I recently watched a 65 year old woman use an iPad for the first time. One thing I noticed right away about her interaction was she struggled to realize that many blue text words were actually buttons (see the Send button in this screenshot[0]). IMO, this has been one of the biggest design failures of recent Apple software.
Apparently the relevant design/usability term is "affordances", which are the "sensory characteristics [of an object that] intuitively imply its functionality and use"[1]. Affordances in UI design have evidently gone out of style, which effectively means that people are just expected to learn the arbitrary (and ever-changing) rules of how to interacting with computers (and other computing devices), instead of having the computing devices cater to how people expect them to work.
I think this also leads back to the idea that the touchscreen as a user interface is absolutely awful in every way except for the one huge advantage of its reconfigurability. (I forget the source for this, but there's a specific article I'm thinking of.)
What kills me is how car radios are getting touchscreens now. What good is a touchscreen in a car radio? The whole beauty of the traditional design is you can operate it without taking your eyes off the road.
I had a 2012 jetta that had this. It was complete touchscreen with only a couple basic function buttons. I found it extremely annoying to use. To make matters worse, when the flimsy screen got a tiny crack in it, I lost all control of the radio because I could no longer touch to interact with it. I guess that would be a big deal breaker from me if I had means to be in the market for a tesla. I'm not big on having everything in a touchscreen.
The lexus I recently purchased uses a mouse essentially to interact with the screen which I find pleasant to use and also offers a full range of buttons for the radio and climate control. I find this to be a much better experience for myself.
Still, the 'mouse' seems like a poor choice. As the op said - isn't the idea of big dumb buttons that we don't have to take our attention off the road to interact with them?
Go through the products available at a site like sparkfun or adafruit and tell me you couldn't make some awesome customization control panel system? Larger buttons for smartphone-specific functions, some with their own small (and low-cost) discrete display perhaps?
A modular control system, featuring 100% plug and play elements (knobs, sliders, mini-displays, buttons) mountable in a range of simple chassis (19"rack, keyboard size, mini desktop box, mixing-console-size control center) that can be configured by a trivial to use piece of software and thereafter control any device (PC, tablet).
I've been wanting this for twenty-five years.
There are things that do some of this, but not very well, and not very cost-effectively. There is software that simulates this on a touchscreen (grr). Anyone fancy a hardware startup?
I realize this probably isn't quite what you want, but have you looked at using MIDI? A lot of modern MIDI gear is quite cheap, connects to your computer over USB, and with the appropriate software can be used to control anything on your computer. There are MIDI devices with just about any kind of physical input you could want - pots, sliders, pedals, buttons, pressure-sensitive pads, optical and sound-based inputs, and much more.
There are a variety of development boards (Livid Brain, uCApps MIDIbox) designed to allow semi-technical users to build their own MIDI controllers. It's also reasonably straightforward to configure an Arduino Micro or a Teensy to work as a USB HID interface. If you can use a soldering iron and a drill, you can build a custom controller.
Absolutely. I've been known to break out the breadboard to that end. But soldering irons and arduinos are a level of too much... erm... wires, for what I'm thinking. Not to mention that making it control what you want takes time and debugging and perhaps a sympathetic operating system or app.
How about forgetting the electronics, and seeing controllers as just another UI problem. Software, implemented in plastic.
What would you want to build if it were that easy? Custom controllers for games? Custom games? MIDI Instruments? Dashboards with physical elements? Home / media center controls? POS interfaces? Hardware usability testing prototypes? Kitchen gadgets?
Maybe I'd be the only market, but I think it would be a big blurring of the line between hacker and maker culture.
I am one of the fools who bought an optimus popularis - it's optimus lame.
Also those relegendable knobs and sliders are very popular in the music industry already 0 - so it would be fairly trivial to bring it to a consumer device.
Not touchscreen related but monitors used to have dials for brightness/contrast control. But now we have these pain in the butt multi-function buttons that you gotta use to pull up the OSD, scroll through a bunch of menus and then finally you can adjust a setting.
Same thing with volume controls, used to be dials now it's buttons. Maybe I just like dials.
Aren't most affordances just arbitrary rules that most people can be expected to already know? I'm not sure that an adult introduced to computers for the first time in their live would figure out convex-looking buttons much faster.
Affordances aren't arbitrary. The idea behind affordances is that the tool fits the user, not the other way around. If something is meant to be gripped in a person's hand, then it should be the right size and shape for a hand to grip it. Conversely, if something is the right size and shape to grip in a hand, that intuitively suggests that it is for gripping.
True, but affordances on screen are a different story. There's nothing to grip there. There are just pixels on a flat surface, that may or may not simulate a 3D effect. And you operate them using a mouse, which itself is a layer of indirection. So I don't buy the concept that affordances in software are not arbitrary - they don't refer to somewhat fundamental concepts for human (like grabbing), they refer to the last 100 or so years, when physical buttons became a common element of the environment one lives in.
I see learning UI in computers as a process that begins with a new user having basic UI concepts explained to him, and then just following the changing trends, where a new trend usually makes "affordances" refer to the previous trend.
When my daughter was two and a half she could use the iPad pretty efficiently, because it was always pretty obvious where you should click. At nearly four she now struggles to click things or to even know what to click. It's a terrible regression for a debatably minor improvement in overall appaearance.
I'm glad it's not just me. Somebody made one of my websites beautiful, but with 'flat' buttons. People became confused on how it worked, and I had to go back and give the buttons depth.
Huh, thanks - I found the option, and enabled it for her iPad. At a glance, nothing much changed, there are just too many flat button-like things throughout the UI for this to help, but we will see.
This is one of my pet peeves -- designers hate styling links so they're underlined, but the underline provides a valuable UI cue to the user that the text is clickable/tappable/whatever. So stripping it out makes a site actively less usable.
