The 1988 shooting down of Flight 655 as a user interface disaster(octodon.social)
octodon.social
The 1988 shooting down of Flight 655 as a user interface disaster
https://octodon.social/@jalefkowit/111490485825183949
651 comments
Another good write up of this incident is on the excellent Admiral Cloudberg blog:
https://admiralcloudberg.medium.com/the-long-shadow-of-war-t...
Another article in that series is also related to a UX mistake: Air France Flight 447.
This crashed, in part, because the inputs from two control sticks (one for pilot and other for copilot) were averaged if they disagreed, unlike on a Boeing (at the time at least) where they're physically connected so you can't have contradictory inputs in the first place, and you'd feel the other pilot fighting you. When the plane stalled, one pilot correctly pushed down to come out of the stall (after which they would be able to pull back up) while the other pulled up instead (which is wrong but does feel like the instinctively correct thing to do). The inputs cancelled out so had almost no effect. By the end both pilots were pulling up, but that hadn't been the case earlier on when the problem could have been resolved.
https://admiralcloudberg.medium.com/the-long-way-down-the-cr...
The captain had been on a rest brake and only re-entered the cabin at the last moment. He finally figured out the problem but it was too late to do anything.
> “Go on, pull,” Dubois said. Was this comment a sardonic resignation to fate?
https://admiralcloudberg.medium.com/the-long-shadow-of-war-t...
Another article in that series is also related to a UX mistake: Air France Flight 447.
This crashed, in part, because the inputs from two control sticks (one for pilot and other for copilot) were averaged if they disagreed, unlike on a Boeing (at the time at least) where they're physically connected so you can't have contradictory inputs in the first place, and you'd feel the other pilot fighting you. When the plane stalled, one pilot correctly pushed down to come out of the stall (after which they would be able to pull back up) while the other pulled up instead (which is wrong but does feel like the instinctively correct thing to do). The inputs cancelled out so had almost no effect. By the end both pilots were pulling up, but that hadn't been the case earlier on when the problem could have been resolved.
https://admiralcloudberg.medium.com/the-long-way-down-the-cr...
The captain had been on a rest brake and only re-entered the cabin at the last moment. He finally figured out the problem but it was too late to do anything.
> “Go on, pull,” Dubois said. Was this comment a sardonic resignation to fate?
Since we're on the topic, you know how much HN complains about touchscreens on cars? Now find out how replacing old school analog controls with software touchscreens UI partly led to the USS John S. McCain colliding: https://features.propublica.org/navy-uss-mccain-crash/navy-i...
> they saw this contact heading towards them labeled as an F-14 fighter.
What's wild to me is the assumption that Iran would suddenly launch a single F-14 fighter to attack a ship. Was there no moment where they thought "maybe there has been a mistake?" - like where Stanislav Petrov chose to interpret the Soviet early warning system telling him an ICBM was incoming as being a result of some faulty instruments.
Granted the stakes were slightly different - downing one airliner is less severe than risking starting a global nuclear annihilation.
What's wild to me is the assumption that Iran would suddenly launch a single F-14 fighter to attack a ship. Was there no moment where they thought "maybe there has been a mistake?" - like where Stanislav Petrov chose to interpret the Soviet early warning system telling him an ICBM was incoming as being a result of some faulty instruments.
Granted the stakes were slightly different - downing one airliner is less severe than risking starting a global nuclear annihilation.
A quite recent UX disaster at Paypal.
The Paypal iOS app used to behave strange with numbers. The interface was designed that you had to type in the amount you wanted to send in cents. If you wanted to send USD 50, you had to type 5000. Paypal then would add a comma after the second digit from the right. What made it even stranger: The numbers were aligned right, so it had the feeling of typing backwards. I never really got used to it.
A few weeks ago, without a note, the whole interface changed. Now you HAVE to fill in the comma. If you just type in 5000 like you did before, you would send USD 5000 instead of USD 50. I personally know of one person who send way to much money and I suspect it is because of this UX change.
Thoughtful design matters!
