Humanized Interface Puzzler #1

Fun UI Design Fundamentals

Welcome to the first installment of the Humanized Interface Puzzler. For your fun, bafflement, and desire for free stuff, we’ll pose an interface design puzzler on a semi-regular basis. To enter, simply send your answer to puzzler@humanized.com by the deadline. We’ll select the best answer and post it on our blog. Then, we’ll send the winner a limited-edition* Humanized shirt and entrance to our beta program.

The first puzzler is about modes and cars.

An interface has modes if one gesture can mean different things, depending system state. Modes are at fault when you miss a call because your phones in silent mode. And there’s little worse than having the final bars of Appalachian Spring – with harmonies as delicate as frozen cobwebs – thrashed by a cellphone who’s owner has forgot to put it into silent mode. Perhaps there is something worse: having it be your cellphone. You can read all about modes, modes errors, catastrophic mistakes, and some solutions in our article Visual Feedback: Why Modes Kill.

Modes can cause misery and sometimes they can be dangerous. Automatic cars have at least two modes: drive and reverse. It’s a mode because the same gesture (stepping on the gas pedal) does one thing if the car is
in drive and another in reverse. I’ve had run-ins with other cars and a wall because my car was accidentally in reverse. I’m sure that others have had the same experience.

So what can be done? I have heard it argued that this mode is necessary. Mechanically speaking, forward and backward are completely different: you are either in drive, or in reverse. They even use different sets of gears. Also, consider the impracticality of turning the wheels 360 degrees. Even if it was possible, how do you know which way your wheels were pointing? How would you know whether the car would be moving forward or backward? Wouldn’t you make just as many mistakes as before, if not more? Forward and backward are being dictated by the driver as two distinct modes.

This is the challenge:

Is it possible to design a car that isn’t forward/reverse modal? If it isn’t possible, why? And if it is possible, how?

The last day will accept answers is January 1, 2007. Of course, if you have a great solution and it’s after the deadline send it anyway! We’d still love to hear from you.

* By limited-edition we mean that one is printed each time someone wins.

(Note: this is a very basic idea)

Off the top of my head, I have one suggestion: a quasimodal steering wheel. Two opposite portions of the steering wheel could be pushed back, or have two buttons that need to be pushed to make the car go backwards.
How’s that sound?

I remember something like this from a human factors class I took in design school. They where saying that the gas pedal sometimes is confused with the break pedal, especially with elders. A solution was one pedal; tilting forward (your toes) moved the car forward, tilting it back (your heel) was the brake.

Remove the hand operated gear shift and replace it with a pedal for the foot not operating the accelerator pedal.

Tipping this “new” pedal forward or backward shifts the mode of travel and puts the vehicle into park in the neutral position.

Not radical enough!

Ever since I was a teenager struggling to learn to drive, I have always wished for a car which you controlled with a self-centering analog joystick. Just push with varying pressure to go faster in any direction. Centering the joystick stops.

In fact, I never really got the hang of it, and now I only ride motorcycles. I never realized that it’s because other than the gear shift (which on a motorcycle with no reverse gear, are all really subselections of the same mode), there are no modes!

The gas/break pedal is a good example of UI that matches the physical system, not the task and intentions of the user. The gas pedal adds more gas the break pedal applies pressure to the breaks. The fact that you can press both while the car is in park or in neutral shows the problem with the design.

There are many ways to solve this modal problem. Here are a couple:
- Use the model of an airplane, instead of gas/break/reverse you just have thrust forward and back. Reverse thrust while moving forward is like applying the break. If you keep reverse thrusting you’ll eventually start accelerating backward.
- Another approach is to have the gas/break always apply to the direction you’re looking. Then make the seat pivot 180 degrees when you want to back up. This is the dump–truck approach.
- A third way would be to simply remove the ability to go in reverse. This would cause other problems but it would be a car without a reverse mode.

A simple solution, as suggested by a friend, would be to simply have a separate reverse-acceleration pedal. Place the pedals far enough apart so they can’t be confused (perhaps putting reverse on the other side of the brake pedal).

Note: this all assumes that only one pedal is pressed at a time, just as we assume people use just one foot in an automatic-transmission car to operate either the gas or brake, but not both.

Computer steered cars where you just input the destination.

“The fact that you can press both while the car is in park or in neutral shows the problem with the design.”

