Tobias Bjerrome Ahlin, an interface designer at Spotify, is a big believer in skeuomorphism. Whereas Apple is a strong advocate of this design concept, Microsoft is clearly moving in the exact opposite direction, while Android is in the process of moving away from skeuomorphism entirely, to a more digital experience. As a passionate hater of skeuomorphism in UIs, I found Ahlin’s examples to be a bit weak.
My big problem with skeuomorphism in UIs is that it feels degrading, like I’m being infantilised, treated like someone who should be taken by the hand and guided through to the big bad scary world of computing. You know, what Clippy was for. Heck, some of Apple’s interfaces these days – and the Photobooth and Find My Friends examples in Ahlin’s article illustrate this perfectly – remind me of Microsoft Bob, just with better graphics.
Ahlin’s central argument is that skeuomorphism makes UIs more fun. “An interface that is not only easy to use, but fun to use, engages the user and creates an experience where obstacles are easier to overcome, and thus an experience where the product is easier and more fun to use,” he argues, “Done right, skeumorphism can retain the simplicity and ease of use of an interface while empowering users to act.”
I feel like the exact opposite. When you look at an iPad, you’re looking at a screen. If the screen pretends to be paper but it doesn’t actually act like it’s paper, it’s just confusing. iCal on the iPad looks like an actual calendar with pages, and as such, I lost count of how many times I instinctively tried to flip the pages to move forward in time (I believe Apple fixed that now). This is a flaw many skeuomorphic user interfaces carry: they look like something from the real world, but they don’t actually act like said object from the real world. All skeuomorphism does in these cases is add another layer of complexity.
He also compares Garageband’s keyboard interface – made to resemble a real world keyboard – to a mockup he made of how such an interface would look in a mixture between Metro and Holo. While he’s actually cheating in that comparison (the Apple version has text on some buttons, while the mockup does not), it actually illustrates what happens when you take skeuomorphism too far.
Apple’s keyboard uses a sort of fancy cursive font that’s virtually unreadable – sure, it may be what the real world object Apple copied was using, but on this display it’s very hard to read. This is typographical skeuomorphism: instead of opting for a font optimised for on-screen reading, Apple dumped a print font on a screen, therefore making it harder to read. On top of that, the fake shadows, perspective, and reflection/gloss are off; they’re supposed to mimic the real world, but it’s just off enough that it becomes distracting. The gloss on the black keys is too bright to boot. Lastly, there’s a whole bunch of wasted screen real estate dedicated to imitating real-world buttons that just distracts from what the tool is supposed to do.
Comparing it to his – as said, not entirely fair – mockup, I know which one I’d choose. His mockup is cleaner, clearer, and therefore, more inviting. The buttons convey their meaning quicker because they’re simpler and more to the point. I’m using a digital device, so it makes perfect sense for the device to feel digital. When I get into my car, I don’t expect having to yell giddy-up or sit on a saddle.
Then there’s an even bigger, more overarching problem with skeuomorphism in user interfaces: the total and utter abandonment of consistency. Every application on iOS looks different and works differently. Buttons are jumbled all over the place, and even something as simple as how to move backwards in an app is inconsistent (arrows? Labels? Both? Top-left? Bottom? Swipe?). Non-skeuomorphic UIs, like Metro and to a lesser degree Holo, are far more consistent, making it infinitely easier and more intuitive to carry over knowledge of one application to the next.
On iOS, applications are visual and behavioural islands. On Windows Phone 7 (the poster child for anti-skeuomorphism), they’re all visually and behaviourally connected.
Now, his first two examples – Paper and Brushes – are far better examples of how skeuomorphism should work: like everything in life, moderation is key. And this moderation is what’s severely lacking in Apple’s recent UI designs, and this being Apple, many developers and designers just blindly up and follow them. I feel so much like a child when using iOS that I’m reconsidering purchasing the new iPad, because I’m pretty sure its new display just means a larger canvas for Apple to smear Clippy and BOB onto.
