I'm Aza Raskin @aza. I make shiny things. I simplify.

I'm VP at Jawbone, focusing on health.


The Seduction of Simple: Hidden Complexity

Simple often isn’t. Spacious interfaces with few controls, artfully placed, may look comforting and inviting (and they often are), but they can also front for a mafia underground of hidden interaction complexity.

I recently heard a respected designer (who shall remain nameless) speaking of what he called an exemplar of simplicity in interface design: the garage door opener. That’s the seduction of simple in action, because it isn’t simple at all. We all make numerous errors using the garage door opener — moving the door in the wrong direction first, pausing the door accidentally, or hitting the button too many times after the door doesn’t respond quickly enough. It’s actually a resoundingly bad interface masquerading behind the innocence face of a single, simple button.

Why’s it so bad? Because the single button is a grunt-and-click interface that tries to hide a more complex set of variables: the door’s velocity and position. The door has six states: Closed, opened, closing, opening, and two types of paused (I’ll come back to this). The button has only one action to control all of these. To know what pressing the button will do, you have to keep track of the state of the system:

When your mental model doesn’t match the actual state of the system, a mode error occurs.

Mode errors will occur frequently in particular because of the two types of pauses: In one the door goes up when you press the button, and the other it goes down. There’s no way to know which state the system is in without remembering what’s come before. We know that this will confuse and befuddle users.

The seduction of simple is an easy trap to fall into as a designer: it gives the appearance of ease-of-use without delivering on the promise. It’s a crutch we can’t allow ourselves. Me most exorcise the devil of false simplicity from our interfaces.

Question: What’s a better garage-door opener interface?

RT @aza The Seduction of Simple: Hidden Complexity | Follow @aza on Twitter | All blog posts

View all 184 comments

You’re right on the money. That reminds me I need to go back and re-read [Design of Everyday Things](http://www.amazon.com/Design-Everyday-Things-Donald-Norman/dp/0465067107/?tag=worfrohombl06-20). The section on errors is excellent.

I’m guessing you know Don Norman, don’t you?

I wonder why they didn’t design it like a light switch?

Switch it up for open, down for close.

I’m not sure a pause state is useful for a garage door, so I’d say a two-position switch for a wired controller and a two-button wireless controller.

The wired controller would have a switch indicating whether the user wants the door to eventually be open or closed.

The wireless controller would have two buttons allowing the user to command the garage door to begin opening or closing the door.

If feedback said the pause state was useful, perhaps a three-position switch and a three-button wireless controller would be a possibility.

A two-way sliding button will be best (to me at least) – push and hold it in one direction – the door opens, the other direction – the door closes. Release it – the door will stop. Much like the control of the electric window.


    Having to press and hold the button on the remote requires constant attention and use of one of the driver’s hands. This makes it is less safe to drive at the same time.

    One of the advantages of having a button that can be pressed once (opposed to pressing and holding) is that the driver can simply ‘tell’ the door to open and meanwhile drive into and up the driveway, focussing their attention and using both hands on the wheel, and by the time they’ve arrived at the garage the door is open. This is less possible and less safe when the user’s hands and attention is focussed on pressing and holding a button.


      The car window button would be an ideal control if implemented correctly.

      Here’s a photo of one from an Audi A3 which I find ideal: http://bit.ly/gG5vi3

      * Recessed 3 Position toggle, push DOWN to close door, pull UP to open door. This addresses the mapping issue.

      * When the door is stationary, Push/Pull quickly and release to start door moving in the indicated direction.

      * Click EITHER Up or Down while the door is moving to “pause” it. To resume simply click the desired direction again.

      * Push/Pull and HOLD and the door begins to move, it continues until it’s reached the end of its travel or you RELEASE your finger at which time the door pauses.

      This addresses the unknown next action because you’ll always directly indicate your intent. It also handles the emergency “bash the button” response since either action results in a pause and you can then immediately reverse the direction if desired with a second click.

      It will work both right side up or upside down without looking since it’s a push/pull toggle rather than up/down.

      Finally, it does not require you to hold your finger on the button for the complete travel of the door, yet gives you the fine control to stop it at any point in its travel if you choose.

      Bonus – it’s simple and cheap. It is a two-pole switch instead of a single, but no LEDs are required nor is two-way communication between the door and the remote.

    This article is novel, I really love it. I also have some ideas written in my article, you can have a look if you have the interest. We can also discuss different topics.
    travel guide

It may *seem*, upon reflection from a distance, that the UI for garage door openers needs improvement.

But I doubt it’s so obvious. For one thing, it seems to me that the core function *has* to be single click. When people approach the door, they need to click a button to open it. No holding a button, no trying to figure out which button to press.

The most typical scenarios are (a) Door us up and user wants it down, and (b) Door is down and user wants it up. It seems likely that the additional modes you describe are so incredibly rare that the 80/20 rule applies and they can be relegated to a “let them muddle through” user scenario in order to permit elegance and ease for the core scenarios.

It’s helpful to consider the potential damage from the mental model mismatch you describe, where a user may panic in confusion. It seems that the danger is minimal — the worst that will happen is that you will either (a) stop the door when you intended to reverse the direction, or (b) unpause the door and cause it to go in the opposite direction you thought it would go. (a) is utterly harmless, and in fact is quite elegant and safety-centered solution. And yes, (b) can cause harm in an unexpected lowering (e.g., you thought you’d raise the paused door but in fact you lower it and crush your dog), but is easily rectified by simply clicking again in a panic to pause, and then lift the door. An unexpected raising is not dangerous.

I don’t normally advocate throwing design ideas right to testing to resolve them but this one seems like one of those examples where we just won’t know until we try, and the additional functions you describe seem to risk adding more baggage then is worth removing.

If you have a properly designed interface, I’d argue that there are actually only three states: moving up, moving down and not moving.

The interface for this would like a three position light switch:


A three-position switch, and two LED’s. Two of the switch’s positions indicate “open” and “closed.” One LED indicates whether the door is opening, one indicates whether it is closing. The third switch position is “paused,” which is indicated by both lights lit at once. Total: three modes, 75% information efficiency (led states used).

