I have found the flat design incredibly difficult to use.
In particular, the new Apple Music app was almost completely undiscoverable for me. It wasn't until I was reading patch notes that I realized that it had a key feature that I had been searching the UI for more than a month. That feature was to show only the music available offline, and it's hidden behind a down arrow next to a heading label. (Rather than the ... that other menus have).
Apple has never been perfect at UIs, but they've always been better than this in my experience. Deciding to hide key information and be as cryptic as possible works great for designers that already know the UI, but it works terribly for users that are still learning it. This type of elementary mistake, all too common by those who are deep in the act of creation, is best corrected by stepping back from the problem and approaching everything with the mind of a beginner. That or giving the device to an outsider and observing them, good old fashioned trials.
That's what UX needs these days, not more fashionistas trying to remove data and UI cues. The industry needs a big wakeup call. Mobile and even the web (like Google Docs) are becoming a churn of bad experience.
I've had a lot of trouble as well. A constant source of frustration is the seemingly unnecessary split between contacts and messages. I don't understand why they're not under one roof and instead frustratingly claw myself around to get to the right section. If I want to send a message to someone why do I have to exit messages to scroll through a list of my contacts? It's ridiculous.
Another one is that for the life of me I couldn't figure out how to access the search for app feature until recently. I knew it existed, and I guess it's a downwards pointing tap in the middle of the screen but it took me forever to find it again and then to actually figure out what I had done to trigger it.
Then there's the loading my phone with apps I have no use for such as Music, Playbook, and the Apple Watch app. I understand these have useful features to people but I don't understand why I can't at the very least hide the damn icons and instead have to either create a folder for "Shit I don't use ever" or delegate them to some far off slide screen.
It seems like it goes on and on and while the old iOS's did look a bit clunky, everything was very clear in my opinion in terms of functionality and I'm not exactly sure why there was a move away from it.
And this is all a shame since Apple's hardware software integrated and closed source environment is a terrific way to make a phone in my opinion where I barely have to worry about malware since apps are checked by their App Store team, everything looks standardized and decent because of the components they provide, and things run really fast (at least until your phone is two generations old) because of their language support and hardware integration.
With regard to “why do I have to exit messages to scroll through a list of my contacts”: are you aware that you can start a new message and either start typing in the “To” field or tap the + button to bring up your list of contacts? Is there something else you’re looking for?
I agree with your general point on the increasing lack of discoverability, but this feature, in fairness to Apple, is "hidden" behind basically the only graphical icon in the Messages app, on the main screen, of a pen on paper, practically screaming to be touched, and once touched, jumps to an email-composer-esq screen with a "to:" text box with the cursor flashing in it.
I have to say, I am a bit surprised I never touched it. It truly is the only icon there, I dunno what I could have thought it meant, but I guess it never occurred to me that it would bring up contacts.
For what it's worth, you also don't need to do that if you've ever had a conversation with them (since deleting your message history), since you can use the conversation search field in the app's home screen - though some users may not know that scrolling up from the initial view of a scrolling list in iOS sometimes reveals a search field. And this also works with non-contacts and group conversations: you couldn't just change the home screen to be a contact list, you'd have to combine them with those other two types of conversation, which sounds more likely to cause confusion than the current setup.
See, and that does work and I know about that feature somewhere, but there's no indication where that works. So if I know it works in one place, it doesn't occur to me that it might work somewhere else.
Agreed on the new iTunes and iPhone Music app. I had to re-learn how to shuffle songs, and how to pull up the album of the song that was playing the other day. Extremely non-intuitive and hidden behind 4-5 interactions.
All for the sake of cramming in new paid features that I'll never use - "For You", "New", "Radio", and "Connect". Why Apple? You make enough money from iPhone sales already. Is Apple Radio really going to add that much to your bottom line?
I probably spent 5 minutes trying to disable shuffling in iTunes the other day. The icon appeared in several different contexts where clicking on it did nothing.
It's just incredible -- stunningly, unbelievably incredible -- that Apple allows this turkey of a Windows application to ship with their name attached.
Yeah, the thing I miss most from Spotify is the "Recent" menu, which was basically just a list of the last 20 - 30 albums you listened to. It really clicked with how I (and a lot of people) listen to music - we tend to repeat a certain subset of our music depending on the mood. This is the rare case where I feel like Apple has barely tried to look at how customers use the thing.
You can make the 'Connect' button into something a little more useful. In the Settings app, go to General > Restrictions and turn off Apple Music Connect. The 'Connect' button will then be replaced with 'Playlists.'
Thanks for this great tip. Again, another totally non-intuitive way to make Music 10x better.
Tangent: I wonder if people who work at Apple browse HN and see these kinds of complaints regularly? It would be weird to work at a big company that gets a lot of press on HN and stealthily get direct user feedback from a tech forum.
For me - yes. I had never needed to use "Restrictions" before, and it would seem much more obvious to be able to customize the display of the application from within the app. It also doesn't make all that much sense that by turning off one of the tabs, it would be replaced with an actually useful tab - playlists.
I don't work in UI/UX and I'm not an iPhone power user, so perhaps it's just me.