I’m sure some people will point to iOS’ sales figures as proof that skeuomorphism works, but that argument is just as flawed here as it is with desktop operating systems, music, film, and everything else. To me, skeuomorphism feels like moving backwards in time, into the wrong direction. I thought the technology world had long ago settled the matter that adding real-world metaphors to user interfaces is simply a very bad idea. I guess I was wrong.
This is actually something I alluded to in my iPad 2 review: the success of the iPad pretty much means Apple has very little – if any – incentive to improve the iOS user interface or move it in different, potentially better directions. When you really break it down, there’s been very little innovation going on in iOS’ user interface since its original inception, and this doesn’t bode well for the future.
As such, I’m fascinated by Metro (on tablets and smartphones, not on desktops or laptops) and Android 4.0’s Holo: they’re the exact opposite of what Apple is doing, and a breath of fresh air for us skeuomorphism critics.
This again?
http://homepage.mac.com/bradster/iarchitect/realcd.htm
http://homepage.mac.com/bradster/iarchitect/phone.htm
http://homepage.mac.com/bradster/iarchitect/qtime.htm
Nope, this hasn’t suddenly become a better idea in the last 15 years. A computer is still not anything but a computer. While Microsoft is applying Metro inconsistently and poorly (in Windows 8), they at least got one thing right: don’t try to make GUIs look like physical things. It’s absolutely a dead end, and it’s pitiful how Apple is embracing these ideas, as they have otherwise been holding the “consistent UI” flag fairly high for a long time (exceptions like Quicktime notwithstanding)
The problem with this crap is the same problem with Unity and wobbly windows or MSFT and adding all that stupid “shrink/grow” morphing crap on their UIs and that is for the vast majority they just want to do a task as quickly and easily as possible and all this crud does is slow you down!
You can tell the designers of all this extra fluffy crud think their stuff is so wonderful yet you talk to the users and they’ll find it irritating, even if they don’t know what it is exactly that is irritating them. Take clippy, I don’t know of a single person that was actually helped by that stupid thing but i know a LOT of people that got the brakes put on what they were doing by el stupido popping up at the wrong time and breaking flow. Or all the UI effects, i don’t know how many times people have complimented me when i’m done working on their PCs “I don’t know what you did but its so fast now!” when in reality all I did was turn off all the stupid effects they were wasting time waiting to finish.
So any developers reading this…just stop okay? We don’t want to be entertained by your stupid program, we are NOT gonna find it “fun” because that is what we have games for, okay? We just want to do our task as cleanly and quickly as possible with as little hassle as possible and I have yet to see any of this junk that keeps getting added everywhere do anything but slow everyone down.
Yeah, dream on. For a while now it’s quite certain we’re long past Ballmer’s developers x 3 point in time (regardlessof his state of mind it really was a fairly good dev. period), and you can’t be surprised to see almost all UIs targeting people who don’t really do tasks, and if they do, their importance is hardly so high not to appreciate the “beauty” and “usability” of the wobbly magic of the new generation of useless and dysfunctional UIs.
You, sir, are spot on!
An interface should be pretty enough to not be noticed as ugly, but also simple enough to not be noticed either. In fact, the aim of an interface designer, much like the aim of a security admin, should be to have his/her work being noticed as little as possible.
I love it when my users’ bug reports, comment, feature requests are about functionality and content, not about the interface itself; this means I’ve succeeded in my UI design.
On another note: “fun” is totally subjective, and this he uses as a base for his whole argument. Fail.
This.
Proper UI design takes the memory of the user into serious consideration.
Oh, he worked for Spotify, btw? Not really a well designed application anyone. Not on my Windows 7 box at least. As bad as iTunes and Safari.
/rant
Well, Spotify presumably looks the way it does also so that the target demographic groups – large part of them supposedly already used to iTunes – can just carry over their experiences…
That’s not really the same thing, though. I agree that fluff for the sake of fluff is a waste, but feedback on actions is not fluff. And animations, if quick and to the point, are a good method for giving feedback: They are fairly ignorable, while also hard to misunderstand. And if those few hundred ms of animation (that does not keep you from immediately using another window) slows you down, well…
Edited 2012-04-23 09:50 UTC
How EXACTLY is this giving any kind of useful feedback? Your argument would have made sense on lets say Win95, where you had a complete new UI paradigm that people needed to learn, but who doesn’t know that programs are going to the taskbar? Even adults who have NEVER used a PC before understand this because there is an icon and the name of the program sitting right there in front of them where it wasn’t before. And I actually know of which I speak because i recently taught a 53 year old man who had never used a PC how to use both it and the web.
