The mouse is an unnatural abstraction. Just because it was better than everything else at the time - and it was - doesn't make it anything other than a spooky indirection.

We get this kind of thing because our superpower is dealing with spooky indirections and weird abstractions. We are not normal.

What is better than the mouse, though? Touchscreens probably only work in your lap, which is not the most ergonomic position for the majority of work. Touchscreens on a monitor are probably too straining since you would have to hold your hand upright all the time.

Not saying the mouse is forever, but I don't yet see the better alternative available for the mass market (maybe those brain wave scanners?).

This is a case where, I reckon, if the mouse is the answer, we're asking the wrong question. What's better than the mouse? Not shoehorning everything we want to do with computing into the form factor of "a computer".

OK, this is going off into left field, and I really should blog this, but take the iPad. For what the iPad does - video watching, web browsing, reading extended text, certain classes of creative apps (eg http://hexler.net/software/touchosc) - it's easily the best thing on the market. It's the best Web content consumption device anyone's produced, ever, hands down, which means it's the best television and printing press ever produced.

But it's not a general-purpose computer. In fact, it's those things because it's not a general-purpose computer. It's in the notebook niche, and by that I don't mean clamshell, I mean spiral-bound; it's the nearest we've yet got to technological paper. If it tried to be a general-purpose computer, it would be much less good at the things it's good at.

That points to the ugly truth of our industry - and let's face it, it's something we all know in our hearts:

General-purpose computers suck.

They really do. They're a kludge. They're a set of dodgy tradeoffs, due to economic and technical limitations. They're unintuitive and fragile. They're expensive. They're rapidly obsolescent.

The computer I want to use for programming isn't the same as the research-and-content device I want to use for writing isn't the same as the piano-keyboard-and-touch-screen-and-dials device I'd want for writing music isn't the same as the book-a-like I want to read on before I go to sleep. One device can't be good at all these things.

When it comes down to it, the mouse is a hack to make a general-purpose computer sort of acceptable for many of these things. It's a genius hack, but it's a hack.

Special-purpose computers create a lot of clutter and confusion. Life before smart phones was pretty lame: you had a phone, a media player, a camera, a notepad, a GPS device.... Well, actually, you didn't have that stuff, especially when you needed it, because you left it at home rather than have a bunch of devices banging around in your pockets. Special-purpose computers are actually becoming less and less common as general-purpose devices like smart phones and tablets displace them from the mobile space.

What we need is not special-purpose computers but general-purpose computers specialized by form factor, running applications that are optimized for the form factor and its style of interaction. Sometimes you want to sit down and type a ten-page report or reply to dozens of emails. Sometimes you want to lounge on the sofa and watch YouTube videos. Sometimes you want to lounge on the sofa with someone and watch a movie on a big screen with a big sound system, and you want your hands free for other activities.

I know that sounds really dumb and non-visionary because it describes exactly what we have now, but what the heck. It's a beautiful world. I also know that by saying "It's a beautiful word" I'm destroying any usability credibility I might have, because actually all normal people are suffering terribly, but again I say, what the heck.

Your first paragraph falls under the "Pocket Exception" (proposed by Jesse Schell). If it's in your pocket, you want it to converge. If it's not in your pocket, its going to specialize. To steal his example: the swiss army knife is a great pocket device, but you'd never use a giant swiss army knife in your kitchen.

You must not live in an apartment :-)

Have you seen the number of buttons on a modern rice cooker? Some rice cookers incorporate pseudo-pressure cooking, and some are even genuine pressure cookers. There are also combination convection/microwave ovens, and I love my combination immersion blender and food processor precisely because it serves two purposes in the same space as either one. It doesn't crush ice or blend as easily as the old stand-alone blender I had, but I compromised on features to save space. (It also has fewer parts and is easier to clean than the stand-alone blender or the dedicated food processor I used to have, so convergence didn't compromise usability.)