I don’t like the following simplification as much, but it is simpler: a three-position switch and one LED. The switch must be pushed and held up/down, else it returns to the middle state by itself. The LED is lit when the door is moving in the direction the user indicates. When the door reaches its extreme, the light turns off. If the user lets go of the button before then, the door pauses and the light blinks rapidly. Total: one mode, two semimodes, qualitatively higher information efficiency (hard to quantify).


2 buttons: “open” and “close”. If you hit one of them while opening/closing it pauses.


I forgot to mention, the reason I don’t like semimodes in a door opener is that it ties up a hand which should be used for driving. This seems like a dealbreaker for semimodal interactions in this case. Does anyone have a counterexample?

The actions that we want to take on the garage door are:

– Open the door
– Close the door
– Pause the door in case of emergency (ie: it’s closing on your cat)

The user shouldn’t have to look at the control when doing this. It needs to be muscle memory.

Once the garage door is paused, it’s basically the same as whether it’s opened or closed. You are going to want to open it all the way or close it all the way. There are some advanced garage door controls, such as pausing it just before it hits the bottom to avoid something that would cause it to “bounce”, but those are not common and tricky to get right with the current system.

The pause is very important – it’s usually done in a panic and not something you do often, so the user tends to just hit it. That’s the advantage of the current system – it stops the garage door so you can step back and think about the problem. The user isn’t going to be looking at the opener in case of emergency – it’s a pure button mash at that point.

I’d propose a rocker switch that auto-centers itself after you let go. Top and bottom should be designed so that top is closer to the user’s body whether upside down or not (this affects the visor clip design, etc).

Push the top part of the rocker momentarily to open from stopped. Push the bottom part of the rocker to close from stopped. Push either part of the button (in a panic) to pause the operation so you can think about how to fix an emergency. Pushing top or bottom in a paused state will move in the correct direction.

I’ll add a bonus feature: garage close confirmation. The garage door should tell the remote what state it’s in. I tend to close the garage door through pure habit and not remember whether I did or not. Having a light I could look at on the remote to tell me what state I left it in would be great.


    “Pause the door in case of emergency (ie: it’s closing on your cat)”

    What if you pause the door too late and it’s paused on crushing your cat to death?

    I would argue that in case of emergencies the best thing to do would be to open the door rather than pause it.

    Pausing is useful for other things though – maybe the maintenance man needs to pause the door half way to work on it, or it’s useful halfway to lean something against it, or let air flow into the garage while you’re working on something toxic, or holding something down (like a burglar)?

A single button that is displayed in different colors to indicate it’s state.

In fact, you only need 2 colors (and one is transparent or a lack of color and to save energy for the batteries).

Red = the door is moving.
Transparent = the door is not moving.

One could assume that the user is almost always viewing the garage door when operating the remote. If it’s closed and you are about the push the button, its pretty obvious which state you are about to enter (opening). If it’s open and you are about the push the button, you know that its about to enter the ‘closing’ state.

I guess you could add a third color to represent ‘paused’ to handle if someone clicks “pause” a moment before the door is completely opened or closed.

Re: semimodes–

A garage door opener needs to be “fire and forget”. If you live in a residential neighborhood with sidewalks, your attention must be on the process of reversing out of your garage. This means:

– both eyes on all available mirrors (or backup cameras)
– full mental attention on the above visual input
– both hands on the steering wheel

Ideally you should pull out of the garage carefully, stop, then trigger the door close. Now that the door is closing, you concentrate on not squishing the neighborhood kids as you back out on to the street.

Asking the driver to spend any additional attention on the garage door opener will likely result in more backup injuries.


A control with a miniature garage door on it; yank it to a position, and the door will go to that position in a safe speed. The garage door always mimics the state of the “mini” garage door on the control.

A single button with a single light.

Lit: moving. Since you were looking at the door when you initially pushed the button, you know wich of the moving states it is in.

Blinking: Pause. Same here; you either wanted to open the door only a little or you were protecting your pet from it. Either way, you know what direction will be next.


I think the design of garage door openers are perfect. I wouldn’t want it any other way.

When I come home from work, I’m not thinking of how I want the door to open, or to what height, I just want it to open.

Keep it simple.

I like Matt’s design of a rocker switch with auto-center. Plus the panic pause by just whacking it. Little LED lights to indicate all closed, or closing is nice too.

I think it’s great, plus pretty easy to construct.

A similar UI example is the power button on a rear projection TV or projector. One button, but always seems to screw me up.


What about voice-activation? Proximity detection? How about a mind-reading system? Yes, I think that would be simpler….

Or you could park in the driveway and forget the garage door!

Hey Aza, I think this is a better way:
Press once, it goes up press twice, it goes down.
Thats it.


    I wonder how many people would think of pressing it twice to make it close? Would and should this require a manual?


      I thought of this idea too, but I think it should also have a pause if pressed once during movement. This leaves only one pause state where you can press once to raise or twice to lower. As for the need of a manual, most people could probably figure it out, if not, you could just have the instructions imprinted on the back or bottom of it or just a sticker on the back since you only need 3 lines to explain it. You could even imprint them on the button itself.

If the pause functionality is only used to prevent accidents, perhaps sensors on the door could take care of that. In that case, a single button to start/switch direction may be enough.

Not experienced with automatic garage doors and their use cases, though.

How about a handle that you twist and pull on to open, and then you can push on the door to close it?

Technology can so often overcomplicates simple interactions. :-)


I like Gary & Boriss ‘s ideas. Direct manipulation, w/ direct correspondence to the object.

At the other extreme, the car should drive itself home, the garage door automatically open & close at the right time. The car door opens itself and you step from car to Segway. Coast through the kitchen, nabbing the hot cup of tea it has prepared already, and into the living room to watch videos already queued by Netflix based on watching habits.

Now that’s a garage door opener!


    Sounds awesome, the only problem is that my mind goes directly to Wall-E with the fat people because they never had to walk anywhere, but as soon as we figure out a way to eliminate the need for any kind of exercise this would be awesome.


Best things are kept simple as usual ;-)

    Of csoure, what a great site and informative posts, I will add backlink – bookmark this site? Regards,Reader.