You have to have some reason to think a setting might exist before you'll go looking for it. If it's unintuitive to think that a particular piece of functionality might be controlled by a setting at all, the fact that it would be easy to find the setting if you knew it existed is of little help.
There is an obvious line between being non-intuitive or unobvious and reasons to think something exists or not. A UI cannot rely solely on intuition alone; investigation must be encouraged too.
>> Is Apple Radio really going to add that much to your bottom line?
They will try to use radio in 2 way:
1. Make it unqiue and hard to copy, to add one more reason for people to use their device.
2. When making content deals, it's hard to get good deals if you come without a large crowd of subscribers.So apple plan to to use your radio subscription as content negotiation tool, and than upsell you a tv service - and there's a lot of money there.
Worse still, since the latest Music update, if there's a way to play through an artist's entire discography, I can no longer find it. And I've been hating for so long the lack of ability to sort by ascending album year so I can match my laptop's iTunes organization.
I cannot tell what kind of repeat is on, or if I hit the target at all. There is no button feedback in Music to indicate my touch actually registered. Hell, it may be my iPhone touch sensors are fucked, but I usually have to hit the play/pause, backward/forward buttons multiple times to register a tap. It feels like the hit boxes are smaller since iOS 7—never had any problems playing/pausing or executing prev/next song actions. Now, it ignores my taps at least 50% of the time. Curiously, the new heart button likes to swallow all my taps meant for the prev/back button.
Music has become the biggest piece of shit since 7 landed. It gets worse every release.
Wow, I thought adding music to your library made it available offline - just realized it didn't (I probably should've noticed)
Even with your cue it took me awhile to find that setting, which is included with the sorting (Under My Music/Library, Arrow is next to the default sort of 'Artists')
The Maps app has been a miserable experience as well. Fortunately it seems a bit improved in iOS9, but it's taken 3 years! It wouldn't be so bad if there was a way to set an alternate app as the default handler for maps/directions.
The problem is that UX is mixed with and influenced heavily by (often simply dominated by) design, and design comes from the creative arts world. The creative arts have, since the end of the Edwardian period, and especially since WW2, stressed stripped down minimalism and originality. A lot of this comes from an innate hatred of the bourgeoisie, some of it is simply fundamentally anti-humanist Platonsim.
It worked fairly well for a lot of industrial design, but has been a disaster for the built environment and, we are learning, software.
what about Holo theme? The action overflow on the top left, and row'd list of features and options was clean, intuitive, consistent, and was by no means confusing.
I call Google's the Messy Desk Layout because it forgoes the function, order, and organization offered in a digital system to emulate the messiness of the real world. Stacked cards, the paper-layers-sliding-motion as you scroll down an app's page in the Play Store on your Android, and especially shoving all your Chrome tabs (my friend [...has problems, ok, but still...] regularly rolls with >50 chrome tabs open on his phone) into the Recents switcher (and poor guy isn't smart enough to figure out how to disable that feature in Chrome)...
anyways, my opinion is that tacit knowledge trumps nearly any level of design you can come up with. Users never look at the label or the picture of what they're interacting with, they remember it's spacial location in relation to the screen borders. For that reason, a good design should be chosen, and iteratively modified-- occasionally, but things should roughly always be in the same spot as before. This business of rewriting the entire presentation layer every year (Youtube app for example) is very mentally taxing and I have given up trying to learn how to use the new one. I suppose I'll figure it out eventually, but I can more productively spend my energy learning a new language. It is, literally, too hard for me to relearn everything to care about figuring out whatever the hell their UX engineers most recently pulled out of their ass.
The thing that really, really angers me is I used to have a keen perception of how a fluid, clean, efficient UI behaved, but as I've been forced to interact with poor designs I've lost that talent.
It's not just google, other products aren't designed as well as they were when I was younger, either. About half of my suggestions for usability improvements in google apps ('Send Feedback') have been implemented after I suggested them, and several within 2-4 weeks. This makes me cry at night. I had no idea what I had, and lost...
So anyways I basically don't update any apps anymore (they're always removing features or moving them anyways, why would I update if this one works fine?), and if I do and find the new interface too confusing I just track down an older APK online and make a mental note to never update that until I have the time and energy to learn the new layout.
bonus mentally-stimulating material: Windows XP's Explorer (likely back from the NT->2k days, and probably entirely thanks to IBM's role in NT development) had a great feature. You changed cursor focus in Windows Explorer with F6--but there were so many panes, so you would hold down F6 till you hit the one you want. Inevitably you were using List View, your hands were on the keyboard, and you were ready to start typing the name of the folder you wanted. The problem? There were so many panes that you'd always miss what you wanted due to the keyboard repeat rate. IBM's (or some other hero at MS) solution? Focus goes to main folder List View pane, to Link bar (ctrl+F6) above the folder List View, and then back to the List View pane. But why?
...because whoever made that decision knew that you were always going to miss it the first time, so they placed it twice so that you'd always hit it...genius...
As far as I can tell, these people don't exist anymore.
