That monostable shifter is just an incredible bad idea. It is baffling why they do it that way. Was shifting a huge problem which needed to be solved in automotive industry? Guessing someone wanted a promotion by claiming "I redesigned the shifter".
I understand when GMail kept changing its interface every 6 months -- it would correspond probably with year / mid-year reviews and ever designer probably wanted to say "I redesigned blah blah". That is just email. But this is a car and it can easily kill people. Why couldn't they just instead add more lights to the dashboard, or rear view cameras or something of that sort.
I understand your point; but we should remember that the redesign of the shifter did not happen in isolation.
Car designers have been tinkering with controls for decades. Sometimes the addition of new ideas can be very helpful. For example, I had a Ford Taurus back in the late 90s. Buttons that had a "positive" effect had a convex portion; buttons with a "negative" effect were concave. Door locks, cruise control, etc., all worked this way. It was very easy to operate that car without looking, or in the dark.
Now add to that the computers-do-everything reality of the modern world. Controls that used to be directly hooked to some mechanism no longer are; software reads the control, and software initiates the desired effect. This means that the form of controls is no longer dictated by the mechanisms they relate to. Designers are free to design controls that users will find easy to use and intuitive.
And so designers have tried all kinds of things. But, alas, some of them turn out to be bad ideas. I don't say this excuses the design flaws at issue here, but I do think it does go a long way toward explaining them.
[Meanwhile, I have a another Ford Taurus. No convex/concave buttons. I have no idea why not. :-( ]
A classic in aircraft design are the levers that control the flaps have a flap-shaped knob, and the control for the nose gear has a little tire for the knob. This is all so the pilot knows what control his hands are on without looking at it.
But as a cockpit designer told me once, everything that seems intuitively obvious in cockpit design was paid for with blood.
> But as a cockpit designer told me once, everything that seems intuitively obvious in cockpit design was paid for with blood.
pretty sure the same is true for a lot of other things that needed improvement, EVEN if the designer(s) knows beforehand a possible failure MIGHT results in human injury / death.
I'm of the opinion that user interfaces should not be patentable, specifically so that when people find safer interfaces everyone else can copy them and standardize on them, rather than trying to carefully avoid them.
The move from a traditional shifter to a monostable shifter is an obvious step backwards for one, painfully obvious, reason:
It gives the user less information.
With a traditional shifter, looking at or feeling the position of the knob immediately tells you what gear you're in. The monostable shifter takes that away.
The designers probably thought that, if the sports-car segment can get used to paddle shifters, perhaps a stick-form of the paddle-shifter could fly with other users. This may have been true were park not thrown into the mix too.
The parking is actually done using a parking brake, not the gearbox.
Cars with manual transmission do not even have a notion of a parking gear.
The car could also automatically engage the parking brake (and optionally a high gear) if the ignition key is removed. Problem solved, lives saved, easy fix?
Heck, make this system mandatory, like airbags and safety belts.
Been riding in/driving Tauruses for my whole life. When I get in another car that is missing features like this I always fail to figure out how things work, and even when I do I have forgotten by the next time I get in. I feel like the somatosensory experience of using a vehicle or a tool is completely neglected by designers who only ever look at a digital model. Yes, it may look good to you Mr. Visual, but have you actually touched it, or used it, played with it? I bet some of designers get an itchy feeling when they see the bumps on the F and J keys.... (end rant).
Astoundingly, in the video in the linked article, the person doing the demo incorrectly reports the gear he shifted into - he reports shifting into neutral when he really was in Reverse! There is also just one chime when opening the door which could just as easily be "you don't have your seatbelt on". Just amazingly bad.
I have a 300 with that shifter. If you open the door with the car not in gear it will bong bong bong bong bong bong bong at you.
But it'll also do that if you open the door with the engine running.
I always found it amusing too. For a car, the cars that have that ZF shifter are really heavy cars.
The other amusing thing is HOW MUCH WORK they put into that shifter. It's doesn't just bounce back to its home position. It's also got some kind of ratcheting mechanism in it, so that when you shift from park to drive you can "feel" the shifter clicking from park to reverse to neutral to drive. That mechanism has lots of clickety parts in it that will probably fail with time.
Why the heck did they do that?! Of all the things you could do with an electronic shifter this seems like the worst thing you could possibly do. They could have used a dial like they do now ,or even buttons. Anything. Why try so hard to replicate something - and take up so much space internally, when they could have just used the real deal?
The gated shifter that they used on the 5 speeds was a better shifter anyway.
Was shifting a huge problem which needed to be solved in automotive industry? Guessing someone wanted a promotion by claiming "I redesigned the shifter".
That is exactly how I feel whenever I drive a recent BMW. Besides the shifter shenanigans, the turn signal switch also bounces back to center immediately after activation, such that it is virtually impossible to deactivate manually (e.g. when changing lanes) and I always end up alternating left and right flashes before getting it to turn off.