No, it shows that the interface doesn’t impede uses the designer didn’t foresee — such as using the gas to keep engine revs high when jump-starting another car, or flashing the brake lights to alert someone approaching you while stopped.

What’s more, the extra-pedal solutions are both dangerous and impractical: the space in the driver’s footrest is at a premium, especially on manual cars, and contradict the very good reason there’s a mode for reverse: to be able to suddenly apply reverse to the wheels while in forward motion would firstly destroy the transmission, and secondly put the car wildly out of control.

(If you’re going to say ‘lock the pedal so it can’t be applied while going forward’, you’ll have to justify the extra expense of the mechanism *and* say why a pedal useless for 95% of the driving time deserves space in the footwell)

The other reason there isn’t a good non-modal solution to this problem is because apart from the direction of travel, all components of the task are identical. The car must be accelerated, the brakes must be applied, and it must be steered.

A well-proven interface for all three things already exists; to switch to another for the already tricky problem of moving backwards would decrease safety and increase mechanical complexity.

@ Joseph Huang

Computer controlled cars pretty much solves all the problems doesn’t it, even drunk driving! Let’s hope they doesn’t “crash”.

grammar correction: don’t

The basic idea is to return the vehicle to neutral whenever it is stopped. Primary issue with this approach is that the automatic car becomes a little more like a manual. So, there would be an immediate marketing challenge. However with good positioning (perhaps a switch on the steering wheel), the interface may become trivial and marketable on safety reasons.

So, new part of the vehicle interface: A forward/ neutral/ reverse switch. It is held in position until the car is moving in a direction and always returns to neutral when the vehicle stops.

Swivel the seat around. It locks in forward or backward direction. Acceleration only occurs in the direction of the locked seat. Perhaps a control panel can popout of the arm rest as you face backwards and provide a simple navigation device like wheel/ joystick to steer and accelerate.

My solution:

Place a mechanism on the back of the headrest of the passenger seat (or some other HF agreed about location thereabouts) that won’t engage reverse until this mechanism is activated (and continually held).

This will force the user to look in the correct direction as they are backing up.

This would be in conjuction with the gear (so your kid can’t just through you in reverse on the freeway.

Downside- requires reverse with one hand on the wheel… common for most, but not everyone.

on second thought I could see a “lock” mechanism to keep the mechanism permamently engaged (thereby, your car operates normally as it does today- gear shift only) but doing so would cause a chime to engage or some equally undesirable trait that would be something the user would only tolerate for a short amount of time (while trying to get out of a mud/snow bank or something)

keep the steering wheel. but change the gas peddle to swivel in forward and reverse directions (like a teter totter). Tilt it forward to go forward. reverse to go reverse. have two gears (park and Go) keep the break peddle too.

I had a Toro riding lawnmower like this once and it was great. It also used rear wheel steering.

Teter totter, yokes and throttles… I have a couple questions about this-

Would this be in conjuction with a brake pedal?

I’m wondering about the “sudden reverse” that could spring up with a teeter.. if you’re going 70MPH and then suddendly your leg is bumped or you otherwise dig your heel into to the “reverse” mode, would you basically lock up and drop your transmission etc.. or would it gradually slow down until you get to stop… then go into reverse? If so, how would you lock up for a deer in the road?

Many of these comments are just different ways of switching modality between forward and reverse.

A single modality could be achieved with two joy-sticks. One to point in the direction of desired movement, the other to point to the direction of the desired orientation of the vehicle.

One solution could possibly keep its focus on the gear shift stick. Move the reverse to the top of the gear shift column (for an automatic) so that it appears as R P N D 3 2 1 (alternately: (for a manual) the order would roughly be R N 1 2 3 4…). To set the car in reverse push the gear stick forward, let the car move backwards (and press the accelerator if you prefer moving backwards fast), let go of the stick when done. The stick would move into park, P (alternately: the stick would move to Drive, D, if that is the user’s intention after everytime Reverse, R, is used).

This would keep the number of operations the same. eg For the current system, under normal operations, the user’s actions would follow

• 1.Park/Neutral to Reverse
• 2. Reverse to Drive.

For my proposed system, under normal operations, the user’s actions would follow as

• 1.Park/Neutral to Reverse
• 2. Park/Neutral to Drive.