The Paypal iOS app used to behave strange with numbers. The interface was designed that you had to type in the amount you wanted to send in cents. If you wanted to send USD 50, you had to type 5000. Paypal then would add a comma after the second digit from the right. What made it even stranger: The numbers were aligned right, so it had the feeling of typing backwards. I never really got used to it.
A few weeks ago, without a note, the whole interface changed. Now you HAVE to fill in the comma. If you just type in 5000 like you did before, you would send USD 5000 instead of USD 50. I personally know of one person who send way to much money and I suspect it is because of this UX change.
Thoughtful design matters!
I wonder when a tesla will show up in a courtroom with a situation like this.
I'm not talking about autopilot.
I'm talking about the continuous (past ridiculous) removal of physical controls from their vehicles.
For example, the original model S/X had dedicated controls for lots of functions - turn signals, gear shift, wipers, autopilot, steering wheel tilt, etc. On the steering wheel, there were two buttons and a scrollwheel on each side of the steering wheel. Press the center of the steering wheel for the horn. The door had mirror adjustment and windows + lock
Unfortunately a few critical controls were on the touchscreen - defrost front and back were big ones, but all the climate controls, and other nonsense too - all pretty much hidden with multiple taps, or small targets or both.
not all of this is bad - putting lots of detailed but non-critical settings like miles vs km are the perfect thing to have on a touchscreen.
but it needed more dedicated controls.
When the Model 3 came out, it started removing controls. There are two stalks, the turn signal also sort of controls headlights and wipers, the shifter is overloaded with autopilot. It has two scrollwheels without buttons, you have to push them left and right.
all other controls are on the touchscreen.
It really needs dedicated controls for important things.
And then the updated model S/X came out. wow.
there are NO stalks. turn signals are touch areas on the steering wheel. so are high beams, horn, wipers. the scroll wheels do different things at different times.
shifter? nope - it guesses what direction you want to go. many more things involve the touchscreen, like going into park. (there is also a touch drive selector in the center console, but you have to look down and touch it to wake, then to select)
Just a mess. It makes you a worse driver.
I'm not talking about autopilot.
I'm talking about the continuous (past ridiculous) removal of physical controls from their vehicles.
For example, the original model S/X had dedicated controls for lots of functions - turn signals, gear shift, wipers, autopilot, steering wheel tilt, etc. On the steering wheel, there were two buttons and a scrollwheel on each side of the steering wheel. Press the center of the steering wheel for the horn. The door had mirror adjustment and windows + lock
Unfortunately a few critical controls were on the touchscreen - defrost front and back were big ones, but all the climate controls, and other nonsense too - all pretty much hidden with multiple taps, or small targets or both.
not all of this is bad - putting lots of detailed but non-critical settings like miles vs km are the perfect thing to have on a touchscreen.
but it needed more dedicated controls.
When the Model 3 came out, it started removing controls. There are two stalks, the turn signal also sort of controls headlights and wipers, the shifter is overloaded with autopilot. It has two scrollwheels without buttons, you have to push them left and right.
all other controls are on the touchscreen.
It really needs dedicated controls for important things.
And then the updated model S/X came out. wow.
there are NO stalks. turn signals are touch areas on the steering wheel. so are high beams, horn, wipers. the scroll wheels do different things at different times.
shifter? nope - it guesses what direction you want to go. many more things involve the touchscreen, like going into park. (there is also a touch drive selector in the center console, but you have to look down and touch it to wake, then to select)
Just a mess. It makes you a worse driver.
Aside: it is the second day in a row that a thread from Mastodon ends up on the frontpage of HN. When was the last time I Twitter thread did the same, and does anyone else doubts that Twitter is no longer at the center of tech-related conversation?
I wonder how this compared to plain air traffic control technology of that era? Were ATC doing it better? (Obviously there are a lot of ATC-driven disasters of that era too).
I first thought this was a post about flat design, then went and read the whole toot. Man, who'd expect UI to be bad enough for life and death operations.
Someone should post more details about the actually confusing UI that lead to this event. Would be a good lesson to most of us.
It also appears that Lockheed won a contract for using this system in 2023 [1]. Can someone share if they actually fixed the UI issues with it recently?