And the simple facts are if it were JUST this one thing? well that would be fine, but in Win 7 for instance we are talking about TWENTY different desktop effects! Its like MSFT and services, if you are running on mains with some monster hexacore then frankly you won’t care how much crud they add, but when you are talking mobile devices all that bling bling sucks cycles and hurts battery life.
Honestly I wouldn’t even care if MSFT made it easy for the end user to choose, but they don’t, they leave it buried under 3 sub menus that the average user will never trip over and with ZERO explanation as to what these things are and what they do. If they gave the user a little wizard at first startup, or even one in ctrlpanel, something that explained what they are and asked if they wanted them? Again not a problem. but all this needless bling added without asking the user and without an easy and simple way for the user to understand what it is, where it is, and how to turn it off is simply inexcusable IMHO.
Well, they did change the panel in Win7, if only slightly. And if nothing else it’s useful if you have a million open apps, in that it allows you to see and then remember where on the panel it minimized to, instead of having to scan through it (ref. the eternal discussion about spatial memory and file managers).
It’s also really not that intensive. The GPU power needed for that scale-zoom effect is something tat would hardly tax a voodoo 3, and I’ve seen claims that aero can use less power than a 2D interface because it uses dedicated and more efficient display hardware for its operations compared to using CPU and 2D operations. Sounds plausible to me, but I’m sure someone has tested it.
And you can mostly disable it; unchecking “Enable transparency” seems to stop both transparency and animations for me.
Just my opinion but bradster should the current iteration of iCal to his list. Not sure who thought the leather look was something that would be appealing to the masses…
No kidding. I have no idea what happened with iCal, but I suspect someone was on drugs. It looks SOOOOO horrible. I can’t imagine anyone actually liking it that way.
Not only did you manage to write an article I fully-100%-totally-absolutely-wholeheartedly agree with, but you also used a word in a title I actually had to look up first, only to find out that that what I hate most about consumer software design, actually has a name…
(And to add insult to injury, I can’t correct my stupid typo in the title. This is not my day.)
Edited 2012-04-20 18:04 UTC
Is not the whole concept of the “Desktop”* an example of a skeuomorphic interface? I think that in moderation it is a cleaver way to allow users to naturally interface with a new paradigm, quickly and easily.
*I grew up with the Amiga, so I had a “Workbench” metaphor rather than a desktop…
Edited 2012-04-20 18:39 UTC
I think Desktop started out as such poor attempt of skeuomorphism nobody has ever realized it until it is pointed out that its supposed to represent an actual desk.
Come to think of it, its this colossal failure as skeuomorphism that makes Desktop work so well.
I’ve always envisioned the desktop on any OS as an always-open folder, a place to drop a file I’ll only be working with for a short time. Call it the Desktop, the Workbench, the workspace, or whatever, I only use it for that purpose so I visualize it as a huge manila folder laid open…on top of my “desk”.
As an aside, I’ll never get over the reaction I got several years ago when I was training a first time computer user (retired man in his 70s) and I told him to click on the icon on his desktop. He began to shuffle papers around on his desk, asking me what the hell an “icon” was and what did it have to do with his computer? It should go without saying it also took me over a week to convince him that the mouse should remain on the surface of the desk and not pointed at the screen like a TV remote.
My screen looks nothing like my desktop. To be specific that’s what I currently have on it:
– a laptop (recursion warning!),
– a computer mouse,
– some books (books, not icons of books),
– my cellphone on a charger (again, an actual phone, not an icon),
– a cup,
– cables.
On the on-screen desktop I have:
– a pointer,
– a panel with some buttons and gadgets on it,
– some movable rectangular windows displaying more buttons/UI controls and changing content,
– some icons acting as shortcuts to applications/documents/directories.