This uses a kinesthetic way of keeping the user’s attention on moving the car backwards– a quazimode.

I just thought of an iteration of my design. I saw two problems: natural mapping and ergonomics. The shift column R P N D 1 2 3 could be inverted to 3 2 1 D N P R which solves the mapping problem– the action corresponds with the car’s movement (which doesn’t nessicarly mean the engine’s function).

This might also solve the ergonomic issue. With the non-inverted shift stick system, the user might need to lean forward some or even need to stretch in order to push the stick forward (the inverted mapping would, of course, require pulling backward). Another solution would be to move the mechanism further back, closer to the user, and raise it some. The latter solution may be needed regardless of the mapping anyway.

Park/Neutral should just be Park.

My Honda Accord with automatic transmission is organized: Park | Neutral | Reverse | Drive4 | Drive3 | 2nd | 1st. The shifter is placed where a manual shifter would be.

This works well for my two most common uses: ONE: I Get in my car and start. Move *down* two to reverse to back out. Move *down* one to drive off. TWO: I need to do a three-point turn: Reverse & Drive3 are right next to each other making it easy to switch between the two; *up* to reverse, *down* to drive, and repeat.

The organization is optimized for the most common use cases. It’s funny how sometimes by solving one problem you create another problem.

If Park/Neutral is just Park, there isn’t a Neutral, this feature can not be eliminated.

I’m curious how heavy vehicles controls are arranged. They typically just have that loud beeping noise, when in reverse, to tell the driver and everyone around their mode.

Or is it: Park | *Reverse* | *Neutral* | Drive4 | Drive3 | 2nd | 1st ? hmmm

I just remember driving a lawn mower that solved this problem. They put two ‘gas’ pedals; a smaller pedal for reverse, and larger one for forward, and of course the brake pedal. This was very handy for lawn mowing because it requires a lot of direction changing. The two pedals required different kinds of motions, so as to help not confusing the two. Also, don’t golf carts work like this too?

Also tractors often will have two brake pedals; left for left tire, right for right tire; and the two can be locked together.

Are we assuming internal combustion engine?

A lot of things change when you start using electric motors…

My rationale for changing Park/Neutral to Park is to simplify my explanation. Neutral is still there, but it has less to do with what I am explaining. The gear is no longer set to neuteral or park after the user has let go of the shift stick; it is instead set park. I guess Neutral can be used instead of Park.

I didn’t explain myself in regard to Park/Neutral. Thanks for having brought up the problem and the problems that I haven’t discussed.

An r/c car has only one mode I believe. Trigger forward–it goes forward, pull back and it brakes and reverses. Turn the wheel and it turns. Pretty sure this has been said before in different words, but theres 3 cents of mine. It seems that only cars, trucks, and boats are bimodal. Motorcycles and airplanes are unimodal only because they dont have a reverse. So of course a car without reverse would have only one mode–but that isn’t a solution. Take out the engine and it would only have one mode too–stopped.

The Warthog in Halo 2 is has only one mode. Left joystick controls acceleration(forward and reverse), right joystick controls steering. All we need to do is put a PlayStation controller in place of the steering wheel in our cars.

The drive/reverse mode is a good design because it saves the transmission from the trauma of a possible high-speed change in direction which would also be dangerous, as a previous poster has noted. Changing into reverse mode also prepares the driver for a change in driving style: reversing is a different skill than driving forwards.

The problem with the drive/reverse mode is that drive mode is used far more often than reverse mode. As a result, drivers reasonably expect their cars to move forwards. Accidents happen when they forget they are in reverse mode and resort to their expectations of drive mode, which is forward motion.

Cars have been around for a long time. Drivers are used to the steering wheel, pedals and gear levers, etc. Messing with so much accumulated knowledge probably isn?t the right way to go.

My solution introduces a risk-free change (as far as I can tell) to the familiar controls:

When the driver puts the car into reverse, the car starts counting. If, after X seconds have elapsed and the car hasn?t moved backwards, the car puts itself into neutral.

This has the following benefits:

1. If you?re distracted and you forget you?re in reverse, the car simply won?t move. You won?t move forward or backwards so you won?t hit anything in back or in front.

2. If you?re waiting for an obstruction to clear, for example a pedestrian, and you use up the X second delay and the car puts itself into neutral, the effort of putting the car back into reverse is minimal.