Same for Android 5 over 4.4. It was an awkward pivot, many regression, many gains. A somehow different way of doing things. I'm less and less fond of the generic and design trend, it takes few, yet far too much interactions to use my phone to... phone. But I forgot, these aren't phones.
Contrary to what the article says, it does explain why various UI elements are designed as they are – not just thoughtless promotion of aesthetics over interaction.
I think people often fall back on 'nobody thought about this and it's rubbish' arguments when the reality is often closer to 'they changed this and I don't like it' – the latter is a totally valid complaint, but it's also qualitatively different.
In my personal experience, I've not seen computer-naïve users of iOS struggle to a greater degree with iOS 7+ than with any of the previous versions. YMMV of course, but I think the extent to which it's a problem is overstated.
There are a couple of exceptions, of course – the reminders app has some stupid UI decisions that irritate me. But what software doesn't? How about that floating 'create a new document' button that was in Google Docs/Sheets/etc. until recently? Every single time I opened it, I had to hunt around for the button to create a new document because it wasn't where I expected it to be. But that doesn't mean Material is awful – it just means that it's a complex, long-term challenge to create a consistent UI applicable to a wide variety of applications. And I don't think modern UX is all that bad.
Amusingly you can change something in accessibility that adds a 1 and a 0 to the switches. For some reason I'm still unclear... Does pressing it make it 1 because I can see the 1 to press, or is it now 1 (yes). Android by default has off and on displayed. No idea how they make this work in languages where the equivalent words are longer.
I guess they stick to "on" and "off". After decades of electronic and electric devices labeled in English, words like those or "play"/"stop" are pretty well understood.
That's strange. The transition to the flat switches is one of the things I like most about the iOS 7 design. Previously, I could never figure out which way was on, but post flat design I have no trouble.
A real life switch is very confusing as well. How many times did switch your light switch without knowing if it was going to turn the light on ?
A switch is symetrical by design. They tried to remove the symetry on ios 7, and with the color it becomes more like "highlighted on/off". Which is better.
In my personal experience, I've not seen computer-naïve users of iOS struggle to a greater degree with iOS 7+ than with any of the previous versions.
Interesting. My experience is exactly the opposite: pretty much all my close relatives --from across the lifespan-- hate iOS7+, and still complain about its (un)usability fairly regularly.
Admittedly there could be sizeable a clustering effect going on, but to me it's been surprising nevertheless how my own family went from loving everything Apple to hating it with a passion.
"Apple used to lead the world in interface design" does not mean their designs were without serious flaws. Nobody would get up and defend System 7 as the pinnacle of usability, it was downright quirky and strange in places, and by the time System 9 arrived it'd gotten downright surreal. Things only made sense in the context of history.
The difference between Apple and other companies is not that Apple gets it right every time, but that Apple genuinely tries. Some other companies literally do not care how their products look, they just ship whatever the engineering team cobbles together with snippets from from Google Image Search.
Microsoft's making similar efforts lately, so that's encouraging to see, and even Google is making strides in reducing the amount of rampant ugly in their applications.
I would say that right now, Apple isn't really trying. Things have always varied from application to application, but right now it's a real hodge-podge. The tiny link targets in particular drive me nuts. I've been using the iPhone for years and sometimes I look at an app and I can't tell what is "click-able" and what's not.
I do think the loss of visual affordances was a mistake on Apple's part. But enabling Button Shapes does offset things quite a bit.
However, I do suspect that not all developers test with Button Shapes enabled. The native Twitter client looked a little "broken" for the longest time with Button Shapes enabled. It took awhile but I think it has been fixed.
> Microsoft's making similar efforts lately, so that's encouraging to see, and even Google is making strides in reducing the amount of rampant ugly in their applications.
I disagree regarding Microsoft. While Windows 8 was not well loved at least the 2 halves of its modal interface were internally consistent. Windows 10 looks like it was designed by committee by picking bits of Windows 8 desktop and Windows 8 Metro and stuffing them together. There's no clear direction or goal to it all. In addition there's just plain manifest ugliness like some of the icon choices.
Not only that but people using 2 products with the exact same function but different form not only prefer using the one with better form but actually encounter fewer difficulties and have a better experience. [1] Or this bit [2]. Read the entire article, but this is where those bits were from.
Soviet-era housing blocks are perfectly livable, they have a floor, a roof, walls, a toilet, heat, running water. What more could you possibly want in an apartment?
They were also brutally ugly.
When you have to live with something, interact with it constantly, you might want it to look nice.
“…users judge the relevancy of identical search results from different search engines based on the brand…Participants in the study indicated that the results from Google and Yahoo were superior to identical results found through Windows Live or a generic search engine.”
Why? Have you studied this? Have you read research that proves otherwise? When did I suggest form interferes with function? The idea that form and function are one says functionality and form are equally important. And there's good reason for this when humans are using your device.
I believe your bias comes from your position as someone who's work is create functionality. You see that as the complete product and by virtue of your intimacy with it, the most important. But you're missing the entire picture.
Start reading here [1] and I suggest you read the entire article.