Hypothesis: Nobody ever wants to stop the door while it’s going up. The only reason to stop the door is if it’s about to hit something, and it can’t hit anything going up. Thus, rather than change the controller, we can just get rid of the pause state.

Door is up, push button: goes down.
Door is down, push button: goes up.
Door is moving upward, push button: no effect.
Door is moving downward, push button: door reverses and goes back up.

Having said this, I can think of two scenarios where one might genuinely want to stop the door going up. There could be an object dangling out of the rafters of the garage that will be hit by the door, but if you’re opening the door from outside, you can’t see if there is any such object anyway, so the right way to deal with that is to have an obstruction sensor that stops the door and disables the regular controls until you reset it (like a circuit breaker). Or, your house could be surrounded by zombies when some idiot hits the opener, but if you’re in a zombie movie you have bigger problems than your garage door opener’s UI.

    I pause the garage door all the time when its going up so I can let the dog in and out when its cold outside. We have a heated garage/workroom.

    I think a fader like on a mixing board, where its weighted to intuitively go from each binary state, but can be overridden to any specific height is the best solution. Even better would be touch sensitive 3″ panel which you could operate with a flick of a finger like an iphone.

    84Most of the rveiews of this product tended to be positive overall with an emphasis on easy installation and quiet operation.21


Gary, you’re right…. that’s one of the benefits of having children. And they don’t need batteries in the remote.


It only needs 3 states: Open, Close and Pause
Opening, Closing and Pause’ can easily be made redundant provided the correct model is used.

Open -> Pause <- Close
Pause does nothing but Pause. Open and Close are both valid starting states.

See http://ff.im/LCEN.

By the way the play button is another great example of an interface with hidden complexity or better said wrong mental model. I like the way you dealt with the problem in Songza by making a button for both pause and play.

This post is a good example of simplicity.

I would say, three buttons: open, close, pause

@Zack is right. Pausing while going up is useless… That said, the fact that the behavior going up would different from the behavior going down (in Zack’s model) might not imprint as readily on the user’s subconscious understanding of the mental model as the simpler model that pressing a button while the door is moving *always* pauses it. The frustrating and useless function of pausing while going up trains the user to understand the general principle that pressing the button while the door is moving does the pause thing.

@Matt Mastracci is also right. “Fire and forget” needs to be the name of the game. Drivers should pay attention to driving, not door opening. Nobody wants to hold down a button because that makes opening the door more of a chore. It’s important that the user finds the UI pleasing to use in its simplicity and ease. Anything more than the one button function will simply not be as pleasant to use — making it effectively an inferior product to most people. You can’t discount this aspect of the UI design, either.

Time to play the devil’s advocate…

The best interface is one that you don’t realize is there. So we get rid of the remote altogether. Instead, lets move to automatic detection. That never works 100% right, so get rid of that too. Now you have to manually open the garage door – thats a pain. Its easier to just not have a garage door – so remove that. However, now the car might get stolen. Simplest way to avoid having your car stolen (ignoring any other factors) is to not have a car.

Ergo, you should just walk everywhere.


Definitely as Brad L said. A 3-position rocker switch nicely labelled from top-to-bottom: Up-Stop-Down. Perhaps with a bit of tension in it so you have to push harder to get it in to the 2 moving states than to pop the switch back to the middle stop position. I think this is the simplest and least ambiguous solution which still retains high usability.

Kenny Rosenberg

The problem with having 2 or more buttons or some kind of directional switch is that people shouldn’t have to think about which way they are holding the remote. Not everyone attaches their remote to their visor; some might just toss it in a cup holder or other compartment. For them, they want to be able to grab the remote and press a button. They shouldn’t have to think about which way they are holding it so that up is up and down is down.

It’s a little bit like when you’re holding your iPhone sideways and you can’t figure out which way makes the volume go up or down.

Maybe one button can have two operations, like on a digital camera, where pushing a little bit autofocuses, and pushing a lot takes a picture. Push a lot makes it go down, push a little makes it go up, either will pause. I can see problems with that too though…a lot people can’t figure out how that works on a camera.

No-one has yet suggested what seems to me to be obvious. Two buttons, one up and one down, and you hold the button to continue the movement – i.e. the interface is spring-loaded. Pause (in case of emergency) is then simple – drop the remote in shock. :-) If you are already at the top or the bottom, holding the button has no further effect (no grinding motors).

This is basically equivalent to the two-direction rocker switch solution but is spring-loaded.

The only interface error you could make is to approach the door and press and hold the wrong button. This is a harmless error (the door just doesn’t move), what to do is obvious – switch to the other button – and could be avoided by making the buttons different sizes, colours, textures, and mounting them in the car so open is up and close is down.

One disadvantage – you couldn’t drive away on leaving until the door had shut. Disappointing for the instant-gratification generation.

Robert O'Callahan

Why not just build a garage door that is permeable to my car only?

@Gerv: “One disadvantage – you couldn’t drive away on leaving until the door had shut. Disappointing for the instant-gratification generation.”

… and thus making the design utterly unmarketable.

And by “instant-gratification generation”, I assume you mean anyone under the age of 80? :-)


This discussion is quite.. illuminating.. I can certainly see why the majority of my software looks the way it does…

There were about 2 posts that actually included any consideration about how an average user would really use the product – ie, during the drive up to the door, press once to open the thing, park, press again to close the thing.

Sure, I would very much like us to solve the problem of presenting a good UI for when you need to do something more than that. Although, as you can imagine, replacing the 1-button system with a 3-button, 2-LED system is a little counter-productive.

Ultimately, you really need to start with a probability distribution –
1., There’s about 99.99% chance that our user wants to open the door because it’s fully closed, or close the door because it’s fully open.
2. There’s 0.008% chance she wants to stop the closing because it looks like someone might get hurt (we did build in nice safeguards, but we forgot to make them obvious).
3. There’s 0.001% chance the user wants to just play around with their fancy new door mechanism. No new design is required.
4. Here’s the remaining 0.001% for which you are actually designing. Now, do you really think it’s worth replacing the UI?


    Bingo, you, and a post by Christopher, are right on the money.


    I’m left wondering if most of the folks commenting here have garages/interact with an opener on a regular basis.