1 - https://news.clearancejobs.com/2023/03/13/lockheed-martin-wi...
Someone should post more details about the actually confusing UI that lead to this event. Would be a good lesson to most of us.
It also appears that Lockheed won a contract for using this system in 2023 [1]. Can someone share if they actually fixed the UI issues with it recently?
1 - https://news.clearancejobs.com/2023/03/13/lockheed-martin-wi...
Admiral Cloudberg has a detailed discussion of the incident, including the user interface behaviour:
* https://admiralcloudberg.medium.com/the-long-shadow-of-war-t...
Reading the different takes on the incident leaves one with the impression that there was a lot more going on there than just the stuff with the user interface.
* https://admiralcloudberg.medium.com/the-long-shadow-of-war-t...
Reading the different takes on the incident leaves one with the impression that there was a lot more going on there than just the stuff with the user interface.
What an insane story. Crazy to think that some of these engineering errors led to such a disastrous outcome...
I do resonate with someone's comments in the above sections reflecting on whether or not such a narrative would be given if it was say, a UI mistake from an enemy country.
I do resonate with someone's comments in the above sections reflecting on whether or not such a narrative would be given if it was say, a UI mistake from an enemy country.
> every large aircraft in the world is equipped with a device called an IFF -- "identification friend or foe."
This is weird, I'm sure I don't have all the facts. In a conflict, why would a military aircraft ever want to identify itself?
This is weird, I'm sure I don't have all the facts. In a conflict, why would a military aircraft ever want to identify itself?
Greetings from Ukraine, country at war.
You may wonder, but computers extremely important in our life, I bet, you don't release how we depend on high tech now (was not such before war).
I will list how I remember, not by importance.
1. Electric grid is core of infrastructure, and it depend on computers. Even if in many cases could use decentralized power sources, but need it to feed bank networks for money payments; water supply, heating, also need electricity. And Russians actively use hacker attacks to turn off power in Ukraine.
2. Air defense. God bless America and developers of Patriot! After last spring we received this wonderful tech, we could feel much safer, just periodically hear "Citizens may hear explosion, as system automatically intercept ballistic missile. Alarm siren was not triggered".
3. For our people in trenches, information is literally life, because Russians have large number old, but dangerous tech, like artillery and air approx 1960s. And now also got cheap new tech from East - most annoying drones (UAV).
So if one side got knowledge, where some opponent military unit, they immediately fire on it if possible.
Even become usual thing, artillery duels, when artillery units on each side, try to destroy each other. And as they are long distance units, they don't see opponent, just know from intelligence, that on some coordinates appear opponent.
You may wonder, but computers extremely important in our life, I bet, you don't release how we depend on high tech now (was not such before war).
I will list how I remember, not by importance.
1. Electric grid is core of infrastructure, and it depend on computers. Even if in many cases could use decentralized power sources, but need it to feed bank networks for money payments; water supply, heating, also need electricity. And Russians actively use hacker attacks to turn off power in Ukraine.
2. Air defense. God bless America and developers of Patriot! After last spring we received this wonderful tech, we could feel much safer, just periodically hear "Citizens may hear explosion, as system automatically intercept ballistic missile. Alarm siren was not triggered".
3. For our people in trenches, information is literally life, because Russians have large number old, but dangerous tech, like artillery and air approx 1960s. And now also got cheap new tech from East - most annoying drones (UAV).
So if one side got knowledge, where some opponent military unit, they immediately fire on it if possible.
Even become usual thing, artillery duels, when artillery units on each side, try to destroy each other. And as they are long distance units, they don't see opponent, just know from intelligence, that on some coordinates appear opponent.
That time when a badly implemented government contract drop down menu UI design was blamed for the false alarm Hawaiian incoming ballistic missile emergency sms [1] only for it later to turn out to have been caused by regular old human communication error and poor safeguards [2].
[1] https://blog.prototypr.io/dangerous-drop-downs-%EF%B8%8Fbad-...
[2]https://en.m.wikipedia.org/wiki/2018_Hawaii_false_missile_al...
[1] https://blog.prototypr.io/dangerous-drop-downs-%EF%B8%8Fbad-...