There's probably something simple that I'm missing, but really, I can't think of a single problem this solves or improvement it makes over the existing convention. It seems like having the physical position of the switch map to the state of the signal is a good thing, and that is now lost for...what exactly?
For changing lanes, you press the signal shifter part way without clicking. The shifter will return to center and the blinker will blink 3 times for your lane change. Only press the shifter all the way so it clicks when making a turn. Then, the steering wheel releases the click after the turn.
Coincidentally, the BMW also has a similar transmission shifter to the Jeep. Push forward for Reverse, then it returns to center. Pull back for Drive, then it returns to center. Park is a button on top.
I once borrowed a friend's car (fiat palio) and I literally couldn't get it out of parking. The reverse gear can only be selected if you lift the shifter and then shift. I felt extremely silly calling my friend and asking him how to shift to reverse.
> It seems like having the physical position of the switch map to the state of the signal is a good thing, and that is now lost for...what exactly?
Digital control, I'd guess. It's like those IoT light switches that are just push buttons sending a signal to the control server, because making them reflect the actual state of the light would require putting a motor inside the switch.
Now what you win by making turn signal controllable by a computer? I don't know. Maybe it's needed for the autopilot/driver assist tech everyone is putting in their cars now?
In other parts of the world where manual transmission is still popular, the leftmost pedal is the clutch. Shifting the accelerator to the left would be an instant recipe for disaster on the roads.
Some comments here completely miss the key point of the article and instead go down the route of whether the shifter itself should change or not.
The blunder (and yes, it is a blunder and not only in hindsight) in the shifter is that it provides poor feedback to the human. While changing technology may render some types of control not necessary, and thus correctly invite new design thinking, the fundamental human need for feedback does not change. In fact, when something that is so habitual to you changes, that is precisely when the design should provide even more feedback because otherwise you are likely to miss it. This is amplified if the consequences of not comprehending ones actions are high - which clearly was the case here.
The designers here dropped the ball. There is no way early feedback through rigorous testing on prototypes would have not identified this issue way before it went into production.
Following up on feedback for the user, for automobiles designers should opt to communicate feedback in non-visual ways.
You know which gear your car is by the position of the shift lever. You know when to shift up/down by the sound of an engine. You know what speed the AC fan is at by the direction of the dial and can crank it up and down accordingly.
On an older car, a driver can be blindfolded and operate many of the accessory functions without trouble. On a newer car, I wager it's much difficult, especially those with touchscreen interfaces.
Yes, there are more functions now, but should they be added at the expense of our safety? I want to be able to switch on the defogger by a simple button, not by diving into a menu.
> Some comments here completely miss the key point of the article and instead go down the route of whether the shifter itself should change or not.
Well, the point of the article is "bad design can be dangerous", but there is another point that can be discussed which is: why change what works?
Regarding the shifter, the traditional shifter is physically connected to the gear box; in newer cars where everything happens via software, the physical connection no longer exists, and so designers thought there was an opportunity to redo it.
But, just because you can doesn't mean you should.
That said, your point about necessary feedback is extremely valid; it is believed that the crash of the Air France flight 447 in 2009 is due in no small part to insufficient feedback (either pilot could take back control of the plane, without the other pilot knowing they didn't have control anymore).
It was actually worse than that in the Air France case -- if the rudder inputs from the two pilots were different, they were averaged before being applied to the physical control surfaces.
It could work both ways: either with one pilot in control or with both.
The averaging thing was not as stupid as it sounds when values are close.
It's completely crazy when both rudders were in the opposite direction at the maximum (average of -100,+100 = 0), but it must have seemed an impossible case to the designers...
I have that shifter on my Jeep Grand Cherokee. It's been recalled but I actually love it. Maybe I'm an outlier but it's not that dissimilar from electronic shifters in BMWs and Mercedes. My girlfriend's Mercedes acts like a turn signal on the steering column and it's 10x more confusing to operate. I hate it. The primary safety difference is that it auto-parks itself when the door is opened. Chrysler's recall adds the same self-parking functionality.
I don't agree that shifters should work the way they've always worked. They warrant a redesign because long throwing shifters eat up a bunch of precious center console space. An area where stuff is often over-packed by the driver or co-pilot; phones, cups, coins, paperwork, keys, wallet. Most of this demands short-term, easy access storage.
If better safety mechanisms were considered in the Chrysler, I doubt we'd be talking about this shifter as a failure. Nobody is criticizing Mercedes' exceptionally poor design – unless you read car enthusiast forums – because they added better safety processes.
I know the BMW shifters you're talking about, but there is one significant difference: the mechanism for entering Park is not like the rest of the gear selection options (Drive, Neutral, Reverse). It's a button on the top of the shifter. As such, when you want to put the car in park, you push a button, you don't move the shifter.