    What also gets missed is that, for safety reasons, the system is designed to help reinforce that it’s not reliable to just trust the button when you can’t see the door. A garage door is something that should be in sight when operated so you know the true state of the the system before you interact with it.

    Also – The response for a button press during the down movement on a garage is almost universally reverse – not pause (to avoid the crushed kitty with a paused door scenario – which the pressure switch should prevent anyways)

    I have to agree with the designer. For what it’s designed to do, it’s pretty good at it. I don’t seem to suffer the challenges of interacting with my garage door listed in the post.

I’ve thought about the alternatives, and I prefer the 1 button remote. The added control just isn’t necessary.

The only change I’d make is a visual indicator for the current state plus acknowledgement of (wirelessly received) button presses, but I assume energy use would be a problem.


In retrospect, my previous comment seems a little vitriolic.. Overall, I just really only want to point out that looking at a state graph of a process can be rather misleading – it’s the Markov Chain approximation of the process which actually begins to capture the underlying dynamic, and therefore point towards the better design..


A garage door should be opened by the use of the two appendages that are hanging from your shoulders.

Motorized garage doors…. only in America


Sorry if someone posts this before but I wasn’t able to read all the comments…
I would compare this to the remotes of car keys. The ones with a single button always confused me and I had to check the doors. Today, I think, every key has two buttons one for opening and one for closing. So I would choose a 3-Button Design:
- Open
- Pause
- Close
Optional some lights added on the buttons, so the button could flash if opening/closing is taking process but may this is already to much.

@Ivan: Wow, using a Markov Chain as a metaphor for an interaction design pattern. Intriguing. You’re saying it’s not the long arc of interaction modes that matters but just the expected next mode?


I definitely think the one-button solution is great – even if we usually have a two button design here in Austria. The main reason is that safety goes first and pausing the door without searching for the right button is most important.
The only thing I could imagine to add would be some feedback about the current state (eg. a three status light: no light for closed or opened, flashing for moving, on for paused; more complex: orange flashing for moving, orange solid for paused, one minute green when opened to indicate a car moving). This feedback-light should be situated at the door, not the door opener. This way others (pedestrians or other cars) are informed and warned.

again: one button is cool, so you don´t have to concentrate to much on the remote. The Information should be found next to the door, so everyone can see it and you (and others) concentrate onto the point of action.
You even don´t have to transmit status-information from the door to the (wireless?) opener, it stays as simple as it was.

@ap: “Motorized garage doors…. only in America” say what – we have them too ;-)

A button to open, a timer to close the door behind you.

Where I live we have three garage doors that operate that way.

Daniel Glazman

My dream interface would be coherent with the object it represents because of the mental model. Here, it’s a sliding door. So I want a slider or even better a clamshell. If the clamshell is closed, it means “if the door is not closed, close it ; if it’s closed, well, do nothing”. If the clamshell is opened 180°, it says “if the door is not opened totally, open the door”. Any position between 0° and 180° means “do nothing and stop moving the door if it’s currently moving”.

My vote is for the 3-state design.
Closed or Paused -> Opening (until Open)
Open or Opening -> Closing (until Closed)
Closing -> Paused


All this seems terribly over-complicated, if you ask me. As far as I’m concerned, the garage door is in one of two states – where I want, or where I don’t want it. In the latter state, one button is all that’s necessary to correct this.

Really, I understand Aza’s point about oversimplification, but the example certainly doesn’t work for me. Yes, there are a variety of states and variables involved in a garage door. But I don’t care about them – all I want to do is open my door, or close it. Everything else is unimportant details.


I’m guessing you’re dropping hints in your text as to your proposed solution, and it chimes with mine:

Ditch the pauses to simplify the model, it’s either going up or down or stopped fully open or fully closed.

You then need two buttons, similar to a modern car key where open is the bigger and easier to get to, and the close one is smaller, and offset so you can find it via tactile response with your eyes on other things.

Multiple presses are simply ignored, though some kind of chirp or response once the door acknowledges your request and goes into ‘opening’ (or ‘closing’) mode would be nice too. Like cars this would probably come from the lock mechanism itself, rather than require 2-way communication with the key fob but maybe you operate these things from round corners, I don’t know, never used one.

You only then need to fret about the occasion when you have two different valuable items about to be crushed, one below and one above the door and then you frantically press up/down/up/down until your battery runs out (though frankly if automated garage doors are as dangerous as made out in this thread then I’d join the US bashers in suggesting it’s easier to live without them.)

@Ben: Actually, I’ve used a garage door opener that works exactly that way. The thing is, the ‘one click’ model is frankly the simplest under /most/ circumstances.

I support the idea of a mini garage door on a small casing, you can slide it up or down and it will respond accordingly.


@Christopher Fahey: At any given point you have the state you are in and the state you want to be in. This is usually the most important part of the process, so you can look at it as a Markov Chain. The long term behaviour is then explained by a stationary distribution and mean recurrence times.

Aza, as far as I can tell, is making the point that when you don’t quite know the current state, this becomes a Hidden Markov Model, and we have a problem, because people suck at solving those – which is a valid point – but the provided example is not great. The cost of overlapping the two ‘pause’ states in the design is the product of 1. expected number of times you find yourself in ‘pause’ while not knowing whether you are going up or down and 2. the cost of getting it wrong. Neither of which is a particularly big number, unless you have severe short term memory problems and like keeping the family china under the garage door.

I think the case here is not for extra buttons or other weird interfaces, but for a simple extra indicator which tells you what your remote thinks about the current state of the door. This will also solve the other ‘mode error’ problem where you drive away and cannot remember whether you closed the door. (I remember ages ago seeing somewhere a key which could tell you whether you locked your door)

    That’s exactly right; while the garage door opener is a simple and not terribly important example, designers often hide the complexity of a problem behind simple-looking designs. Yet those designs are facades for hidden Markov models, which people don’t do well with. A canonical example are digital camera interfaces that have just a couple buttons but whose f stop changer is hidden five levels of menus deep.


I like the lightswitch idea. We used to leave our garage door open just a few inches for the cat to get in and out. I just needs up, down and middle to hold the garage door in place. perfect!