There's nothing to study. What could you even mean by that? I'm sure there's an entire caste of useless "designers" who do all sorts of "research" to make themselves feel important, but in the real world people buy devices to perform tasks.
If that's not what you want it for buy some jewelry instead.
There's actually a lot to study backed by quantitive evidence. Your business will suffer if you only focus on the function and fail to consider form.
Do you think the Tesla Model S would have been a success had it the same design and form as previous attempts at an EV? The function of the car was far superior, obviously but the form was as well.
Both of these elements matter and if you only have 1 or the other, you lose.
I think that's the entire point of "they're one". You're not working with a finite amount of status points here. Giving to one does not take away from the other and if it is then you're doing it wrong. The idea is that the form does in fact matter. A lot. But form at the expense of function would be function follows form which, although a philosophy, is probably suboptimal.
What I'm trying to say is you don't say "lets make it work" and think of the form as an afterthought - as a "nice to have". We need to make it work and make it beautiful. Both are important and comprise the essence of what we are trying to build. Otherwise what's the point?
There was always a "time-wasting dial." The labels were always "styled with more visual impact than the actual data." All four different types of data elements were always "styled the same, with the same visual weight."
Thanks for posting the images side by side. It was tremendously refreshing to see the previous interface. To me the affordance on that is way way higher than the new flat trend.
Urgh. At least in my case, when I set an alarm, I already know what time I want and really don't benefit from any kind of visualization - so just give me a keypad and let me tap in the digits, something everything from the phone app to my microwave has trained me to do quickly, rather than confusing me with a one-of-a-kind UI element.
Luckily, I don't have to care because I just use Siri to set and manage alarms.
You do benefit. When typing the time, you have to tap at least four times. With Androids visualization, you can enter the time with two taps. I'd call that a benefit...
Imho, the problem is that flat design is intensely restrictive. Suddenly things that used to be available to freely design with have become UI cues... the designer can no longer play with color and layout and let the buttons stand out by button-bevel and the like... now the color and layout are part of the UI language.
Apple, being a design-oriented company, can't keep fiddling with individual app layouts, which means they can't work within the hyper-restrictive design language of flat.
Microsoft actually does much better with flat, I find, because I think there are less cooks in the "design" kitchen there.
Meanwhile, back in the 'I'm not superior to everybody else' mere mortal's world…
Discarding flat design as 'abomination' by 'idiots' is rather missing some of the wider design context.
Take a look back as Swiss design from the 50s, for example. Lots of use of minimalism, bold colour, simple typography (Helvetica, even!). It's remarkably similar in some senses; one of the reasons it became popular was because of the perception that design had become fussy and overburdened by ornamental detail.
I think we saw the same process of design evolution with modern flat designs. There was certainly a perception among some designers that it was difficult to build modern interfaces in part due to the cruft which had been accumulated over the years. Look at an iOS 6 device, for example; even though it's been only a few year, it looks utterly baroque to my eyes now. Needless lines, textures, gradients and shadows that obviously served a functional purpose at one point, but with that purpose gradually diluted as the functionality and interaction patterns of software evolved. Building to modern requirements – with things like touch-oriented interfaces and responsive applications – can become much more complex with these constraints.
The thing is, design approaches will evolve over time. Designers, developers and UX folk will continue to tweak and adjust designs to offer a better experience. It's not always going to be right, or even visually appealing, but it's also a field that experiences constant change. Look at Google's Material, for example – it's not simply flat, but involves the use of animation, subtle gradients and shadows.
> I think we saw the same process of design evolution with modern flat designs. There was certainly a perception among some designers that it was difficult to build modern interfaces in part due to the cruft which had been accumulated over the years. Look at an iOS 6 device, for example; even though it's been only a few year, it looks utterly baroque to my eyes now.
It's funny, the more time goes by the more I miss iOS 6 and company. Comparing flat UI design to Swiss design from the 50s is missing the obvious difference that minimal Swiss design was about presenting information, not creating an interactive interface.
As the linked article points out, discarding affordances has a very real cost to someone learning the interface - it becomes harder to separate interactive elements from non interactive ones and harder to tell "on" from "off" (I STILL can't tell if iOS's shift is on by looking at it [1]). The desire to present information in a modern/minimalist way is at odds with this.
My personal opinion is that what you call "wider design context" is pure fashion. Yes, flat is fashionable, but it's just not as good at the job of communicating the nature of UI elements to the user.
[1] Edit: Just to add, Apple's difficulty with making the state of the shift key obvious is a microcosm of this overall problem. The expressive power of flat design language is so limited that indicating the state of a single element is problematic despite several attempts. iOS 9 finally solves this by changing the case of every letter on the keyboard to make it obvious.
Comparing flat UI design to Swiss design from the 50s is missing the obvious difference that minimal Swiss design was about presenting information, not creating an interactive interface.
Yes – they're different. Obviously so – there's 60 years between here and there. The point is to compare the minimalist approach, and point out that there is precedent for it. There is no absolute, guaranteed, 100% 'ideal' user interface, and it's something that we are still working on – iterating on design is how we do that.
it becomes harder to separate interactive elements from non interactive ones and harder to tell "on" from "off" (I STILL can't tell if iOS's shift is on by looking at it). The desire to present information in a modern/minimalist way is at odds with this.