In both these cases, the driver gets two sources of good feedback: a revving engine and a car that doesn?t move. The driver then reverts to why-isn?t-the-car-moving behaviour, diagnoses that the car is in neutral, and puts the car into reverse again.

Should the time-delayed-return-to-neutral action apply to drive mode? It could do, but it probably shouldn?t for the following reasons:

1. Drive mode is the most frequently used mode so forward is the expected direction of movement; it shouldn?t be surprising that the car moves forward.

2. Cars that put themselves into neutral from drive would irritate drivers waiting at traffic lights, etc. and no device should irritate its user—especially on the roads.

How long should the car wait before putting itself into neutral? Testing with drivers should give us a good answer.

Reverse should be a quasi-mode, while accelerating and breaking should stay as actions (effected using the pedals) as they are now. The reverse quasi-mode should be effected by pulling the direction lever backwards. The lever should be conveniently positioned to make it ergonomic to steer at the same time.

A separate Reverse pedal may be even better, but there may not be enough space for it.

@Jeffrey Morgan: There is no problem with gear trauma. In case of a sudden change of direction, such as pulling the reverse lever while going forward, the transmission would disengage and make a warning sound, as it does today.

Messing with poor design is the only way to improve it. The suggested new design will be very easy to learn, and expose the awkwardness of the old design.

How about the other modes on modern cars? So should they be eliminated, or are they OK?

On central locking systems, the driver’s door can only be locked using a key when it is open. On automatic transmissions, you have to press the brake pedal to start the engine, and park the transmission to pull out the key. On manual transmissions, you have to press the clutch to start the engine.

Clearly these states are modes, because my attention is elsewhere.

They annoy me immensely when I drive an automatic transmission car with central locking. I am habituated to stop the engine before parking my manual transmission into gear (which is safer), then pull out the key, get out, put the key in my pocket, and lock the door without a key (which is much faster than using a key). I never lock the door without checking that I have the key.

This is a two foot gestural interface. There are two pedals; one pedal for reverse, one pedal for forward. If you press both together it acts as a brake.

I’m in strong agreeance with Eddie. The best way to stop people from going forwards when they want to go backwards is to have an extra button that must be held down to go backwards. This button could be on the wheel.

There are a few different ways that this would be used.

1. The car is in drive (forward gear) and the user accelerates, they go forward.

2. The car is in drive, stationary, and the user accelerates, but has the “reverse” button down. The car moves no-where but activates a warning light and sound.

3. The car is already moving forward, the user accelerates and has the “reverse” button depressed. The car goes forward. Here the user knows that they are in a forward gear because they are already travelling in that direction. No mode errors (hopefully?).

4. The car is stationary and in reverse gear. The “reverse” button is not depressed. A warning light and tone is heard. The car goes nowhere.

5. The car is stationary and in reverse gear. The “reverse” button is pressed. The car goes backwards.

If the user lets go of the reverse button while reversing the car keeps going, unless it is travelling below a certain speed. It keeps going so it doesn’t try and suddenlt stop (for no apparent reason). It needs to be depressed when below a ceratin speed, to make sure the user knows that they are still travelling backwards. If you are reversing at 0.5KM/H you may not have the feedback of going backwards.

I think this nicely solves your problems. Sorry I didn’t get a chance to read your article before the 1st. I’ll make sure I get onto the next one quickly.

PS: Just finished The Humane Interface and found it very interesting and useful. Thank you. I’ve started to implement a version of the inline CALCULATE command in JS for text fields and it works quite well.

The “reverse button on the wheel” solution does seem to be the best way to turn the reverse mode into a quasi-mode, but there is one glaring problem.

If this button is on the wheel (rather than on a stick shifter behind the wheel seen on some automatics, or some other location) then you come across the difficulty of holding the button while turning the wheel around.

Maybe you would only need to hold down the button while stopped, as it was pointed out that while the car is in motion you don’t usually mistake what mode you are in.

Another key point is that you would have to get rid of shifting into reverse, in the conventional method, for this to really count as quasi-modal. Basically that when going from a stopped position, holding that switch is what actually makes the engine accelerate in the reverse direction.

All in all, this is very fun stuff to think about.