Wow. Really?
I bet there are many comments like this one, but here goes:

The one button model works fine if you ignore the pause stuff, which everyone says is about safety. (Don’t see it myself, though…) So why not take that out of the equation? Have a sensor on the door which detects when it’s hit something and reverses direction or stops.

In other words, don’t change the controls, change the intelligence of the object.

@everyone that thinks there should be a “paused” state:
This is not something that exists in a garage remote today. There is a safety feature that automatically opens the door when something trips the safety sensors. I don’t think users need to be inundated with modes they won’t use.

Time to hack together a “light switch” garage opener.

Why does this post exist? I have never seen a human being have trouble using a garage door opener. This seems like a clear case of over-analyzation. Who wants to have two or more buttons to reach up and fumble with when you’re pulling into your driveway? Who wants to hold a switch down until the door is done moving?

Someone presses a button and the door opens. They hit it again and it pauses. The hit it again and it goes down. The person feels dumb once and learns exactly how it works forever.

This is the same process that occurs when you have a dimmer light that can be preset – you set the maximum brightness and when you press ‘on’ the light brightens up to the max. Press it again before it gets there and it reverses and dims to off. You do that once, learn your lesson, and now you know how it works.

What’s the problem?


    “Someone presses a button and the door opens. They hit it again and it pauses. The hit it again and it goes down”

    Because from a design point of a view there are other ways of making it easier for the user to learn. But you’re right, in reality it’s not that hard to learn and once it’s learnt it’s not hard to use.

    I think we need to remind ourselves of the reality of the situation and the people using these things rather than just seeing how far we can simplify a button.

Two buttons — open and close. Pressing open while closing pauses. Pressing close while opening pauses. But from either pause you can know exactly what state you’re initiating. Simple.

I now notice that there were already comments (including people who already suggested the two button interface). So, umm — note to @azaaza: there may be an interface problem. I didn’t look close but a quick glance made it appear there were no comments.

    Aza Raskin

    That’s a good point; I’ll think about better ways of making it clear that there are already comments left but without overloading the page. I think the answer it to show a couple comments near my pretty comment visualizer, which can then expand. Show by example.


Well, when it’s completely up or down you only need one button. And when it’s moving, you still only need the one button to stop it moving.

It’s only when it’s paused halfway up that you might wish for two buttons. And of course, this almost always implies that you’re now consciously manipulating the remote control and paying attention to what you’re doing.

So ideally you’d have a control that has one button that does the obvious action, but somehow has a different physical state when the door is paused. I think maybe I’d go for one large button with two smaller, harder to hit up/down buttons next to it? When the door is paused the large button wouldn’t do anything. Obviously you could come up with a more elegant solution when the interface itself can be reconfigured, so perhaps someone could come up with a clever trick there :)

The benefit of a design like this is that it works exactly like the one button design in all the common cases, but when the door is paused it gives the user direct control over what happens next. Because of safety issues it has to be easy to throw the door into the paused state, so it’ll inevitably happen that a user will wind up in such a situation. I don’t see any way around that. But you can minimize frustration by giving the user a clear choice on what to do next.


No buttons:

When the car is started, the door opens. When the car has driven out of range, it closes. When it senses I’m approaching, it opens again. When the engine has been turned off, it closes.

OK, and maybe a button as a manual override. :’)

an analog clock crown ? ^^

I guess a 2 or 3 buttons remote would be better (open, close, and maybe pause. But pressing any button while the door is opening could pause the action).


I propose:
- A 3-state rocker switch (open-pause-close)
- The switch stays in position (being either open, pause or close, thus not requiring the user’s constant attention by having to press and hold a button and allowing the user to do more than 1 thing at a time, such as driving the car while the door is opening.)
- The ‘open’ end of the switch should be big, attractive and easy to hit to make it more accessible in emergencies (when the user is in panic mode)

And of course the remote should have MP3.


    It would be better if the states were (open-neutral-close)
    Pressing a button (1 sec) would cause it to completely change states
    Holding a button (3 sec)would cause the garage to open or close until released.
    It would default to neutral when not pressed.

    This solves the issue of no feedback mechanism between remote and garage -
    leaving it in open when it is fully open would cause issues.

This reminds me of the design for the knob for the sunroof of the car. Users can choose which state the sunroof stays by the degree of the knob it turns. The garage door opener can draw some insights from the design of the sunroof knob, I think.

I’m actually not so sure one-button clicker is a bad design.

It’s FANTASTIC for the primary use case:
“This door is either open or closed and I want it to be the opposite.”

And of the error cases I can think of that would cause a user to want it to stop mid-transit, the most important one would probably be:
“The door is closing and I just realized it shouldn’t be, STOP! UNDO!”

The single button allows people to not have to think about how to stop it, because there’s only one option. Also, It’s reasonable to assume they want to go in the opposite direction if they stopped in-transit (again, UNDO), which is what most garage door openers do.

Intentionally stopping mid-transit, then continuing in the same direction feels like an edge case.

As for hitting the button too many times after the door doesn’t respond quickly enough (I do this all the time), I think that could be error-proofed without altering the design too much. Perhaps having a one-second timeout after a press registers until the opener accepts any more commands? That way, I’d have time to get the visual feedback of seeing the door move.

It shouldn’t require any interaction from the driver at all. Make it automagic.

Christian B.

a two-buttoned controller. the open button shaped as an “O”, and the closed button shaped as a “X”. this makes the controller independent of orientation – arrows won’t work – and language and you won’t have to look at it when using it).
both buttons pause the door, when it’s opening or closing.

Stewart Alexander

Add a color changing lead that mimics traffic signal lights (red, yellow, green, the led on, the led off) that shows the states of the system.

one button, one LED.

Adding a now-nearly-ubiquitous “Play/Pause” icon to the single button may clear most confusion about paused states.


My personal preference for a UI would be one that parallels the action of the door: a slider, gradual motion up, or down, stopping at the ends.

Some have old-fashioned-looking (high-tech) gate-style garage doors. For them, analogue UI would be rather difficult to imagine.