So then maybe the pocket exception should be renamed space exception, and now it explains Japanese appliances well.

I disagree - I actually want my Computer to do as many things as possible. I don't want to clutter my house (or my purse) with a multitude of devices.

I think it is a fallacy to conclude that just because Windows sucked, all computers suck.

Also I don't see the iPad replacing computers yet. I don't own one, so I am not even sure I consent to it being the best device for watching movies - the screen is small, and presumable I have to keep it on my body at all times to have a good viewing distance.

Also, I don't understand your argument that it is the wrong question. Apparently there still are tasks that require a mouse - how do you suggest replacing them? That there are tasks that don't require a mouse doesn't invalidate the concept of a mouse.

While you say you disagree, it sounds to me like you two are in agreement: one device doesn't fit all needs. Tablets can be fantastic. That doesn't in the least bit mean that "computers" as we've traditionally known them will or should go away. It's just that their role is somewhat reduced and clarified, empowering new types of device.

Touchscreens on an upright monitor would be too straining on your arms as you hold them up. Touchscreens placed flat would be too straining on your neck as you look down. Finally, your hands get in the wa of the screen.

Theres plenty of reasons not to prefer touchscreens as an all-purpose input device (though I DO like touchscreens...).

PS: Yesterday, I used an ipad for the first time. It left a pretty good impression. Yes, it has the hands-in-the-way problem and I wouldn't want to type on it for too long, but as an I can use it to read stuff and interact with it quickly with my hands when I need to device, it seems to work very well.

For me, there's a bit of a natural distinction: "workstation" tasks need a big screen and, at least for the currently foreseeable future, a mouse. "Casual" tasks need something personal, intimate, and "touchy". The world will benefit from having both.

This isn't to say that all tasks can be easily assigned to one of these categories, at least without further parameters specified: fingerpainting with Brushes and compositing 3D renderings for a motion picture are "the same task" in a sense. But each fits nicely with one device instead of the other. And, of course, there'll always be a gray area.

This is something I noticed when teaching my dad to use a computer; when I said "right" click, he always press the button with his index finger and will continue to do so until I tell him he have to press using the middle finger.

So my guess is, it has something to do with the fact that people are used to using their index finger to press something. Multiple buttons on digital camera are pressed using the index finger. Buttons on a TV remote are pressed using the index finger. Right click? Press the button? The intuition kicks in, and the index finger is used.

The "right click" that involves alternate pressing apart from what they are used to do so daily might be unfamiliar concept to those people. A vague, unfamiliar concept that is not intuitive to them are more likely to be forgotten. However, a set of rules for a specific task might be easier to remember even though it is unintuitive. A radio dashboard, for example.

So you don't understand that someone who was never bought up with computers, big screens, or even buttons can get confused with new technology?

What I don't understand is all the people who haven't quite grasped the fact that people are different, and think about the world differently. While right clicking would be second nature to many of us, a lot of people still don't "get" computers, and a lot of what happens is magic to them.

What's the distinction, though? What's the difference between right-click and all the other things she can do?

She can work her digital camera. She can work her cell phone. (She even sends text messages now!) She can work the DVD player, the TV, and her fax machine. She even managed to program her VCR to record shows a few times back in the days when programming a VCR was a real accomplishment.

Also, she considered majoring in math in college and became a high school math teacher in her forties, so she's never been averse to abstract thinking. (It also proves she can work a graphing calculator.) She's had Windows in the house since 3.1, and before that we had an Apple IIgs, which had a GUI finder and other mouse-centric apps.

So what's the distinction? To say that people who came to computers late have trouble with stuff is an observation, not an explanation.

Pretty much all of those things require you to memorize some arcane series of steps to accomplish anything. Most likely, she is taking that mental model of how to use electronics and applying it to the computer.