The iOS shift key is a good example of where this had failed (and I do hope it is fixed). But here's an alternative take – presenting information in a minimal fashion can help to draw attention to interactivity, rather than distract from it, thanks to the elimination of visual noise. The point is, it's far from clear what the best approach is.
My personal opinion is that what you call "wider design context" is pure fashion. Yes, flat is fashionable, but it's just not as good at the job of communicating the nature of UI elements to the user.
I don't think that's a case, or that such design can be called a 'fad'. Design is still an evolving field – devices and the way we use them is changing, and design is no different. I won't deny that there's an element of fashion, in the sense that the popularity of design styles can influence other designers. But isn't that equally explainable as a desire to avoid completely 'off the wall' interfaces that will be unfamiliar to users?
FWIW, I don't find flat interfaces to be an issue. Metro, for example, was simple and elegant, while allowing for useful approaches to clearly displaying and exploring information. And the iOS approach, though it has inconsistencies, actually makes a lot of sense, with the 'content is king' and 'depth through translucency' approaches. Most of the problems with the latter appear to be to do with the expression of state.
> And the iOS approach, though it has inconsistencies, actually makes a lot of sense, with the 'content is king' and 'depth through translucency' approaches.
I think that's the key difference. The 'content is king' approach is great for consuming visual content but less good for applications that are not about content consumption but about interaction. If I'm reading a website, I want the UI chrome to get out of the way and not distract me visually - on the other hand if I'm trying to reschedule something in my calendar I want the options and interaction points available to me to be obvious so that I can make the change quickly and accurately.
> Comparing flat UI design to Swiss design from the 50s is missing the obvious difference that minimal Swiss design was about presenting information, not creating an interactive interface.
What about Microsoft, one of the first companies to bring flat design to phones? Their motto for Windows Phone 7 was "Content before chrome".
One interesting nuance here is that "content" actually refers to visual content. I think maybe that's why so many here are expressing their frustration with the Music app. The content in question is NOT visual, but the UI is still designed with the same UX philosophy that tries to emphasize visual content instead of favouring richer interaction.
For a comparison to a minimal design that has to support interaction consider the 606 Universal Shelving System [1] designed by Dieter Rams. Note that even though it's minimalist, every openable door/shelf/drawer has affordance that indicate where and how to open it - whether they're handles, holes for your hand or ridges you can grab onto.
Modern UI design has pushed past this stage of minimalism into something that's user hostile. Like those shelf doors where you have to push one side or the other to make the door pop out a bit and become actually grabbable - and it's not clear which side of the door needs to be pushed.
Take a look back as Swiss design from the 50s, for example. Lots of use of minimalism, bold colour, simple typography (Helvetica, even!). It's remarkably similar in some senses; one of the reasons it became popular was because of the perception that design had become fussy and overburdened by ornamental detail.
Those people were designing tangible objects, not computer programs.
At some point it's going to occur to someone that software is a different kind of thing that requires different kinds of controls. But it hasn't happened yet. We're still debating which archaic school of design we should be slavishly adhering to.
Microsoft literally founded the modern flat UI design movement with their Metro/tiles interface but like anything happens in the world of computer and electronics, it doesn't get recognized till Apple blesses it with their touch and get credit for in the process.
I've had conclusive data that a 3D-looking button was converting significantly better but have been overruled because of the "sleek", "clean" bullshit that passes for design these days.
That's a bit of a leap. There is nothing wrong with having your call to action pop, and a lot of thought is given by designers to position, colour, context, text, and texture, in order to ensure that it does.
Maybe a button that doesn't look that much like a button doesn't entice us as much as one that does.
It's funny to read this, because I personally find iOS to have by far the best UI of any mobile OS, and I consider the flat design of iOS7 to be a big step forward.
Other people feel differently, and they consider Android superior, so clearly there is a difference in taste.
The conundrum is that nobody seems to be able to clearly articulate why this is, and this article doesn't really help. For example, it says the timepicker is "awkward, time-wasting, inaccurate". Compared to what? Has anyone been able to measure how much time it wastes and how innaccurate it is?
In the end, this article takes a thousand words to say little more than "This doesn't feel quite right to me", and doesn't reveal any root causes. Do people prefer different UIs because of differences in finger size, manual dexerity, visual attention? Does Apple's UI cater to a specific minority of users? Do biological differences make it impossible to satisfy everyone?
If you think iOS 7/8/9 are hard to use, try Windows Phone 8. I had a Lumia 620 for a while and it was pretty hard to understand what was going on for some applications (aside from the fact that their back button behavior was something that was completely confusing). It's odd too because some of the non-Microsoft WP8 apps were very easy to use by comparison, but the Microsoft designed apps were the worse in terms of usability for me. So, I'm not sure if its the design aesthetics at play here or maybe just that some developers are better are conveying what certain UI elements do and what to expect from them. Either way, I think there has to be a middle ground between removing clutter from UI and avoiding what amounts to snow blindness for computers. In my opinion, the Material Design spec by Google is closer to the sweet spot than what I've seen so far (I think it needs more UI hinting of some kind and less empty space which makes it harder for me to focus on what's going on in an app).