Tarwin and Henry-

I’d like to adjust my thoughts a bit based on Tarwin’s comments- I too agree that once you’re “in motion” it should be safe to release the mechanism since you’re already travelling and aware of what mode your in. I like Tarwin’s 1-4 steps.

But like Henry says,the steering wheel is probably not the best place for this control. I think that it should be right next to the gear shift.. You move the gear selecter into reverse, then pull and hold a switch to get the car in motion… then you can release. I sitll think you should be able to “lock” this switch into place, but it still should have a flashing light or annoying sound… or at the very minimum reset it selft each time the car is started.

…I’m thinking about the frustrated (poor) parallel parker that is constantly moving from forward to reverse. The driver is likely frustrated already with the act of parking, let’s not compound the problem with a hindering UI.

A Solution? An extra pedal for reverse. It’s a pedal, or rather a button, off to the way left side that you have to push and hold with your left foot, and then use your right foot to push the gas pedal as regular. And there is still the standard brake pedal.

Eddie, Henry, Tarwin, Anndra:

The designs you propose are modal, since you still have to “move the gear selector into reverse”, and pressing the reverse button does two different things depending on the state of the “selector”.

So this is not a solution to the problem of designing a car that is not forward-reverse modal.

Jeffrey Morgan: Your design is modal too, since I have to check (or count) to know which way the car will go if I press the pedal.

Pgan:

In my design, the car will either reverse if the driver accelerates after putting the car in reverse, or the car will not move at all if the car times out the reverse and puts itself into neutral.

Yes, my design is modal. It is a refinement of an existing modal design. My suggestion addresses the problem some drivers have of forgetting they have put the car in reverse and hitting something behind when they expected the car to move forwards.

The original question asked if a non-modal design was possible. Yes, of course it is—as some of the suggestions on this blog demonstrate.

However, the real issue is whether such non-modal designs would be easy for existing drivers to get used to and use and, more importantly, would they be safe. That is less clear.

Hello All,

I want to apologize for taking so long in answering the puzzler. Enso is just a stone-throw away from being released and everyone at Humanized has been busy enough that we have to remind each other to eat (and remind each other to remind each other).

The discussion here has been great and has far exceeded my hopes for. Keep up the great comments!

I agree that if there is a solution, it’s a quasi-mode. Many of the suggestions here are extremely dangerous from a practical standpoint, though they may be very clever in theory. The ones that stand out are the airplane/joystick model and the tilting gas pedal model.

One of the reasons the accelerator in a car is at the driver’s feet is that it’s by far the safest place for it. If you’ve ever driven with a small dog and it jumped down under your legs while you were moving at 60 mph, just remember the sudden panic you felt. At that moment you realize that little alcove is sacred. Bringing that space further up into the cabin allows for a whole new array of potential disasters. When the difference between 10 mph and 100 mph is a slight amount of pressure, you simply can’t allow anything near that accelerator but the body part you use to control it.

A pilot can get away with this partly because he doesn’t have other planes immediately above and below him. He has a huge margin of error that a car simply doesn’t. And the wheel doesn’t trigger the engine to accelerate or decelerate anyway.

Braking also becomes extremely problematic. When someone runs out in front of your car, you want to STOP. You don’t want to worry about accelerating into the car behind you after you stop. You don’t have time to.

That’s all really just the beginning. It works great in video games, but going over every real world concern, I think we’d all agree that this model couldn’t work.

There is the possibility of having the gas and brakes control movement while the steering wheel controls direction, but I wouldn’t personally want to spend a long car trip pushing the wheel in the whole way. I’d start asking, “Why can’t this be done automatically?” which is exactly what’s achieved today by putting the car in “D”. Make it so the wheel only pulls outward for “R” and now you’re in quasi-mode territory (and there are simpler ways to accomplish it).

The tilting gas pedal model has a similar problem. Unless you’re a woman used to wearing high heels, pushing the pedal down at a forward angle would be agony on a long trip. Currently the weight of your leg does all the work. Anything that doesn’t allow the driver to relax for most of the trip decreases enjoyment and increases fatigue and risk.

In situations that require split-second decisions, possibly in a panic, I don’t think making the pedal more complicated than “less” and “more” is a safe way to go.

Holding something down for “R” in a quasi-modal way is excellent for feedback, but it shouldn’t require a hand (try parallel parking one-handed), a foot (in a world where a clutch still exists), or your gaze. So it seems to me so far that “D” and “R” need to be lockable for safety reasons (let alone mechanical), and as it is now, they are.