[2]https://en.m.wikipedia.org/wiki/2018_Hawaii_false_missile_al...
There seems to be a lot of trust in IFF in this incident, what would stop Iran from reprogramming their F-14's IFF to identify as as civilian aircraft (or if reprogramming isn't feasible, retrofitting a stolen civilian airliner's IFF system into an electronics pod on their F-14, I'm sure the Iranian government wouldn't have any trouble getting an aircraft from Iranian Airlines)?
Do warships do any verification beyond reading the IFF ping aside from looking for attack patterns like climbinb/decending, etc, it still seems possible for an attacker to get within missile range while still appearing on radar to be a civilian craft.
Do warships do any verification beyond reading the IFF ping aside from looking for attack patterns like climbinb/decending, etc, it still seems possible for an attacker to get within missile range while still appearing on radar to be a civilian craft.
I'm so conflicted reading this story. On one hand, yes, there were choices made during the design of the system that directly contributed to this tragedy. And a lot of innocent lives were lost, so saying that's "shit happens, it's an edge case" rings very hollow.
On the other hand, this was a very peculiar set of circumstances, very much an edge case. Is it reasonable to expect designers of combat systems to triple check their choices and run more test scenarios to identify and address such edge cases? I'd say yes. However, I think it's unreasonable to expect them to design a perfect system for a highly volatile and chaotic use case such as war.
On the other hand, this was a very peculiar set of circumstances, very much an edge case. Is it reasonable to expect designers of combat systems to triple check their choices and run more test scenarios to identify and address such edge cases? I'd say yes. However, I think it's unreasonable to expect them to design a perfect system for a highly volatile and chaotic use case such as war.
Is this a mastodon instance? Mastodon is a twitter clone that doesn't allow long posts? All the posts but the first come collapsed and I have to click on each to read it. Is someone measuring engagement?
This could have been an article worth reading, instead it is chunked in tweets (or whatever they are named now) on X. I'd be interested in the topic, but hate the reading experience to much to read it.
Everything's possible, but there would be no debate about UI mistakes if it was Iran shooting down a US plane. They would've done because they are evil by nature, or at least perceived as such. In that case the media and the public buys into its own reality, but of course the UI discussion could be a distraction from the public maybe starting to question if that's actually the reality.
Also from the Wiki page about this shootdown:
In 1991, political scientist Robert Entman of George Washington University compared U.S. media coverage of the incident with the similar shootdown of Korean Air Lines Flight 007 by the Soviet Union five years earlier by studying material from Time, Newsweek, The New York Times, The Washington Post and CBS Evening News. According to Entman, framing techniques were used to frame the Korean Airlines incident as sabotage while framing the Iran Air incident as a tragic mistake,[67] stating "the angle taken by the U.S. media emphasized the moral bankruptcy and guilt of the perpetrating nation. With Iran Air 655, the frame de-emphasised guilt and focused on the complex problems of operating military high technology."[68][a] By "de-emphasizing the agency and the victims and by the choice of graphics and adjectives, the news stories about the U.S. downing of an Iranian plane called it a technical problem while the Soviet downing of a Korean jet was portrayed as a moral outrage."
Also from the Wiki page about this shootdown:
In 1991, political scientist Robert Entman of George Washington University compared U.S. media coverage of the incident with the similar shootdown of Korean Air Lines Flight 007 by the Soviet Union five years earlier by studying material from Time, Newsweek, The New York Times, The Washington Post and CBS Evening News. According to Entman, framing techniques were used to frame the Korean Airlines incident as sabotage while framing the Iran Air incident as a tragic mistake,[67] stating "the angle taken by the U.S. media emphasized the moral bankruptcy and guilt of the perpetrating nation. With Iran Air 655, the frame de-emphasised guilt and focused on the complex problems of operating military high technology."[68][a] By "de-emphasizing the agency and the victims and by the choice of graphics and adjectives, the news stories about the U.S. downing of an Iranian plane called it a technical problem while the Soviet downing of a Korean jet was portrayed as a moral outrage."