I'm not saying this is a great shifter design overall, but it DOES mean that you quickly learn a completely different routine for putting the car in park. Basically: "If I want the car to be moving, then move the shifter. If I want the car to stop moving, push the button."
There is a separate Park button, and I think there is no risk of confusion. I have actually occasionally powered down the car when I just wanted to put it in park, but there is no real danger in that.
It automatically engages "Park" when powering down, but not when exiting the car, but maybe it should as long as the car is not moving when the door is opened.
Actually, as the car is aware of that no one is sitting in the drivers seat, it would be trivial for Mitsubishi to program it to prevent the car from moving, or at least starting to move from a stand still using the ordinary brakes.
I guess that would be true for all cars with airbags, which would be every car sold, at least in in USA and EU.
Actually, even now the car can not be towed if it is out of power.
To tow, you need at least a booster battery to boot the 12v systems in the car. Or you need to lift it. Not a problem for an road assistance service, but potentially irritating if you are on your own.
The traction battery does not have to be online though.
An automatic gearshift doesn't belong in the center console, it belongs on the steering column. I've driven Ford trucks and SUVs too long for me to think otherwise.
Steering column shifters have worse feedback re: what gear you're in than center console shifters. Neither one is as bad as a monostable shifter though.
Idiots? nah. More that one develop mental myopia if one wander the same proverbial halls all day. One start to expect things others may not, or overlook details.
Its how the mind can operate with such apparent efficiency. After all its primary concern is danger, procreation and food in that order.
And danger is in nature most often something that is not present every day, and thus stands out like the proverbial sore thumb once it shows up.
This angers me as a consumer of automobiles. I abhor poor automobile design and this takes the cake. This is also why I hate
a) air conditioning and radio controls that do not have clear positions for their status, forcing me, the person driving a big automobile in traffic, to divert my gaze to inspect them to ensure I have the correct setting
b) Tesla's monolith faceless screen which forces me to stare at it constantly, taking my eyes off the road when I drive one
On the flipside of that, my LEAF has a big button in the middle called "Auto". Hit that, and it's usually what I want (or damn close enough) if I'm at a time when I can't glance at them.
Front defog is the only thing that I might need to look at in any kind of a time-sensitive fashion. In my experience, setting defog on any other car is harder than mashing the two buttons the are on in that area (defog and rear defog).
I agree that if I want to manually set AC compressor on/off, fan speed 0-7, face, face/feet, feet, feet/windshield, windshield-only, that it's harder on the LEAF. Most times, I don't want that. I want either "defog" or "I'm a little uncomfortable; please make me more comfortable".
Tesla also has a monostable shifter. I've found myself in a different state than I thought I was in on more than one occasion. It's remarkably easy to bump it and trigger a shift out of park.
This seems to be something the medical industry hits every month with medications and materials. The human is prone to an incorrect interpretation of the specifics and may take action or not take action improperly because of that.
How to identify similarly spelled medications as different? Is the dosage or dilution magnitudes apart without visibly distinct and obvious clues?
More information about how the medical world handles these kinds of issues is at the Institution of Safe Medical Practices [1].
I heard recently of a patient that took some prescribed medication before her surgery, but instead of taking 5 mg as prescribed she ended up taking 5 grams, which induced a near fatal stroke. You could argue that better design could have prevented this from happening.
Edit:
The mistake was between mg and ml, so the person drank 5ml of a 5X concentration drug. Either way a terrible, preventable result.
Boeing had just transitioned their FQIS from lbs to kgs (aircraft fuel is always weighed, volume changes with temperature). The flight crew hadn't had enough familiarisation and miscalculated.
All aircraft now calculate fuel usage in SI units.
Jet fuel is typically planned by weight. Avgas is typically planned by volume. In general aviation, both are typically priced and sold by volume.
> All aircraft now calculate fuel usage in SI units.
All 3 of my aircraft track/calculate fuel usage in imperial units. Likewise, everyone I know in US general aviation also still uses imperial units (lbs, lbs/hr, gallons, gal/hr).
I find it fascinating how many chat applications exist, and how success is almost completely due to design. Most chat features you find in newer chat apps, like Kik or Whatsapp existed since AIM. Yet users are very willing to jump ship and grab onto an application with the same, perhaps fewer, feature if it's more eye-catching, or intuitive. It's a nice reminder that part of innovation in the tech industry is UI, and if you innovate a new interface for an existing service, it can be as ground-breaking as any other innovation in the industry.
AIM and similar gen IM protocols expected that you were using a desktop or laptop with a persistent connection. This would break on iPhone because Apple didn't allow backgrounding.
Kik and similar took over once Apple introduced their push notification service. This allowed apps to behave closer to text messaging.