The more I think about it the more I think the car designers have made the necessary sacrifices (which always seem to exist) in the proper places.

Incidentally, I remember reading somewhere (I wish I could track it down now) that in the very early days, car makers tried to replace the steering wheel with a joystick because in accidents people were being impaled on the steering column. But the new interface was largely rejected — it just didn’t feel right. Part of driving a car competently is the feeling that it’s merely an extension of your own body, and I suspect for some subtle psychological reason the steering wheel is here to stay.

One mode = Direction
Sub mode = Neutral
Sub mode = Detect Motion

With transmission in neutral and motion dector active ( mounted on left of driver’s seat); Driver leans forward to activate “forward” and leans backward to activate “reverse”.

Change of direction can only take place in the neutral sub mode..

Audible alarm when reverse sub mode is chosen.

The prototype would be round, as in circular , and have three wheels positioned at the points of an equilateral triangle. The wheels would be able turn a full 360 degrees and would be locked to turn together. Steering could still be done with a steering wheel and the cockpit would turn to orient the driver (and passengers if you can find any willing to ride in the contraption) to face in the direction aligned with the wheels. The wheels would need only revolve in one direction and the vehicle would be capable of motion in any direction.

I think this idea has been solved…or well on it’s way.

Design a car that can only go forward (and turn left or right, of course). But make it so that the driver rotates, and ‘forward’ is whatever direction the driver is currently facing. For the time being this may require forcing the driver to switch between facing-forward and facing-backward. Once facing backwards, the acceleration pedal will cause the car to go in reverse, but the driver will be facing that way, so there won’t be any surprises to the driver. While the car is in motion the driver seat will be locked in place, and can only be swiveled while stopped. And parents would see the benefit when traveling with children as the front passenger seat could also be made to swivel, thus helping us tend to screaming children during road trips.

In the most reductive of terms, that pedal makes the car go. The other pedal makes the car stop.

When driving, we’re almost always moving forward. So let’s assume that’s the basic state caused by accelerating: “go.”

I like the easy fix employed by your average garbage truck: loud beeping noises that tell others that the truck is about to move backwards. Guess what–that noise also signals the driver that she or he has moved in reverse. This is, alas, only a preventative solution.

A better option is the voice command. The vehicle simply won’t go in reverse (heck, let’s throw in turning right, left, and accelerating from a full stop) without the driver saying “back up” “turn left” “turn right” and “go.”

Either way, it’s often easier to redesign the interface than the mechanics that live below it.

Almost every idea posted fails the practicality test, which unfortunately for all academics everywhere, must take into consideration what people currently do. You won’t retrain millions.

Andrew Clark’s analysis is spot on.

The fellow with the audible alarm has a workable idea - and it has already been implemented, commercial vehicles (at least in Canada) require an audible backup alarm to warn others (not just the driver) that the vehicle is backing up.

Modes are sometimes the right thing to have.

Pgan:

You pointed out that the quasi-modal reverse button method would indeed still be modal if you kept the conventional Reverse setting on the gearshift.

This is certainly true, and I did point out that for the reverse button to be a truly quasi-modal solution, it would mean getting rid of the Reverse option on the gear selector.

I suggest that while the automobile is stopped and the button is held, the car will accelerate in reverse. While in motion, it (might) no longer be necessary to hold the reverse button. This really works to the advantage in parallel parking situations where you go from forward to backward, always stopping when you are going to switch. There is still the question of where to place this button, but if it isn’t necessary to hold it while in motion, it would allow reversing with two hands on the wheel.