(You can try to invent your own UI cue to take the place of the underline, of course, but then the user has to learn a new cue for each separate site, whereas the underline is still common enough for people to have already learned it somewhere else.)
At work they introduced a page once with links that were not only not underlined but also the same color as the regular text. I sent them a hoary old Nielsen article or two describing why that's a bad idea and people don't want to play Minesweeper with text but they dismissed the article as "too old" and no longer relevant to today's tech-savvy users. Then they had to change it within a couple weeks because no one could find the links.
"We're going to make a door that's perfectly flat and seamless, and we're going to paint it white just like the walls. It's going to be the most awesome door ever and people will love using it."
If web designers could actually stop and think about the virtual things they do in physical terms they might be able to see how foolish they're being, but I don't expect that to happen any time soon.
Stackoverflow was one of the biggest offenders to me. They would state that something was a duplicate of something else, with no link to the other thing. Except there was a link. I just couldn't find it without 'view source.'
I must confess I've been guilty of it too. I 'invented' a style of link whereby the link has a border on all four sides and is filled with a transparent pastel. I've even seen this around some (apparently it is kind of an obvious idea). But all in all, I've gotten away from trying to invent my own UI.
I'm an Android user since v2 and have watched it become so much less usable over time. The average person has no idea what is clickable and what is not in v5. This causes fear and confusion. My wife (aged 30 and intelligent) cannot reliably send SMSs on her Nexus 5.
Agreed, I loved the old iPhone 3 UI, since then, it has gone to hell. Android is also in that hell, with the "Flat UI", and my brain hurts trying to find what's clickable.
>> Agreed, I loved the old iPhone 3 UI, since then, it has gone to hell. Android is also in that hell, with the "Flat UI", and my brain hurts trying to find what's clickable.
The whole point of going with raised buttons and such was to provide a visual cue as to what was clickable. It seems UI people who missed the early 90's are getting to relearn this lesson. When I saw flat I just shook my head.
It really seems to go in cycles. We've gone from flat UIs (think ncurses-based) to using texture and shadow (90's - mid 2000s), and now all the way back to flat again.
I'll never understand why these designer types think that discarding all of that context is a good thing.
It happens in all areas of everything. I'm convinced. It comes in cycles, because of generational retirements/career transitions. Younger/newer folks don't have the experience to know what the older/more experienced folks do. The older/more experienced folks do a bad job of documenting their experiences for the next group.
I'm thoroughly convinced I'm correct, but have no data to back it up.
We should do a head-to-head competition between two Grandmothers, one armed with an iPad2 using iOs6 and another one using iOs8 and let them do the same tasks, and measure the time it takes to do those tasks.
IMO, Android 4 was the pinnacle of stability and usability in their lineage. I had a Moto X and 2012 Nexus 7. The upgrade to Android 5 on my Nexus was saddening; it was different for the sake of being different, and it fixed problems that didn't exist (and introduced many new ones).
Even 4.1 was much better than 4.4 - all the migration from text and contoured buttons to symbols. an open eye for "turn off snooze" on the alarm is indicative of what I think is really poor design.
I haven't used v5 so I cannot comment on that transition. However I was around for the transition from v3 to v4 and I'm pretty sure the common sentiment was it was strides more usable. Also, without malice, I find it pretty shocking your wife cannot reliably send an SMS.
Apple seems to rely on long term user training sometimes. I knew that was a send button because I have used previous versions of iOS, it's never even occurred to me that it doesn't look like a button at all. Similarly, I was late to the iOS party, I joined around iOS4-ish, and there were features I discovered completely on accident. Like double clicking the home button on the lock screen to bring up audio controls. Great feature, but I had no idea it was there. People joining in the next few years probably won't know to swipe from the top or bottom until they do it by accident.
I definitely agree that the UI of buttons in iOS is broken now. The blue is only a useful hint if you are 1) a long-time web user, and 2) not color-blind.
You don't have to be 65 to have that problem, at first I couldn't tell, flat UI isn't accessible in any way, is definitely a step backwards in accessibility.
I agree 100%. But I also remember posts on different forums about how button shapes are a skeuomorphic anachronism that nobody needs anymore. So if Apple didn't address that audience, they'd still be screaming about Apple's backwards design.
I've been looking for a particular quote about this, and just found it. I thought it was a random forum poster, but it was actually John Gruber from Daring Fireball:
"The design of the iPhone software was entirely informed by the fact that this was a new experience, it was nothing like using an existing smartphone, nor anything like using a Mac or Windows PC. It needed training wheels to get people up to speed. Thus, to name one small example, why iOS buttons have tended to look so very button-y. To inform the user, as clearly as possible, that this is a button that can be tapped.
Look around you. Any street corner. Any office. Any shopping mall. Any restaurant. You will see people tapping on touchscreens. We all get it now. iOS-style computing is no longer novel; it is now the standard interaction model for personal computing.
The primary problem Apple faced with the iPhone in 2007 was building familiarity with a new way of using computers. That problem has now been solved. It is time to solve new problems."
I remember a few years ago here there were lots of front page posts criticism skeumorphism, and commentors criticized it as well. Flat design was all the rage.
THIS! "What is clickable" is one of the first things the mind wonders when presented a new page. If the true answers aren't obvious in 0.3 seconds, it's a bad design.
The latest UI trends to remove "button style" and go to "flat" and just text is a prime example of how we've gone the wrong direction.
The backlash against skeuomorphism has really led to some serious regression in design affordances. In a similar vein, there are text inputs in the latest OS X that are indistinguishable from labels. I'm not anti-fad but these are real usability issues.
[0] - http://313e5987718b346aaf83-f5e825270f29a84f7881423410384342...