> Minimalism in software is achieved by simplifying feature sets,
> not stripping away pixels.
<rant>
Simplifying feature sets should not mean reducing functionality. That is the lazy way. Simplicity should remove the extraneous, redundant, and inefficient; functionality is not that. Functionality with simplicity requires things generality, orthogonality, composability....
Good software engineers eventually learn to be able to do this for the code they produce (whether the business case allows it is a separate question). It should be possible for good UI designers to do likewise.
</rant>
Edit: I am not disagreeing with the article author here:
> It’s not minimalism to rip away the very things your users need.
Even if they do research it's a vastly different context. <70s was niche and rich. Nowadays it's for the mainstream. I recently learned that Xerox GUI had an OO copy function. A generic one, for any object. When Jobs and Gates took the GUI, they mostly brought the metaphor, not the underlying ideas.
Star also designed the mouse to be used with the keyboard, rather than instead of it. The generic object functions were arranged to the left of the alphanumeric section, to be used by the left hand to operate on the object selected by the mouse in the right. They had a column of wide keys for the main object operations (DELETE, COPY, MOVE, PROPERTIES) and a single-width column further to the left with secondary operations (AGAIN, FIND, SAME [copy properties], OPEN).
They also did not have an invisible clipboard. COPY meant what today would be copy+paste, and MOVE, cut+paste. If you wanted a 'clipboard' document it would be manifest and direct like everything else, and you would COPY or MOVE things to and from it.
IIRC in a demo they were copying a UI element, a whole panel or a window. It felt like a live object playground. AFAIK other OSes where limited to semi-type strings.
Nope, it can be used for general data. On OS X (since that's what I'm most familiar with) you can put any data you want on the clipboard, and even list the types that you can provide it in, so when it gets pasted, the target can specify which format they want it in.
For instance, copying and pasting files to duplicate them is a common operation on many operating systems.
Once a UI design is perfected, as is the case often, companies continue to look for ways to change it simply for the sake of changing it and so they can announce something new. This is true of both Apple and Google in the mobile phone space. Every single time, the new UI is worse because it hasn't been tested and no designer can think of everything. There was nothing wrong with UIs before the flat design. Flat design didn't solve anything. But once Apple implemented it, everyone had to have it. If you don't have it, your app is not "slick" and "cool." And yes, those subjective qualities matter way more than the quality of your app or really, anything else.
Oh, man, I have never wished so much that I could upvote a submission more than once. I wish this article could be pinned at the top of every discussion site used by web and app designers for the next... forever.
The only thing in the design world that I find more infuriating than the current trends, and the accompanying blandness and usability issues, is when people try to justify those trends as being somehow superior to what we had before using the worst kind of retro-fitted mumbo jumbo. At least let's be honest that most places have adopted flat design because it's cheap, easy, and quick.
At the bottom end of the market, making UIs a commodity is in itself no bad thing. For web applications, you can implement flat design in pure CSS, cutting down the bandwidth required for images. More generally, typical flat design elements are nicely scalable and Retina-ready, because everything is all done with such trivial vector graphics that no real effort or creativity is needed. You can adapt the simple layouts more easily to small screens as well. In fact, why develop anything original or even hire anyone with design skills at all, when you can just slap Bootstrap on it and charge the client an extra 200% for making the site responsive?
Unfortunately, for anything above the bottom end of the market, and particularly for promoting UIs that offer better usability and/or more distinctive styles, the current trends are awful for all the reasons this article sets out.
Not that Apple has perfect UI's (The clock dial is awful) but I feel like the ideas expressed in this article are that a UI should be designed as if the user is always using it for the first time. And that simply isn't true. With such a small screen a lot of things need to be considered, such as how easy is it to press a button without moving the hand and if this means sacrificing some natural ease of use - so be it. The user will still quickly learn how the UI works and adapt quickly.
While the old iOS look was getting a little cheesy by the time iOS 6 rolled around, it was indeed very clear for the most part. Wooden bookshelves in iBooks might not have had much function, but the shading and glassy look on controls certainly did.
As an example, coming from the angle of an individual who'd never approached a smartphone in his life, the function of the iOS 6 picker/spinner was immediately obvious. The shading and glassy highlights made it look like a real spinner and practically begged the user to interact with it. Distinct section separators made it perfectly clear that each section can be spun separately.
Compare this to the picker in iOS 7 and up. Not only is there no shading or highlights to suggest how to interact with it, but now there are no separators — even if one presumes that it can be spun, it looks like the whole thing would spin. To make things worse, oddly skeumorphic 3D perspective has been added into the mix, presumably to try to suggest spinnability, but without partner cues it's just confusing. With this design, so much is left unknown until the user attempts interaction.
I personally feel that Apple struck a nice balance between design modesty, usability, and aesthetics with Mavericks desktop, but that's gone with Yosemite. Interestingly though, El Capitan adds in subtle hints of shading and depth in a few places. I wonder if we'll see things start to tilt back in the other direction with OS X 10.12 and iOS 10.