Unless I miss my guess, most of the comments here are related to movement and vehicular motion: forward and backward and left/right/neutral/stopped considerations. Also, most of us think in terms of the cars and devices we have had experience with: from airplanes, and boats, and motor vehicles and motorcycles and rc ‘toys’ - everything mechanical we have ever experienced has contributed to our thinking about this problem: an alternative to forward/reverse modality. Forget the problem, for the moment, of training the world to use a new device. Forget (again, for the moment) the need to train mechanics and repairmen who must support this revolutionary device. Focus on the need, and determine if modality (forward/reverse) is essential to fulfill that need.
The need is to enter a vehicle, for the purpose of transportation, from point A to point B. Some basics here: transportation means one or more people, plus various finite sized objects (clothes, luggage, tools, food, etc.), and point A would be any relative distance from point B making transportation in this vehicle a desired option.
Vehicle power is worthy of consideration, too, because battery power presents certain limitations while combustion engines present other requirements (and limitations). (The use of steam is not totally discarded, but for the moment we will ignore H2O per se.)
Thinking about modality only, in terms of forward/reverse mechanisms, I believe the answer is simply that we must think about vehicle design in a new way. Because a vehicle, as we know it, is operated to move in a forward/reverse direction it is hard to think of any other way to migrate from point A to point B.
The human body is designed to move forward skillfully, rapidly, and safely. The ability to move backwards is awkward, limiting, and hazardous. Sometimes (rarely) humans move backwards ? but only rarely, and seldom if ever, is this mode of human travel desirable. So what do we do? We turn. Even if on a dime, it is far simpler to think about moving forward in the direction from which we just came than it is to move backwards, however short a distance, in that same direction.
I submit that from a modal standpoint it is far easier to think about the ease of moving always in a forward direction than it is to deal with the problem of physical limitations in proceeding through a 180 degree turn. I believe the Reverse mode problem stems from the mechanical limitations of the vehicles we use. Note the absence of the Reverse mode on aircraft and motorcycles as testimony to the benefits of this simplistic answer.
Before someone complains that I have stated the obvious, and ignored the practicality of accepting the last century’s vehicular design considerations, let me hasten to make one more point: although it is easier to THINK about the simplicity of a single direction (Forward) mode, it is impractical to do so for another reason: physics doesn’t permit a vehicle to turn on a dime at any speed, and (certainly for boats) forward momentum must be negated before a safe turn in any direction can be made. Taking into consideration just whether or not we can abandon the Forward/Reverse modality based on the above arguments, I submit that until we have vehicles that can read our minds, shift (safely) from one direction to another without compromising life and limb, we must ACCEPT that dual modality (Forward/Reverse) is immutable.

Paddle shifters. They only come in expensive sports cars, but I’d like them in my car too.

So, we modify the idea just slightly, so that instead of just toggling the shifters to go up or down the gears, we have to perform a slightly abnormal sequence to get reverse, such as both paddles clicked toward you twice then away from you once.

This really shouldn’t be that hard…I learned to drive on a VW bug where you had to push the stick shift down toward the floor before it would go back into reverse. Definitely slightly abnormal behavior but quick and not difficult, uh, like, uh, you know, enso. ;).

Sorry for the above wordy dialogue. This is a shortened restatement:
The problem: an alternative to forward/reverse modality.
.

The need is to move a vehicle, for the purpose of transportation, from point A to point B.

• Transportation: one or more people, plus various finite sized objects (clothes, luggage, tools, food, etc.),
• Points A / B: any relative distance apart , making transportation between the two a desired option.

Thinking about modality only, in terms of forward/reverse mechanisms, look at this modality in human terms. The human body is designed to move forward skillfully, rapidly, and safely. The ability to move backwards is awkward, limiting, and hazardous. Sometimes (rarely) humans move backwards ? but only rarely, and seldom is this mode of human travel desirable. Instead, we turn. Even if turning on a dime, it is far simpler as humans to direct our energy into forward motion than it is to move backwards.
Enter the mechanical issues of forcing a man-made vehicle through a 180 degree turn. The Reverse mode problem stems from the mechanical limitations of our vehicles. The absence of a Reverse mode on aircraft and motorcycles is testimony to not having a need for this mode in every transportation situation. We cannot omit designing for the Reverse mode because physics doesn’t permit a vehicle to turn on a dime at any speed, and (certainly for boats) forward momentum must be negated before a safe turn in any direction can be made.
I submit that dual modality (Forward/Reverse) must be inherent in vehicular design.

Get a horse. Nuf said.

Sorry to come into this so late (after 1 Jan). My car has a “gear shift” that looks like a little joy stick on the dash (a knob sticking up about 2″). It “lives” in a neutral position. You can push it to “D” or pull it to “R” (I’m not 100% certain of the designations, here) — when you release it, it pops back to the neutral position. Every time you put it into “R”, a little “beep, beep, beep” starts to sound to remind you that you are backing up.