I believe you said that, once given specific instructions to accomplish a specific task on the computer, she can generally remember and repeat those steps in the future. This fits the model of those other devices, where you must memorize some arbitrary button or combination of buttons to make it do what you want. VCRs, TVs, fax machines and digital cameras do not have a "right click" menu attached to them, and these are the devices on which she has formed her mental model of how to use electronic devices.

There's no point in denying the fact they have trouble, though. If a similar group of people keep having trouble with the same interface of a user application, then the usability is just plain bad. It's not the fault of the users, it's the design's fault.

Don't get me wrong, I feel with you, though. I don't want to count my problems explaing the concept to beginners^

Or maybe they are just not motivated enough and like a good excuse for not having to do stuff? "Strategic Incompetence" is an amazing concept :-)

I wouldn't say that, Tichy. If people buy a computer for their homes, they are motivated to learn new things. If it doesn't work out for them, they certainly could manage with more motivation.

But so could anyone achieve a BSc. Everyone with a BSc could have an MSc. All having a MSc should be working on their PHD. You get the idea. If you can go one step, you _could_ always take two.

Sometimes you don't want to go that extra step, because there's not enough ROI. And sometimes you don't need to, because it was a design flaw.

Or you don't need to because you have a son you can ask for help instead...

Maybe it is one of the few excuses to get in touch. So the iPad will lead to more loneliness in the lives of old people, because suddenly they don't know what to ask anymore :-)

Or they could for instance share pictures they took of their beautiful summer garden instantly with their grandchildren - including an invitation for dinner on the weekends.

As a techie, I want to believe that technology enables communication sometimes. Especially for edge cases like older folks.

I'm only available on the phone, and she usually suffers with something for a while before asking me. She's good about that. Also, the questions have become less and less frequent every year as she adds to her repertoire of concrete know-how.

Didn't mean it in a bad way - just some people don't know what they are missing if they don't use computers, so they don't make that much of an effort.

That doesn't explain why, after being shown a task and explaining that the same action enables other, similar tasks (and in this case, saying that right clicking will open a menu and telling the user to simply try it when looking for a new action), someone would continue to not be able to perform such tasks. A handful times is normal, but after a while, "right click opens a menu of available actions" should sink in.

These same people can do much more complex things, theres no reason that they couldn't learn things like right clicking after a while. How hard is it to right click as an automatic "see if it does what i need" response after having been shown and told to do so multiple times?

Same here - it is really just one rule to remember, instant PC expert. That is also why I don't understand everybody raving about Macs. Macs are not consistent with the second mouse button rule (they used to have only one button), therefore there is no way to become an instant expert on a Mac. I always hate using a Mac because it makes me feel so helpless.

My mother learned the second mouse button rule after a while, btw. Don't give up :-)

"That is also why I don't understand everybody raving about Macs. Macs are not consistent with the second mouse button rule (they used to have only one button), therefore there is no way to become an instant expert on a Mac."

Although you may have answered your own question, as it seems that the right-click button does not fit the mental model of non-computer-experts (at least, that is what we have been discussing).

Maybe the Mac works better for those people because it does not assume that the user will just right-click to figure out how to do something?

Right-click does work well, of course, in most contexts on the Mac. Interestingly, though, I've been finding myself appreciating having a big obvious button to click even when I know the functionality I want is also in the right-click menu. I suspect Fitt's law has something to do with it.

Yes. There has traditionally been a drive toward putting all major functions of a Mac app in the menu bar — even if they also additionally got a shortcut such as through a "right-click" contextual menu. This leads one to expect to be able to find any available option through scanning a single set of options (categorized roughly by the thing or type of action they effect) rather than having to mentally union all the options from several disparate places to get a complete mental model of the app's actions and options.

And even if they only have one button, they still have the command+click combo (which, IMHO, is much more complicated than simply having a second dedicated click button).

command+click is horrible, as are the overhyped gestures. There is no way to accidentally discover them.

For several weeks I had to Google whenever I wanted to rename a file on my Mac. I still have to Google for the key combination for showing the desktop (cmd+some function key, but there are 12 function keys).