Here's a secret. Apple has never designed truly wonderful user interaction. But they have been unafraid to say "no," less horrible than most others, and opinionated. These discussions don't benefit from a false narrative of them falling from a grace they never held. (Mac owner from plus to pad)
the dramatic loss of color and warmth when the new look came about took me back to the days when I had a PS/2 50z with the VGA gray scale monitor. While everything can look well defined it does at times look a little too stark; if software could have a dystopian air to it compared to what came before they did well.
I could not agree more with this article. Human brains are hard-wired to infer 3D structure from shading. Taking away shading means taking away 3D structure, means taking away one of the most important visual cues there is to help humans grasp interfaces.
Also, coming from all the print media, human brains are conditioned to detect and separate important/immediate from un-important/less immediate content by looking at overall page structure and relative weights. Deciding against bold or large fonts for the looks takes that away too.
Finally there is a ton of research pointing to the fact that fonts with a healthy amount of thickness is more easy to read than thin fonts. The debate is still on about serifed fonts I guess but using Light Helvetica for kind of everything quite certainly is a step in the wrong direction.
At this point, a lot of decisions that were once skeuomorphic are now a matter of tradition and user expectation. For a whole generation, the image of a floppy means "save" not because of it's physical history but because that's the icon everyone else uses for "save".
Abandoning choices like "button means clickable" and "colored and underlined means link" isn't just a move to flat design, it's an attempt to retrain users on software conventions that have transcended their physical origins.
Sure, the floppy doesn't communicate anything to anybody except as a cryptic artifact of the past.
But other design language, like buttons and underlined links, communicate based on the necessity for some sort of language. Changing this language to something else should not occur unless there's some tangible benefit to the change.
Language changes all the time, but normal language change happens in the course of two-way communication between people. UI language is more one-way, it's like a speech that's delivered to an audience. If a speaker delivers a speak using an entirely new language that the audience doesn't understand, there had better be 1) some benefit to the change of language, and 2) clear communication to the audience about what the new language is.
Apple, Google, and Microsoft have failed on both points 1) and 2) in too many cases, IMHO. Languages are fashion, to some degree, but they are also functional. Flat UI has in too many cases placed the emphasis on looking cool and pushing "forward" over the primary point of a UI, which is to communicate to the user.
This is a big part of why I've been reluctant to buy an iOS device.
Originally it was because of the requirement to use the first-party keyboard software, but now it's because they've changed the design language in a way that conflicts with everything but iOS - and as a result, the UI feels remarkably unintuitive and confusing, even if it makes sense internally.
My company does a lot of paperwork. It turns out people want to work on something that looks like "paper" or the final printed out result. The idea of an interface with buttons may be commonplace to us, but there are many people for whom this is equally impossible.
> There was never any evidence that a few decorative pixels hurt the user.
Is there any evidence that removing them does?
> The HIG wasn’t about aesthetics, it was about interaction.
> It was based on research, not trends.
It's not about interaction now? It's not about research now? In which ways isn't it?
I believe skeumorphism is just about aesthetics and trends. Well, actually I don't, I'm not sure what I should think I'm not aware of any research and can't make a good argument either way. Unlike the author I don't pretend I can though.
There might be a problem, it might just be a figment of the authors imagination. In either case articles like this one certainly are a problem.
That's a weak complaint – even if one of their products was not the best, it doesn't prevent the rest from being.
That aside, Apple's keyboards aren't bad IMO. I'm pretty happy with them generally – they have decent durability, enough key travel, and are relatively quiet. Certainly better than the keyboards provided as standard by any other major computer manufacturer.
I think everybody is entitled to their own opinion. Personally I see nothing wrong with the keyboards and actually prefer them to some regular keyboards. I can still type ~100+ wpm and the keys aren't particularly missing anything I need.
I've even bought extras for work just so I don't need to use any of the crazy things they have here. Long travel keyboards annoy me, I vastly prefer the chiclet style keyboards.
I have used the full size apple keyboard and MBP keyboard for years and like them both just fine. I have a friend with RSI and he needs a keyboard with a longer stroke, but that's a personal thing.
My pet conspiracy theory is that Apple is trying to kill desktops by giving its desktop users the equivalent of a laptop keyboard (and for an exorbitant amount too.)
> My pet conspiracy theory is that Apple is trying to kill desktops
I'm not sure that part is a conspiracy theory. Tim Cook just said directly that the maxiPad “is the clearest expression of our vision of the future of personal computing”. You're being hit over the head with the writing on the wall.
The author is not claiming that all of Apple's hardware products are home runs (I certainly agree with your take on the keyboard, and can't stand the Magic Mouse), but that some of them have been absolutely world class. Do you disagree with that?
Well the keyboard is the main device I interface with when I am actually using the product (besides the screen, which is just a flat panel). So one could hope that they could at least get that one right.