So, if I have the terminology right, this is definitely a “modal” solution, but it has been reduced to the two possible “modes”, going forward (where you want to look at what’s in front) and going backwards (where you need to look behind you). Not only to you make a kinesthetically different movement, you get auditory feedback for the more “unusual” motion (R). And this is in an existing car, with an “ordinary” steering wheel, brakes, gas pedal, etc.

[This is part 2 of my post. I tried to say this “all at once”, but when I hit “Post”, got an error message saying I tried to submit too much. Sorry.]

If you are curious about where the other “gear” positions are located, there’s a big button marked “P” that you push to engage “Park” — I haven’t tried this out, but I suspect that you need to be stopped in order for this to work. There is no “sub-gearing” on this car — the transmission takes care of itself, leaving only “D” and “R” as choices. Well, there is also an “N”, but that’s only there when you first turn the car on (by pushing the Power button, of course), sort of the “logical default choice” — I don’t think you can put it back in “N” once you commit to “D” or “R”.

Here’s a dumb idea… it doesn’t remove modality but it might remove accidents:

If you’re in neutral and you hit the gas pedal while looking backwards, you go backwards. If you were looking forward, then you go forward. Once you’re in motion, the direction you look has no effect.

Bob Schor: it’s not forward-backward modal.

If one really wants quasi-mode; Make the gear box have a function I found in toys…

Something like; the gears change normally upto N; but beyond that it has a spring’d mechanism that requires you to push it back as long as you want the car to run in reverse. Once you’re done; it jumps back to normal.

Interesting stuff. I have not thoroughly read through all comments, so pardon me if I am repeating anything.

The first thought I had was that there are three important aspects to consider when determining the feasibility or possibility of such a concept:

1. the interface between the driver and the vehicle
2. the mechanics
3. the psychological impact to all drivers

The way cars drive now do more than limit the ways in which we can move. It also easily helps to create a level of comfort based around the predicted movements of other drivers.

The reason I am comfortable cruising down the interstate at 80mph is not because of how well my car handles; instead, it is because I like to think I can predict if/when other drivers around me plan to change direction.

Making vehicles more versatile in this case may in fact make them seem more limiting.

This would also apply to vehicles moving 5mph through a parking lot. This last weekend my wife and I were pushing our son in a stroller at the zoo, and during a nice weekend, it was busy! We hardly even think about walking next to moving cars because everyone just knows that cars move forward and they pivot around the front wheels.

So for people aiming to invent a vehicle that moves around like a crazed killer in an FPS, think about that :) People drive cars forward, and moving in reverse is really not used all that much.

Requiring that a car be stopped and that the clutch or break be depressed pretty much means that as long as the break-lights work, a passerby will know when a car is about to begin moving backwards. Switching modes in this case forces a mental model on the driver that can provide an additional couple of seconds of awareness to person or car that may be directly in the way. Other things that may give a direction change away could be exhaust coming from the tailpipe or seeing people in the car if the car were forward-facing; however, one of those factors will be gone with electric or more energy-efficient motors.

Perhaps we should invent a way where people don’t have to drive and can walk everywhere they need to go? haha. We invented laptops didn’t we? Let’s try to create a better interface for working from home :)

Paul Ingemi: neat idea about reversing automatically when you’re looking backwards from a stop.

What if someone were to turn around and reach towards the back of their car for something? In the process, they accidentally hit the gas pedal.

Perhaps it’s not likely that someone would do that, but it is less likely that the he/she would be in a position to correct the mistake if they did.

By requiring a person have the car in “reverse” you add an extra level of security.

Also, the reverse mode in this case shifts the responsibility from the car to the person. To make the car responsible and just as safe and predictable means taking all human factors into account.

Perhaps we have a more general “drive” mode?

Something that would be helpful regardless of the existence of a “mode” would be to have the seat of the car sense a change in weight. If the person were moving around while the car was in a standstill position, the car wouldn’t move.

This could be dangerous if the car were moving, however. Even if the weight system was working as expected, let’s say the driver decided to change spots with the passenger while on the highway. Okay…that is dangerous in itself, but perhaps the people are skilled and pull it off. The point is that car’s reaction itself should be predictable and not be the cause of an accident.

It is arguable that the knowledge of such a system might prevent a driver/passenger from doing something so stupid, but you cannot rely on a person knowing a “feature” like that exists. That is another issue.