Just hit enter when you clicked on a file. That's the same F2 does in Windows. Just the enter button is closer to regular use than F2.

As to showing the desktop: FN+F11. But you can configure that in System Preferences -> Exposé and Spaces.

Easy enough. I use it every day. Besides you're a linux guy. You should take a look at the visor terminal (visor.binaryage.com). It enables a Quake2 like terminal via hotkey all the time. That's a great productivity booster!

Wait.. are you trying to tell us its easy by introducing yet more odd and complicated buttons to the mix? Command + Click is too complicated - use Enter or FN+F11 instead? What??? I'm really not buying into your argument here!

I can not remember cmd+f11 (my guess would have been f3). I don't use it often enough. I can remember the little button next to the start menu that looks like the desktop. I can remember to try second mouse button when I want to do something special with a file.

Key combinations are impossible to discover by exploration. With the right mouse button, you can solve every problem by exploration.

To be fair the macs have a second button now but you need to enable it I think. (though it might be enabled out of the box I cannot recall) if you want to use another mouse with more than one button you are certainly able to and it will work as you expect it to. But if you cannot do any of that the context menu is only a ctrl+click away.

People just need to play some Minesweeper and they'll learn about the existence of right click. That's why Minesweeper was created, isn't it?

I have to say that I have watched many people play minesweeper without the right click.

In fact it was a while before I realised that you could leave little flags by right clicking and I still rarely do it today if I'm playing.

So have I, actually. Many people also don't know about clicking with both left and right buttons simultaneously. If you do it on a square that has all of it's flags set, it clicks all other unclicked squares around it.

This is very nearly the only reason I ever used flags. It took me a while to accidentally flag something, then a tiny bit more to discover how to unflag. Fast forward quite a while, and I discovered the auto-sweep, and now use it nearly exclusively (way harder to misclick that way).

In Germany we also had this game called "Moorhuhnjagd".

My personal opinion is that clicking with the mouse is a flawed design. This makes it hard for beginners.

Why would you need to double click to start an application on your desktop while you only need one while you're in your webbrowser?

But even more so, I believe that the concept of files is hard to grasp for a real beginner. Modern applications (like Google Docs, Evernote, ..) do not use them at all. The concept of "content" of a certain application is a lot easier to understand.

Take the iPhone for example, there's no files at all (for the enduser). There's no need to "rename", to "open with another application". Does this infuse aggrevation in developers because they feel drained of power? Yes, of course! But those are devices created to be used, not to configured.

Files are useful if you want to share data between applications. I thought about it just today - maybe they could just be seen as the most general interface (a blob of bytes, basically). I suppose Apple & Co want to replace that by handlers and plugins - want to send a photo by email? Either you need an email plugin for iPhoto, or an iPhoto plugin for your email program (and how does it work if you use a web based email program?).

I don't think this approach will scale well in the long run (there are a lot of file types...). The iPhone is a bad example - I suspect in the long run apps just don't interact enough, which is why they will be superseded by web apps again. They are just good for isolated use cases.

Android's implementation of Intents and its sharing menu are probably a better example. I take a photo with the camera app, then hit share. Android automagically knows that Gmail, Messaging, Picasa Uploader, PicSay Pro or any number of image manipulation apps can do something with that image. Reading a cool story in the NPR app? I hit share and can, again, select Gmail or Twitter or whatever else the system recognizes as being able to handle that data type.

I don't think it's a replacement for filesystems, but it is a very powerful way of handling data on a platform as limited as mobile.

I like the intents concept, too. Unless there are > 100 apps that claim they can do something with your intent.

The thing is, there aren't actually that many file types which you'd want to open in multiple programs. Maybe fifteen or so which you'd expect to do data interchange with - a handful of picture formats, ditto movies, ditto audio, ditto text/DTP, and Excel and Powerpoint.