No, they aren’t the same thing, in fact, they are as different as they possibly could be. Computer screens just happen to display 2D images so it makes sense to construct user interfaces that way.
So, we are left with just the name. But that was just someone’s arbitrary decision to call it a “desktop” (for a good reason, we don’t want to describe it every single time we mention it). After all, just because someone else called a pointing device a “mouse” doesn’t mean we should now be comparing them to living mice.
This is so true. Living mice do not have tails attached to the front!
It might as well be the back of computer ones… as far as they are concerned. In fact – who hasn’t seen, at least once, some computer mouse with trademark snout and whiskers drawn at the user-facing end?
(and now I’m thinking about very unethical project of trying to fit computer mouse innards into stuffed not-anymore-living mouse; but hopefully me being perfectly aware of too small size of the latter will stop me from ever trying / luckily it’s not “computer rat”)
From the opposite point of view: See the “Realworld Desk” (including “screenshots”) at The GUI Gallery!
http://toastytech.com/guis/desk.html
Spot on. A crude and simplistic version at that, because on the earliest systems carrying GUI’s, resources were extremely limited.
These days, we have oodles of resources available to display so much more than these early UI’s, so more refined and realistic versions become possible.
Skeuomorphism is one of my favorite words, but it is also one of my least-favorite things. The first really bad example of it I remember was Quicktime 4(?), with that damned wheel for volume.
Whoever thought that was a good idea should be dragged from his cubicle and beaten like a crappy office printer.
The largest problem I see is that the real world part is only implemented as far as it isn’t very hard to do. Yet the harder things aren’t implemented at all, or in such a way that it is half analogy, half cyberian.
Take something simple like a PDF or eBook reader. On real books, I have an edge where I can immediately go somewhere near a page – It is the edge opposite the spine. Also sometimes the spine gets cracked so my favorite pages are stickier and I can feel or see them.
There is nothing like that on any reader I can find. There is some way of navigating, usually involving scrolling – somehow I can read the “book” as if it were on a roll of paper and have to drag/spin it a great deal to get to something near the end. Or switch modes and find a seek box.
Pick ONE paradigm, and floodfill all the features that make sense into it. If something important doesn’t fit, switch paradigms, or admit you will have a clunky and hard to use but pretty UI.
I think whether you’re an advocate or detractor of Skeuomorphism, I think your focus is wrong. When designing user interfaces, we should strive to make them obvious and unsurprising. Whether that requires an entirely novel UI element or something that resembles a physical object we’re familiar with depends entirely on the application, and we should not be afraid to mix themes. In a similar vein, although I like pretty interfaces, I will advocate making someting uglier if it makes the app more usable. This allows us to make “nonsensical” (as in stylistically incompatible) combinations of UI design elements when they have clear benefit.
With regard to Photobooth, I agree with Apple’s objective here. Yeah, it looks kinda stupid, but if an application is a toy, then it should have balloons and bunnies in it. The app becomes an environment with a theme that has a certain desirable emotional impact.
For musical programs it is a tradition to mimic real things.
But still not the only tradition (trackers, most notably; and I believe that graphical signal flow editors look more like “digital” diagrams, typically not attempting very much to look and feel like manipulating cables and plugs on screen)
Thinking about this, there’s actually good reason to have skeuomorphism in a application UI. For users that are unfamiliar or intimidated by computers, having a UI that familiar to a physical object that they are likely already familiar with lowers the fear factor and encourages more use. We as experienced computer users tend to forget how often people can be terrified of “breaking” their device just because they wondered off of a known and familiar path.
While you do make valid points about OS UI consistency, the mindset above already has to deal with vastly different ways of interacting with physical objects already. OS UI consistency matters less in this mindset than do keeping things familiar. When skeuomorphism wasn’t so popular, we heavily relied on UI consistency to keep us grounded until we mastered the interface. Skeuomorphism simply takes that perspective and points it outward to the physical world instead of the digital one.
There’s a big problem with looking at things like this.
It won’t last.
While growing up we may have used something like a tear-off calendar, or a legal pad to keep notes, there is an entire generation of users that never have used these physical objects. They’ve used the digital counterparts only and have much less of a barrier of fear as do older users. As use of these physical object falls away for their digital counterparts, we run into the flaw of skeuomorphism.