Ah, Blair McBride, devil’s advocate. Magnetic fence, like we use for pets?


a slider like the checkbox in iphone, the surface mimics the door

|=====| <- grip

state closed
| |
| |
| | forground white(=you look at the garage door)

besides color one can add ribbles to the white (mimics the garage surface) , for blind people and additional feedback.

the door starts moving after the grip reaches end positio and the door is not in that position
if you move the grip in the opposite direction to the current movement, while the final state is not reached the door pauses. p.e. you pull to the middle during movement. Only when your reach a new desired end position with the grip the door will move again in that direction.

so the slider has 3 zones representing desired states: a “do close” and a “do open” zone, the middle zone is the “do stop” zone.

The size of middle zone should be big enough to request a certain commitment of the user for a certain desired state.

The do close/open zones at the ends should be a not just be at the final ends, but a bit bigger, to enable the remote to work even if the lid does not perfectly slides to the ultimate end. (dirt etc.)

- no led, or text needed: color,surface+position shows last state.
–> remote can be flipped upside down or on a side, and is still understandable
–> is usable international and for blind people

- not following each movement of the remote exactly but only switching between two states: open/close
- double open/open or close/close is not possible, no logik required
–> keeps it simple for 99% cases.

- in panik the reaction is often to prevent the current movement, by doing the opposite of the last movement. a pause in panik is better then moving directly in opossite direction. So we wait for the next clear command.
(stop zone must be big enough)
–> intuitive panik mode


Interesting design dilemma! While I haven’t had time to read all the comments, I didn’t see anything on auditory feedback.

Personally, without changing the design from a single button, what I’d add is auditory feedback that the garage door received the signal from the opener. Of course, this would require two-way communication between the opener and the garage door receiver, but it would eliminate quite a few error conditions. Like the double-push: When you push the opener button a second time, thinking the first push wasn’t received by the door (hear a beep, the signal was received), or when you push it before you’re in range (don’t hear a beep, push it again).

As long as I’m dreaming of bi-directional communication and feedback, I’d also like a secondary auditory confirmation that the door is completely closed (and this would be transmitted over a longer distance than current opener’s range). This is for time-crunched people who like to push the button to close the garage door as they’re speeding off to work and don’t want to drive slowly, looking through their rearview mirror, to confirm door closed properly. It would be nice to get an auditory confirmation that the door finally closed as you’re block-and-a-half away. While this may seem extravagant, you only need to return to your home after work to find that your garage door was wide open all day (because something was in the way of the safety sensor) to realize why this is an important scenario! :-)

I couldn’t agree more, Aza. “Simple” is a much-abused word and design principle. What *feels* simple (i.e. straightforward) is not always that which *looks* simple.

For this argument applied to web page design, see Jared’s Spool case for the “link-rich homepage”:

Another example: the apparent simplicity of the one-button mouse landed Apple Macs with an extra modifier key on the keyboard.

thanks for share. weldone. iyi gunler diliyorum canim

Excellent point! Now that I think about it, it’s odd that it isn’t a slider or at least a switch. I expect it is possible to make a slider with just the right tension so that nudging it up or down from either side would flick to the opposite position like a switch, while still allowing for manual overide, if for any reason you wanted to close it half way, or just want direct control over the speed of the door, or to mess with it like a DJ ;)


A slider. The point at which the slider sits determines where the garage door will stop. If the slider is moved from the top of it’s range of motion to the bottom, the garage door will do the same.

Thanks for text very good

A two-way sliding button will be best (to me at least) – push and hold it in one direction – the door opens, the other direction – the door closes. Release it – the door will stop. Much like the control of the electric window.


The market economy has clearly voted that a single button garage door opener is the preferred design.

Quick searches of Google show a variety of options, should you desire 2 button, 3 button, touch pad, light switch style models as well. They’re out there.

That said, nice topic for provoking out of the box thinking. I like it.

I vote for a Theremin style opener. :O


Why not just eliminate the need for a garage opener?
Embed it in the car and you’re done and automatically trigger it when close to home.

(When I’m walking home I do not usually take the garage opener with me… I just use the keys)

Ed Henighan

Might be a bit late to the party, but why not the following:

When fully closed/open or paused, doubleclick to open, click once to close (or vice versa). When moving, click once to pause.

Most you ever have to do is two clicks, and you always know what the door’s going to do because you tell it explicitly. Don’t have to concentrate on holding a button or anything, no need for any kind of feedback. Maybe a small LED that stays on while the door is open just to make it a bit more difficult to accidentally leave it slightly ajar.

Actually there is no problem in simplicity. The problem comes from slow responsiveness of the system. The garage door is too heavy and we can’t expect fast response but it’s not so difficult to create small process monitor. It will show last command entered. Door will open, door was paused, etc… Few graphical symbols. This could be cheep solution and will remove confusion of using this interface. One button is perfectly fine! ☺

Nice Post

I’m not sure a pause state is useful for a garage door, so I’d say a two-position switch for a wired controller and a two-button wireless controller.


I’m not sure a pause state is useful for a garage door, so I’d say a two-position switch for a wired controller and a two-button wireless controller.

Actually, most garage doors I’ve seen built in this century (I’ve seen way too many recently-built garages) have no pause state, only the pause’ state. The pause state is very unlikely and still available at pause’ so there is no loss in removing it, and in emergency cases opening is 99% of the time what you want to do. So there can be no modal mismatch anymore and the button does what you want as long as you know the current state of the garage door.

The only problem I can think of can occur is when you are not within eyesight of the door, such as when you are driving and click the button assuming the garage is closed when in fact your housemate/friend/wife/husband has opened the door and is now frantically running toward the in-garage button to stop the door.

Given the impact of the downturn in the mens belts labor market, GHD, cheap gucci belts has undergone many workers have no other choice, fixed-term employment cheap louis vuitton belts for men or agency. Therefore, some could change cheap desiger belts from low gucci belts on sale labor costs all on a regular louis vuitton belts cheap basis. In other words, if they could move from poverty to escape C temporarily. The charity says that government policy and legislation, MBT, on the basis of the phenomenon and requires change, including improved rights and conditions for employees to wear.

Question: What’s a better garage-door opener interface?