Everything else is special-purpose, so as long as you can mail the document, you're fine.

iTunes and iPhoto already handle the majority of those. Add a personal office-document-management tool and you're already pretty much there.

Isolated use cases is just what the average user needs. He doesn't want to be distracted by "byte blobs" and "program interaction".

Do one thing and do that well, I say.

The thing is, I am often very confused when I want to do things on a Mac. Maybe it is just because I am used to the old fashioned way of just attaching a file to my email. Like with the right click, files always work the same way. Unless every app comes from Apple (which is probably their goal), not all apps will adhere to my expectations in the same way. And even Apple will probably have a hard time to pull it off.

Some tasks simply are complicated by nature. I don't think all use cases are isolated. For example mailing an image is not an isolated use case (or rather, would it really be less confusing to have different apps for mailing images and mailing text?).

Since when can't you mail images or other files with OSX? Just drag and drop a file into your mail and you're good to go. There's also the "attach" button when writing a new mail.

How do you drag a photo there if it is in iPhoto? Where does iPhoto store it's photos? Is it even legal to just copy fotos out of the iPhoto folder?

I also tend to have problems on OS X because the finder sucks and dragging files to the task bar -> window opens -> drop on window often does not work for me (I suppose it should work?). If you open the finder, it hides the email window...

Just drag the photo from iPhoto to your new mail. Be strong, there's a lot of good magic involved(;

Also, iPhoto stores the pictures pretty conveniently as you might expect: ~/Pictures/iPhoto Library. And of course you can copy the files out of there.

And finally, you might want to get a screen that can fit a new mail and a finder window next to each other. A 13" monitor should do that just fine, btw(;

I identify with you to a great degree: I have to work to imagine how mousing could be hard.

Nevertheless: none of our bafflement changes in the slightest way the fact that there are people who are, indeed, so baffled.

It delights them to use something that doesn't baffle them. And that delights me.

I don't disagree at all, and I'm mystified why everybody assumes that just because I show a little frustration on a message board with other techies, I must be bent on haranguing users into compliance.

I feel a little vindicated by the fact that people have suggested several different explanations here, none of which seem to be universally accepted. Is the mouse unnatural? Are abstractly-minded (with respect to computers) people different from concretely-minded people? Is it a matter of being exposed at a young age? I think there are credible objects to each of these arguments.

For instance, the first suggestion, that the mouse is simply unnatural, doesn't account for variation among users, and it's only a constructive criticism to the extent that it stimulates development of superior alternatives.

The second suggestion is circular, because "abstract-mindedness" in this context doesn't seem to correlate with any other measure of abstract-mindedness. Are mathematicians or physicists naturally better at right-clicking than salesmen or physical therapists? That doesn't seem to be true, at least among folks who are middle-aged and older.

The third suggestion implies that usability problems with well-designed WIMPy interfaces are a historical blip that will naturally disappear unless there's sustained radical innovation in interaction technologies and UI paradigms. I could believe that, but some people strongly believe otherwise.

If a single one of these is responsible for the bulk of the difference between me and my mom, then conclusions can be drawn about the future of user interfaces. If the mouse is unnatural, problems will persist until someone invents a superior replacement. If abstractly-minded people are different from concretely-minded people, a way will have to be devised to cater to both. If it's a matter of early exposure, then there are usability generations washing slowly through dimensions of class, culture, age, and geography, and different usability guidelines apply to different generations.

It's an interesting question to think about, especially since new form factors have created more room for innovation. Right now, less complexity seems to be the wave of the future. It will be interesting if a new form factor emerges to prove that people can comfortably handle more complexity than current technology allows them to.

I realised the web browser (from Netscape onwards) is easy to use mainly because of the mouse cursor changing to a hand when it hovers over a link. Or a text thingy hovering over an input form. Without the mouse changing cursor, you wouldn't know what was what...

Who the hell down-mods honest confusion? If it's so obvious, then you try explaining it.