Skeuomorphism isn’t a fad, or infantilizing, it’s a phase.
You know, like your parent’s used the phrase, “It’s a phase”. Replicating physical objects in digital UIs helps the most people who are familiar with the tear-off calendar or legal pad. The younger you get it becomes “cute”, or “fun”, but that supposes that they even know those things exist. In a world where someone has never seen a tear-off calendar or a legal pad, it just becomes noise that unnecessary bloats up the application.
Apple is desperately clinging to user friendliness and the crowd that would find skeuomorphism enjoyable. Eventually, it’ll become the better choice to drop these concepts because they get in the way of using the machine. WinPho7 and Android are already doing this. I suspect that with time, impersonating physical objects will hold less and less appeal with iOS devs.
And I believe it’s the reason people were so happy with Gnome 2. A HIG and consistency. I really like what the elementary guys are doing in this department. I sometimes wonder what a GEM UI would look like today. The simplicity back then really made it shine in regard to clarity.
As for the iPad sales: It looks like a toy and it acts like a toy. That’s really ok! Just like it’s ok for a RPG game to use a gothic font.
But for a desktop environment it doesn’t work. A DE should follow guidelines, get out of the way and enable users to quickly adjust to new applications and dive in to get their work done. That’s why I’d pick the Mockup UI with the keyboards anyday over the piano interface. I expect it to have other advantages as my piano (e.g. looping, multichannel, whatever…) and therefore it should be fit to represent those and not replicate my piano 1:1.
Edited 2012-04-20 20:58 UTC
“When you really break it down, there’s been very little innovation going on in iOS’ user interface since its original inception, and this doesn’t bode well for the future.”
Or Mac OS X’s interface, or classic Windows… Apple have basically been flogging the same Aqua interface since Max OS X’s conception in 2001 (11 years!) and they don’t have a problem shifting Macs. And what about Microsoft? Every released version of Windows since Windows 95 has the basic GUI layout and experience, and all “themed” versions of Windows (XP and above) can have the theming disabled to be brought back to a Windows 95-esque interface. Microsoft has been flogging that horse for 17 years!
Now I know Microsoft is trying to reinvent the GUI with it’s clean and stylish Metro interface, But people aren’t buying Windows Phones and people aren’t responding well to Windows 8. Most users ARE dumb and will buy products because they have pretty graphics, not because the GUI works well. The fact that Apple can’t shift iOS devices quickly enough to keep up with demand is proof of that (I recently bought the last black 64gb iPad in stock in the whole of Cambridge, UK), as is the fact that Windows Phone 7 has smaller percentage of mobile phone market share[1] than Linux does of desktop market share[2].
Not everyone is a power user, and most people who buy into computers neither know nor care how a computer fundamentally works. If it helps them do it and looks pretty whilst they’re doing it, they’ll buy it. If it’s blank and sparse, they might not.
[1]http://www.pcmag.com/article2/0,2817,2398695,00.asp
[2]http://www.pcworld.com/businesscenter/article/252552/if_desktop_lin…
Edited 2012-04-20 22:19 UTC
Thom, don’t make me give you a time out!
For me it is the faux materials more than anything. The 3d keyboard looks fine but the fake leather, etc … What are they thinking. Makes me think trashy 90s microsoft.
The Lion Photobooth looks so…cheesy. it certainly does not make it clear that the intended purpose is “to fool around and have fun”. It does tell me one thing though: “Use another application, any application, NOW”.
Many good points in the article. And yes the inconsistency in Apple’s iOS is confusing. I can cope with most of it on the iPad, but iCal really makes me mad.
So Tom’s central agrument is that ui elements that look like real things are confusing and Ahlin’s argument is that they are more fun. They are BOTH right, more or less.
To be honest I do find some Skeumorphed interfaces fun on handheld devices, but I would find it confusing and terribly irritating on a desktop matchine, especially if I had work to do.
I think (IMHO) that skeumorphed ui work in handheld devices because a smartphone fits in your hand, and you feel the weight and size of the interface you are using. When I am working, an inconsistent skeumorphed ui on the desktop would drive me bonkers. So you are both right in a sense.