I am intreaged now about the possibility of a simple editor that is able to keep track of revisions in documents. Sort of like a source control embedded into the document.


Isn’t the comment system on your blog an example of the complexity. The multiple call outs are a nice way of visually showing how many comments are here, but it is confusing to me whether I should click each one of them to view the comments. Isn’t the number ’98 comments’ enough to understand the number, if that is what you want to showcase.


Was this post a gimmick to see how ridiculous your readers are? Most of these ideas are idiotic. The current design works well.

University on her work; that lecture was similar to the one

University on her

Three buttons, one for going Up, other for going Down and a small one to Pause either said actions.

The nearly 400 million current Firefox users is a testament to our ability to make those tough calls and change towards the better. As our user porno izle base continues to grow, those calls will only get tougher.

I know this is off topic but when I first saw your comments secion with the flickering speech bubbles I thought that I had to click on each speech bubble to view the comment – Okay I didnt but being a web developer that has given me an idea!

University on her work; that lecture was similar to the one

There is something more about this.

The behavior of “what happens when you click” depends on what the door is currently doing, or what it has done until recently. Or even worse, you need to look what it will do, because you can’t really tell.

This means, you need to carefully look at the garage door to do anything reasonable with the remote control. After some time it will be hard-wired in your brain: Don’t touch the remote, without looking at the door. And be careful if the door really does what you intended.

A nice educational effect, isn’t it?
Having more buttons could make your life easier, but could also make you less careful. And then one day you your dog is being crushed, and you don’t even notice.

Not entirely what your article is about

the promise. It’s a crutch we can’t allow ourselves. Me most exorcise the devil of false simplicity


i really love this post can i add it to my website trendy boy

I just wanted to write a brief remark to appreciate you for these marvelous information you are sharing on this site. My considerable internet search has finally been recognized with high-quality facts and techniques to share with my good friends. I would state that that many of us website visitors actually are extremely endowed to exist in a really good network with many marvellous professionals with great guidelines. I feel rather fortunate to have come across your web pages and look forward to some more amazing minutes reading here. Thanks again for all the details.

Some truly great articles on this web site, thank you for contribution. “I finally know what distinguishes man from other beasts financial worries. – Journals” by Jules Renard.

WOW they are such beautiful photos Anel

Recently I have been looking for website about this topic for a while and yours is the best I have discovered so far. Thank you.


I like such topics

There are certainly a lot of details like that to take into consideration. That’s a great point to bring up. I offer the thoughts above as general inspiration but clearly there are questions like the one you bring up where the most important thing will be working in honest good faith. I am sure that your job is clearly identified as a fair game.

Outstanding post. As quickly as I noticed this website I went on delicious to share some of the love with them.

Aw, this was a very wonderful write-up. The theory is that I’d like to generate such as this additionally – using moment as well as real attempt to manufacture a great content… nevertheless what can When i declare… My partner and i delay doing things a lot rather than often receive a thing accomplished.

With the whole thing that seems to be developing within subjects, all of your scenes are generally quite refreshing. Even so, I apologize, but I can not subscribe to your entire plan, all be it exhilarating none the less. It seems everyone your opinions are generally not completely Justified and in reality you are generally yourself not really fully convinced of the argument. In any case I do not appreciate a look.

This is the perfect blog for anyone who wants to know about this topic. You know so much it s almost hard to argue with you (not that I really want . HaHa). You definitely put a new spin on a topic that has been written about for years. Great stuff, just great!

I was very encouraged to find this site. I want to thank you for this special read. I definitely enjoyed every bit of it and I have you bookmarked to check out new stuff you post.

The beauty of these blogging engines and CMS platforms is the lack of limitations and ease of manipulation that allows developers to implement rich content and skin the site in such a way that with very little effort never see why the site tick all without limiting content and effectiveness.

What you say is absolutely true. I know that everybody must say the same thing, but I just think that you put it in a way that everyone can understand. I also adore the images you put in here. They will fit well with what you re saying. Im sure you ll reach so many people with what you say.

This is my first time i visit here. I found so many entertaining stuff in your blog, especially its discussion. From the tons of comments on your articles, I guess I m not the only one having all the enjoyment here! Keep up the good work.

Simple, wonderful what you ve done here. It is pleasing to look you express from the center and your clarity on this significant content can be easily viewed. Extraordinary items and expect your future updates.

Aw, this was a post that was really good. In theory I d like to write like this too taking time and real effort to make a good article . but what can I say . I procrastinate a lot and never seem to get something done that.

Far, this post is really sweet about this important topic. I am in harmony with the conclusions and are greedily looking forward to the update entry. Saying thank you will not just be sufficient, for the wonderful clarity in your writing. I will immediately grab your rss feed to stay informed of any updates. Wonderful work and much success in your business dealings! Please excuse my poor English as it is not my first language.

Let me start by saying beautiful post. Im not sure if this has been discussed about, but when using Chrome I can never get the entire site to load without refreshing many times. It may be my computer. Thank you.

Not? T better written. Reading this post reminds me of my old room mate! He was always talking about it. I will forward this article to him. Pretty sure he will have a good read. Thank you for sharing!

Resources like the one you mentioned here will be very useful to me! I will post a link to this page on my blog. I am sure my visitors will find that the most useful.

I thought it was going to be some boring old post, but it really compensated for my time. I will post a link to this page on my blog. I am sure my visitors will find that the most useful

Hrmm that s weird, my comment got eaten. Still I would say that it is great to know that someone else also mentioned this as I have trouble finding the same information elsewhere. This was the first place that told me the answer. Thank you.

Hey, just looking around some blogs, seems a pretty good platform you are using. I m currently using WordPress for some of my sites but looking to change one of them to a platform similar to yours as a trial. Anything in particular you would recommend about it?

You can not intended to do so, but I think you have managed to express the state of mind that a lot of people entering Taste want to help, but not knowing how or where, is something a lot us are going through.

You got a really useful blog I have here reading for about an hour. I am a newbie and your success is very much an inspiration for me.

This article gives the light in which we can observe the reality. This is very nice one and gives in depth information. Thank you for this beautiful article

Keep em coming . you all do a great job at such concepts . can not tell you how much I, for one appreciate all you do!