Or the perfectly round mouse? Or the iPod nano with its unusable, imprecise clickwheel and the lock switch so small and inset into the body that you can hardly find it without taking the damn thing out of your pocket? Or the iPhone with its single button on the front that to perform its various functions needs to be either clicked or double clicked or triple-clicked or clicked and held or clicked and pause and clicked again? (see http://blog.codinghorror.com/the-one-button-mystique/ ).
"so apple legal call was not a threat. it was a request. b.c. jony ive was personally offended by our soundboard. what world do i live in?"[1] - by the author Amy Hoy
The author wasn't around when MacOS was new, otherwise she'd remember that the things she thinks are obvious are just conventions that people had to learn. Easier than a CLI, but something to learn just the same.
Microsoft, Apple and Google are all pursuing variations of what can be called flat design.
There are many variations in flat design though. In the small details, Google and Apple have as many similarities as differences.
One of the differences is that Material relies heavily on shadows (one of the very few good design ideas from the first versions of Android). The most important buttons are usually elevated.
Color is also used to inform the user. The accent color of the app is used to color clickable elements.
I am currently in the process of implementing material design in a large scale mobile android app. So far it is going extremely well, the new design direction respond to many of our gripes with Holo and is good enough in order to be able to make a strong point with the product team in order to implement it instead of making a frankenmonster between iOS design & Holo like we did in the past. Time will tell how well it performs with our users.
I'd argue that MS did a great job with flat design in Windows Phone 7 because they were willing to sacrifice aesthetics for consistency. Everything on WP7 was stark black backgrounds, which made it really obvious what was a button - a colored box was always clickable.
The moment you let graphic designers in who care about making things look good, flat design falls apart because they want to change it up.
Flat works if you let the designers construct the design language and then never let them near the actual individual UIs and get tempted to deviate from their original language.
Youtube on a phone makes you realize why they're pushing 5.5"+ phones: https://i.imgur.com/2d0Neg2.png (yes, that whole large card is an ad, while top 25% of the screen is static)
Yeah, it's weird how non-vendor apps take their design cues from Apple, Google, and Microsoft but they tone down the obsession with empty/negative space separation which makes it wholly easier to use.
Seriously, I don't know why but every time I use the Google Youtube App it's hard for me to focus on what I'm doing. I'm not sure if it's all the empty space or the lack of any explicit separator between items (videos and what not). But it's just harder to use the app than to just open the web browser and get the video that way. :/
Actually no. Take a look at the Google clock app - it actually addresses a lot of the negatives in this article. The article argues that Apple is bad at UI design, not that flat & material are wrong.
Hear, hear. In exactly the same way that some of the current bad design decisions were made as an overreaction to the worst excesses of skeuomorphism, people often exhibit an unwarranted negative reaction to 'flat design', just because some of it -as in Apple's case - is so awful. It is entirely possible to adhere to the concepts behind flat design, yet still differentiate between a label and a button. Just because Apple can't do it, doesn't mean it can't be done.
Software is harder than hardware. Both are incredibly hard, but software is harder. Software has to rely on the hardware, that's why any truly serious builder of software deeply understands the hardware and Apple understands that, with limited resources, hardware is more important to get right.
You can always change software later. Hardware stays in the consumer's hands for years. Bad decisions there will go uncorrected, even if you take the extraordinary step of issuing a recall, not everyone will bring in their kit.
This dynamic means that every company that has limited means, that is to say, every company, is going to have shitty software. Because hardware is more important, so it gets all the design attention. But software is harder. And if you're not doing hardware, like Microsoft, then your software is still going to be shitty because you don't understand hardware.
It's why your VCR was famously hard to program, it's why the entertainment system in your car is horrible, it's why the old stuff that doesn't have any software at all is way more reliable than the new stuff that does.
The answer here is not to criticize Apple, but to help them out by engineering better software. I used to hate texting on an iDevice because I loved using Swype and iOS didn't have it. Apple responded to the market with Smart Keyboards, they're not perfect but it beats the old way. Now they have an ad blocking API for browsers.
Apple is the tech world's quality champion. No other company in the world can pump out quality like Apple. But again, software is really fucking hard. So let's work together here, folks. Show the world a new way.
Perhaps in support of your point, Microsoft's hardware products (keyboards, Xboxes, the Surfaces) are excellent. Despite being an ardent Linux user, I always buy Microsoft keyboards.
In particular, the new Apple Music app was almost completely undiscoverable for me. It wasn't until I was reading patch notes that I realized that it had a key feature that I had been searching the UI for more than a month. That feature was to show only the music available offline, and it's hidden behind a down arrow next to a heading label. (Rather than the ... that other menus have).
Apple has never been perfect at UIs, but they've always been better than this in my experience. Deciding to hide key information and be as cryptic as possible works great for designers that already know the UI, but it works terribly for users that are still learning it. This type of elementary mistake, all too common by those who are deep in the act of creation, is best corrected by stepping back from the problem and approaching everything with the mind of a beginner. That or giving the device to an outsider and observing them, good old fashioned trials.
That's what UX needs these days, not more fashionistas trying to remove data and UI cues. The industry needs a big wakeup call. Mobile and even the web (like Google Docs) are becoming a churn of bad experience.