This is why I prefer to use CLI as much as possible. My colours, my preferred font, my preferred #333333 backdrop, everywhere!!
Seriously though, good write-up, Thom. I agree with almost everything you’ve written.
You might be surprised just how skeumorphic your car actually is, Thom, even after we’ve been using them for a century. That surface in front of you that holds all the instruments? It’s called a dashboard. The original dashboard was a simple plank on which the driver of a stagecoach would put his feet so he could brace himself when the horses made a dash. Sorry, my Dutch is too rusty to know if that works in your language.
Does your car have speedo and tach dials that move in a circular fashion? They WILL move clockwise as road or engine speed increases. Why? Because that’s the way clocks have moved for hundreds of years. A decision made by a clock maker hundreds of years ago still influences designs today. Or do you have digital bars, well, they WILL extend from left to right to indicate an increase, never the other way round. Just like they did on steam trains in the nineteenth century.
Another example: luxury cars have long bonnets with the front wheels way out front. There’s no real engineering reason for this. But that’s where the horses used to be and rich people had more horses pulling their carriages than poor people. By all means try to market a luxury car with the wheels directly in front of the driver. It won’t sell.
But back to computers. The real problem with skeumorphism is that it constantly runs into obsolescence. We see this in iPhoto for OSX, for example, which used to categorize photos into “rolls”. As emulsion film cameras gave way to digital cameras, more and more consumers asked “what’s a roll?” and Apple had to change it to “events”. Strangely, the Photos app in iOS still refers to a camera roll.
The same thing will happen to agenda apps that resemble filofaxes. “a filo-what?” Skeumorphism is something that constantly needs to be redone as digital replaces the analog world it initially tries to imitate.
Now can we talk about the most glaring example of skeumorphsm of them all: the QWERTY keyboard?
That’s not what skeumorphism is all about. Going through your examples:
– Dashboard – just a name. No one is putting legs on it anymore, at least not while driving.
– Clock and dials – that’s a convention. No one is trying to imitate an actual clock (twelve digits, two hands, tick-tack and a woodpecker)
– Long bonnets and wheels in the front – leg room and a big engine.
– Filofaxes – no idea what’s that. This analogy wouldn’t help me at all.
– QWERTY – again, a convention. Physical keyboard may look similar to a typewriter because of similar physical constraints but the last thing I want from an on-screen keyboard is to render a photo of a keyboard (at least not while they lack tactile feedback).
This helps prove his point.
That’s good. It means we agree on something.
A picture paints a thousand words:
http://en.wikipedia.org/wiki/File:Myfilofax.jpg
OK, I’m sure you see the inspiration for dozens of skeumorphic PIM apps there.
That’s not the point. If this was about efficiency, we would all be typing on the far more efficient Dvorak layout now, on tablets, on laptops, on everything, not on a QWERTY layout that was developed for nineteenth century typewriters. The QWERTY keyboard layout perpetuates an illusion that we are still working on an ancient typewriter. And we all go along with it. Resistance to learning something new is a very strong factor even among geeks, and I don’t exempt myself from that!
http://en.wikipedia.org/wiki/Dvorak_Simplified_Keyboard
If anything, I am arguing that the general definition of skeumorphism is too loose and not radical enough. It goes far deeper than just how photo-realistic a depiction of a real-world object it is. If you don’t analyse those deeper levels, then no matter how abstract and non-representative you make the depiction, the skeumorphism is still lurking underneath and waiting for a graphic designer to pretty it up and make it look like a real-world object again.
The question is not whether the virtual keyboard on, say, your iPad is a photo of a typewriter keyboard (actually, it is damn close to a photo of an Apple wireless keyboard, but never mind). The point is that as long as nobody asks the question “isn’t there a better way to do text entry?”, the possibility exists for someone to make that photo of a typewriter keyboard and paste it in there. And as long as that possibility exists, someone, somewhere, is going to do it.
And if that “better way” does catch on, watch out for someone to make a skeumorphic representation of that, ten or twenty years later! This is going to be a long debate.