Thank you for taking the time to discuss this, I feel strongly about it and love learning more on this topic. If possible, as you gain expertise, would you mind taking updating your blog with more information? It is extremely useful for me.

Hi webmaster, commentators and more! Blog is absolutely fantastic! Lots of great information and inspiration, both of which we all need! B Keep em coming . you all do a great job at such concepts . can not tell you how much I, for one appreciate all you do!

Have you ever considered adding videos to your blog posts to keep the more entertained the audience? I mean, I just read your entire article and it was quite good but since I m more of a visual learner, I found that to be more useful well let me know how it turns out! I love what you guys are always up too. The clever work and reporting! Keep up the great work man I added you guys to my blogroll. This is a great article thanks for sharing this informative information . . I will regularly visit your blog for some latest post.

Pretty good post. I just stumbled upon your blog and wanted to say that I ve really enjoyed reading your blog posts. Any way I will be subscribing to your feed and I hope you post again soon.

That fantabulous post this has been. Somehow seeing this kind associated with useful post. I am grateful to you and expect much more associated with posts such as. Much obliged.

Thank you so much for writing all of the excellent information! Looking forward to checking out more posts!

Wow, this is a post that is really good quality. In my theory, AOD want to write like this also, AI took the time and effort to make a real good post. but what can I say. I procrastinate a lot and never appear to get something done.

This post is quite interesting. I really never thought I could have a good read by this time until I found this site. I thank you for writing given. your information is also very nice. Thank you for the great post. From tons of comments on your articles, I guess I m not the only one having all the enjoyment here! continues to work well.

Interesting topic for a blog. I searched the Internet for fun and came on your website. Unusual items. Thanks a ton for sharing your knowledge! It is very nice to see that some people still put effort into managing their websites. I am sure

I am really satisfied with this posting that you have given us. It is truly an amazing feat made you. Thank you and looking for more posts

I love your blog, i have it in my rss reader and always like new things coming out of this

    Sounds like the striker plate on the door frame may be recsseed too much. I had a door that did the same thing. I put a couple of small washers behind the striker plate and the door closed and latched fine.Next step is changing the door knob assembly to one with a longer bolt coming out of it.mike

I must tell you I was impressed. Very rarely do I find a blog, both educational and entertaining AOS. Just wanted to let you know that you have most definatly hit the nail on the head. Your mind is a good one. Thx is all I can say.

Most powerful, I just given this to a colleague who was doing a little study on this. And he actually bought me breakfast as a result of I found it for him . . smile. So let me reword that: Thnx for the deal with! But yeah Thnkx for spending the time to discuss this, I feel strongly about it and love learning more about this topic. If achievable, as you turn into expertise, would you mind taking updating your weblog with more information? This, AOS is very useful for me. Big thumbs up for this blog put up!

In fact, I learned about all of this, but with that in mind, I still think it is useful. Good work!

Hello I found your blog by mistake when i was searching AOL for this matter, I must tell you that your blog is really helpful I also love the design, which is cool!

With the whole thing that seems to thrive in the subjects, all your perspectives are usually quite refreshing. Even so, I apologize, but I can not subscribe to the whole plan, all be it exhilarating nonetheless. It seems everyone that your opinions are not entirely justified and, in fact, usually do not really trust fully the argument. Anyway thank examined.

I see you put a lot of work on this site! Keep writing!

An interesting dialogue is worth comment. I think you should write extra on this matter, won, Aot be a taboo subject but generally people are not enough to talk about this issue. To the next. Applause

I was impressed with the quality of information on this site. There are many great resources here. I am sure I will visit this place soon.

It is a great resource that you are providing and you give it away for free. I enjoy seeing websites that understand the value of providing a quality resource for free. I really loved reading your post. Thank you!

Super blog post, I count on updates by you.

I was very encouraged to find this site. I want to thank you for this special read. I definitely enjoyed every bit of it and I ve marked you to check out new stuff you write.

Thank you for another essential article. Where else can you get this kind of information in such a complete way to write? I have a presentation next week, and I am in search of such information.

The beauty of these blogging engines and CMS platforms is the lack of limitations and ease of manipulation that allows developers to implement rich content and skin the site in such a way that with very little effort is how I see it makes this site tick all without limiting content and effectiveness.


Finally, an issue that I want. I have searched afor information of this caliber for the last several hours. Your site is greatly appreciated.

This is the perfect blog for anyone who wants to know about this topic. You know so much it s almost difficult to argue with you (not that I really want . haha). You definitely put a new spin on a topic that has been written over the years. Great stuff, just great!

This is a smart blog. I m serious. You have so much knowledge about this issue, and so much passion. You also know how to make people rally behind it, obviously from the responses. You have a design here that is not too flashy, but makes up for what you say statement. Great job, in fact.

I admire the valuable information you offer ina your articles. I will bookmark your blog and my children check here often. I am quite sure they will learn lots of new things here than anyone else!

I was very encouraged to find this site. I waant to thank you for this special read. I definitely enjoyed every bit of it and I have you bookmarked to check out new stuff you post.


I was imprssed with the quality of the information on this awebsite. There are many great resources here. I am sure I will visit this place soon.

An interesting dialogue is worth comment. I feel that you ashould write extra on this topic, it won, Aot be a taboo subject but usually people are not enough to speak on such topics. To the next. Hail

Although I am not a noob in the website industry, your siate is truly unique and features some useful insights. Enjoy it fully! I, ll have entered my blogroll, I think it will give more value to the visitor.

Technological revolution in all these years has however changed
the entire facet and practice of brand promotion. Simple ads,
with few graphics and a voiceover, can cost as less as $20.
The media kit will present all information about your offer to potential advertisers.

in my opinion , comprehensive interface that can support us as user performance . Especially if it is supported with a good screen , would be the maximum .

I’m really enjoying the theme/design of your website. Do you ever run into any web browser compatibility issues? A few of my blog audience have complained about my site not working correctly in Explorer but looks great in Chrome. Do you have any ideas to help fix this problem?

foto bugil Kebaikan hanya bisa disampaikan dengan kasih sayang foto memek

Leave a Comment