick deflection, turn coordinated and dampening are calculated by ELAC)
 * Bank Angle Protection (67 degress left to 67 degrees right)

Alternative Law
Control Surface behavior Change in Primary Flight display Changes in speed protections and stability
 * Pitch - load factor demand (Normal Law with less protections) except when gear down.
 * Automatic pitch trim and yaw damping (with limited authority) is available except with ADIRU @ 3 failure.
 * Turn coordination is lost (Turn coordinator is the compensation with rudder on aeliron turn to prevent horizontal sliding of the aircraft, it keeps the nose pointed in the direction of the 'normal' airflow, the coordinator is a pilot option on older jet aircraft) and * Roll degrades to Direct Law - roll rate depends on airspeed.
 * Load factor maneuvering protection-lost
 * Amber XX's replace the green = attitude limits on the primary flight display.
 * Airspeed scale is modified.
 * A nose up command is introduced, overridable, any time the airplane exceeds Maximum Operating Limit Speed or Mach to keep the speed from increasing further. Unlike a road speed limit and aircraft speed limits is referred to as the never go over speed.
 * Aircraft may stall and stall warning capacitated
 * Certain failures cause the system to revert to Alternate Law without speed stability
 * low speed stability function replaces the normal angle-of-attack (AOA) protection (prevents stalls and overspeeds, no fun :^) )
 * System can introduce a progressive nose down command which attempts to prevent the speed from decaying further.

Direct Law

 * Flight yolk and pedal directly (unmodified by flight computer) controls the control surfaces.
 * Flight yolk and pedal sensitivity depends on airspeed
 * Manual trimming is required, An alert USE MAN PITCH TRIM appears on the primary flight display.
 * No protections except overspeed and stall warnings.
 * Primary flight display is as Alternate Law.

Mechanical Backup
PB666 yap 03:37, 5 June 2009 (UTC)
 * Pitch control is made by manually trimming aircraft(requires hydrolic power) with warning in primary flight display.
 * Lateral control is accomplished using rudder pedals(requires hydrolic power)


 * How is the above text related to improving the article? Please note that this talk page is not a forum for general discussion of the article subject. Offliner (talk) 03:53, 5 June 2009 (UTC
 * Agree with user Offliner. This is definitely out of the scope (as you know). What is your point? I don't see why adding this bunch of text improves the article. Not only is that a lot, but it's also very technical. It might be deleted. Just because it is a talk page it doesn't mean you can just write anything you feel like.--Camilo Sanchez (talk) 05:00, 5 June 2009 (UTC)


 * Someone asked the question as to what "Alternative Law" was so this is the answer of what the laws are. IMO if the people who were questioning what Alternative Law is are now satisfied that they have a clear enough definition this can be removed.PB666 yap 12:10, 5 June 2009 (UTC)


 * mh, i think the point being I also wondered if not they went for a more bumpy ride, even that the joke be, 'hear them laugh', this will get them quiet (not aloud I assume). When the dozen or so ignored warnings concerned the weather mostly, this appears the more logical position.


 * Given my rather intense fear of flight and skepticism over the powers of new technology I can only argue that the "Black Swan Theory" applies to the most advanced jet aircraft and thunderstorms. We have to expect that 3 pilots were not careless in their duties.PB666 yap 13:43, 5 June 2009 (UTC)


 * If the reports it was much to far north are correct (errors started in the navigation system) apparently the autopilots (steering) not responding to meteorological report could be blamed(?). I guess it combined, the navigational errors made the pilots take a wrong 'bumpy ride'. This leaves questions wether the meteorological and steering systems shouldn't interact and or warn more independedly (from the navigational set and (its..) pilots interpretations). not improbably it was also struck by lightning while suffering from unexpectedly heavy turbulence limiting the pilots window for reactions, and somewhat explaining the lack of floating victims. Obviously they did put on belts, before entering the worst turbulence. 24.132.171.225 (talk) 05:32, 5 June 2009 (UTC)


 * In answer to the question of weather and Alternative Law. It is possible if turbulence was high enough that then IR and ADR became overwhelmed and were no longer giving consistent information to the computer then it is possible that the flight was automatically switched to Alternative Law and warning lights for various malfunctions occurred. These are computers and the inertia references algorithms probably assume that their will be no 'food processor' turbulence scenarios. As per too far north, and the last Waypoint past, I should remind everyone, there has been no official verified citing of AF 447 debris and we don't know if the coordinates given in the ACARS are correct, although I cannot see why they wouldn't be, or where the aircraft disintegrated. There is alot of contradictory information on the geographic positioning of the final moments of the incident, and now it appears the state of the aircraft as it hit the water. There is absolutely too much speculative information however well refernced on the Main on search and rescue so Caveot Emptor. Wikipedia is not a newspaper.PB666 yap 12:10, 5 June 2009 (UTC)