There is an obvious point the pro Dvorak camp is missing. QWERTY was designed to space out commonly used keys on a typewriter but it is still useful today because it provides exactly the same function for peoples fingers.
QWERTY keyboards allow the user to spread their hands out in a more natural position where they are too cramped on Dvorak keyboards.
Only when you have a very small keypad and using a single finger to type on does it start to make sense to redesign the keyboard and even then not always. I use swipe for instance on my phone and it determines what I am writing from the shape that I draw so there is still an advantage to separating common letters.
There’s also http://en.wikipedia.org/wiki/Dvorak_Simplified_Keyboard#Controversy section and its links*… Or at the very least – overall efficiency is what matters in the end (most people don’t even touch-type after all)
*or more succinct summary of sorts, of one : http://wwwpub.utdallas.edu/~liebowit/harvj/harvard.html
V. EMPIRICAL EXAMPLES OF STANDARD CHOICE
A. The Fable Of The Keys
Plus Dvorak layout is a bit of joke, vs. internationalisation & our modern very connected world, in its quest to be supposedly very optimised for English …apparently it’s community thinks it’s bad enough for other languages to warrant language-specific Dvorak variants.
Still, my 1st language (of 40+ million speakers, with more letters than EN) doesn’t seem to have its layout.
My 2nd language has… more than one Dvorak layout.
There’s enough of a (mild) mess with QWERTY/QWERTZ/AZERTY.
Edited 2012-04-27 23:16 UTC
My wife really dislikes the new iCal… I bonded with the author at that mention.
I don’t think OS UI’s should try to be anything but what they are… on-screen displays of information… they need to find their own navigational paradigm.
yep, I totally agree that his keyboard mockup is the better one, it just looks worse than it could because of the hideous color of the keys.
The problem with Skeuomorphism is that is very difficult to do right. I completely agree with the author that it has a lot of value, the problem is that 99% of designers are not good enough to do it right.
The paper app on the iPad is beautiful, and it’s just way more fun to use because of the skeuomorphism. I also enjoy the calendar app on the iPad and I don’t think any usability is taken away by the fact it uses some textures.
The reason why most skeuomorphic apps are hideous is not because of the graphics, but because of the non-standard controls. If you look at this screenshot, http://tobiasahlin.com/wp-content/uploads/fmfSmall.jpg it doesn’t actually use any different controls than any other iPad app. They’ve textured it, but everything works as expected. No one will have a problem operating that app, you don’t have to learn anything new.
The problem comes when people invent their own controls, which work worse than the standard ones without fail.
Photobooth and the keyboard are great examples. iCal is taking it to an unholy and criminal extreme. People often wonder what the big deal about Apple software is (as the same *functions* are available in competing platforms), it’s that ability to invite/inspire users to use the software. It’s subtle but it’s there and it’s one of the factors behind Apple users occasional Kool-aid-drinking behavior.
There is a similarity when you make a User Interface to get a similar effect as an uncanny valley.
If your UI is too abstract or basic. Then people really don’t get it much, and are intimated when they look at the screens and will react poorly to it. Then as you make it feel more real people begin to like it… Then you get into an area where the UI start getting too close to matching real life and then you start not liking it much.
Animations are good in UI to help show your eye where something went. A good example is the Genie effect when minimizing a window, But a shrink down works just as well. It doesn’t just vanish and a button appear, But it goes down to where that button is and you know where it went to… The same thing with Shadow Effect, It is actually much easier to find your active window based on a good shadow then the color of the title bar. However when you get to showing a real paper look or different methods any little glitch we will catch and have problems with it.
Ultimately, it’s just the wrong path to go down – it’s a red herring.
What technological progress has always been was to make things work the way our MINDS work. Trying to make things work like they do in the physical world only works as long as the physical object was designed well enough to work like our minds like them to work.
With human-computer interfaces, we have a greater opportunity to emulate the way our minds work directly rather than by the proxy of pre-designed physical objects.
Though “make things work the way our MINDS work” is a slightly double-edged argument – in some ways, partly, it can also go in favour of the discussed approach.
After all, our minds evolved, over a looong time of half+ billion years, to the way physical words works.