It's important to note there are misrepresentations in the Wikipedia article:
https://en.wikipedia.org/wiki/Iran_Air_Flight_655
For the most truthful view of what happened that day:
https://www.britannica.com/event/Iran-Air-flight-655
https://en.wikipedia.org/wiki/Iran_Air_Flight_655
For the most truthful view of what happened that day:
https://www.britannica.com/event/Iran-Air-flight-655
When a foreign country kills US citizens, it's terrorism. When the US kills foreign civilians it's collateral damage or a UI disaster ... business as usual.
Would just like to jump in and remind folks of the Therac-25 incident just a couple years before this (1985 and 1987), and the user interface was also identified as a contributing factor.
https://www.cse.msu.edu/~cse470/Public/Handouts/Therac/Side_...
https://www.cse.msu.edu/~cse470/Public/Handouts/Therac/Side_...
Tl;dr
- some friend/foe “cursor” was not locked to a jet trajectory
- a fighter jet flew through it later, classifying it as a threat
- dashboard shows no altitudes
- identifier reuse led to invalid ascent/descent check
- captain decided it’s a classic attack profile
- some friend/foe “cursor” was not locked to a jet trajectory
- a fighter jet flew through it later, classifying it as a threat
- dashboard shows no altitudes
- identifier reuse led to invalid ascent/descent check
- captain decided it’s a classic attack profile
"I hope it's not a civilian flight" -- low ranking guy in the CIC who was ignored.
The Wikipedia article on the incident reads like it was written by an Iranian government employee.
The whole thing would have been avoided if the US didn't push their Navy around Iran
why the hell is there a US navy ship there to begin with
and why the hell is it that easy for them to shout at whatever is moving in the sky
yes it might have been a UI issue
but the root of the problem is "a US navy ship next to iran's border"
i wonder if ideas like side by side deployments were on their radars back then. maybe pretty expensive, but it seems like with systems of people and machines that complicated you'd want live validation of everything before cutting over.
If I remember correctly, the Lockerbie / Pan Am attack was in response to this.
Very impactful mistake, but apparently not a mistake worth apologizing for.
From Wikipedia: "the U.S. government did not admit legal liability or formally apologize to Iran"
Pinnacle arrogance/exceptionalism. Disgusting.
Oh, and the US back then supported Iraq in its horrific attack on Iran.
https://en.wikipedia.org/wiki/United_States_support_for_Iraq...
From Wikipedia: "the U.S. government did not admit legal liability or formally apologize to Iran"
Pinnacle arrogance/exceptionalism. Disgusting.
Oh, and the US back then supported Iraq in its horrific attack on Iran.
https://en.wikipedia.org/wiki/United_States_support_for_Iraq...
I’m guessing it’s not a popular opinion around here, but what’s described is so insanely dog shit I would have been fired and sued if I had been responsible for developing anything even close to how stupid this system was.
I know people who have worked with military systems, and as dumb as the process around developing these things tend to be, due to the asinine cluelessness of the decision makers, none of them would have accepted even a fraction of the idiocy on display here.
I mean, not “slaving” the cursor still feeds garbage data into the system, immediately reusing 4 digit codes in a way that is pretty much guaranteed to create confusion, no real cooperation or coordination between the same system running on ships in the same battle group, supposedly essential data displayed without the actual essential details shown to decision makers in charge of death and destruction…
It would be laughable if it was the worlds strongest military we were talking about.
The people in charge of deciding to make this “system” in this way should be in prison.
I know people who have worked with military systems, and as dumb as the process around developing these things tend to be, due to the asinine cluelessness of the decision makers, none of them would have accepted even a fraction of the idiocy on display here.
I mean, not “slaving” the cursor still feeds garbage data into the system, immediately reusing 4 digit codes in a way that is pretty much guaranteed to create confusion, no real cooperation or coordination between the same system running on ships in the same battle group, supposedly essential data displayed without the actual essential details shown to decision makers in charge of death and destruction…
It would be laughable if it was the worlds strongest military we were talking about.
The people in charge of deciding to make this “system” in this way should be in prison.
The two original founders recently retired and our new CEO is a former Captain of the USS Zumwalt.