Does anyone know how NASA modified presentations after the report mentioned in the article? Only show one single piece of data per safety slide? Mandate large fonts and color coding on safety slides?
The design of pilot controls in aircraft has a great deal of effort going into making them intuitive, and still serious mistakes are made. It's a very hard problem, and what's obvious in hindsight escapes nearly everyone in foresight.
It just doesn't make sense for -so many- reasons. Besides the obvious that the cleaner literally looks identical to a neon-colored children's drink, why would anyone possibly decide to put them on the same aisle? The cleaner even has similar pricing to the beverages.
Some stupidity I can understand, but this? How does this possibly get past anyone in the chain of command? Feel like this requires a dedicated, concerted effort in avoiding any thought whatsoever to occur.
First gear is all the way down and to the right. Then left for 2nd, up and left for 3rd, up for 4th, right (!!!) for 5th (aka drive). Now it gets dangerous, if you go up and right you get reverse.
Visually, it may seem confusing, but tactilely, it's intuitive:
Move your hand from the bottom-left corner to the top-right. That diagonal motion is shifting from D-N-R-P, what most people will use.
Now pretend you're driving in the mountains and you're on a long descent. Instead of riding your brakes, you want to use the transmission to limit your speed.
You're in 4th gear (D), so just bump your lever down to third (3). It's a simple vertical motion. Now you're in 3rd gear, but forgot and think you're still in 4th (D), and attempt to downshift once more. Fortunately, you won't accidentally drop the car into second gear. To get into second gear, you need a downward and horizontal motion. It requires intention to move from 3rd to 2nd gear.
And once you're in the rarely used 2nd gear, to switch between second (2) and first (L) is a horizontal motion.
So it makes sense once a person feels it out and actually drives their car. Designers with a background limited to just visual aesthetics need to learn this. Run through your interfaces; interact with them under less-than-ideal conditions.
You're right, but I've only looked at it and run the design through my head.
If I were sitting in a car, I would would know that 4-D is basically shifting between fourth and overdrive. And if someone accidentally did shift from overdrive down to fourth, it won't kill the car in most circumstances. It's better than the old O/D button that asks your thumb to press it.
Ultimately, my proposal that this shift pattern groups gears by use-cases still holds true. D-4, 4-3, 2-L. If a person can't understand that, it's perfectly fine to leave it in D.
Consider this scenario: you're shifting up from 1st gear and forget what gear you're in. If you're in 3rd gear its safe to shift up and to the right. If you're in 5th gear its not safe, up and to the right is reverse!
I don't agree with your analysis. Toyota's shifter is unsafe.
First off, please explain to me the scenario where you're drag racing someone and limiting your upshifts in an automatic transmission car. Okay, so you're in 5th gear (D), which happens to be overdrive. If you know what overdrive is, then you'd know that the possibility of redlining in that gear is next to none.
Lastly, with any automatic-transmission car made in the past 20 years, they won't let you shift into reverse from drive.
It might look crazy, but I've never had a problem using it, without looking. Certainly much, much better than the ridiculous Jeep shifter.
Re: the gear order, since it's an automatic you spend 99% of your time going between park, reverse, and drive, so the only one you regularly go through is neutral. It's very rare to need, say, first gear.
Edit: Oh, and I've never left it in drive thinking it was in park. Park = all the way up = hard to screw up.
This guy goes to a dealer and buys a fully decked out luxury BMW. Fifteen minutes later, he calls the dealership complaining that they sold a defective car and that it broke down on the highway. They apologize, bring out a tow truck, check that yes, the gearbox is dead, apologize again, and give him another car. Fifteen minutes later, he calls again, totally angry, stating that this one is defective as well. The dealers are surprised, bring out a third car, and join with him on a test drive to see what happens. So, they sit in, and the guy starts showing the gearbox problem - start the car, put in 1st gear, drive off, put it in 2nd gear, put it in 3rd gear, drive onto the highway, floor the gas, put it in 4th gear, enjoy the ride, put in 5th gear, and now the Rocket mode! ...
Bad UI can happen to everyone and for a lot of reasons: I think the minimalism thing can get to stupid levels, the music controls on the ipad mini has the controls so cramped that I get a friend to prove than less than 50 percent of the time I can move the sound without pressing the rewind arrow.
Speaking of bad design, on mobile the text on that page slightly flows off the right edge in a way that interferes with reading and can't be fixed by zooming.
That monostable shifter is just an incredible bad idea. It is baffling why they do it that way. Was shifting a huge problem which needed to be solved in automotive industry? Guessing someone wanted a promotion by claiming "I redesigned the shifter".
I understand when GMail kept changing its interface every 6 months -- it would correspond probably with year / mid-year reviews and ever designer probably wanted to say "I redesigned blah blah". That is just email. But this is a car and it can easily kill people. Why couldn't they just instead add more lights to the dashboard, or rear view cameras or something of that sort.