Nielsen Norman group slams gestural interface usability, ironically points finger at iPad and iPhone
Nielsen Norman group has slammed gestural interfaces, in an article entitled A Step Backwards In Usability:
The usability crisis is upon us, once again. We suspect most of you thought it was over.
Given that two-year-olds and centenarians are using iPads, I did, yes.
Well you are wrong.
Oh.
In a recent column for Interactions (reference 2) Norman pointed out that the rush to develop gestural interfaces – “natural” they are sometimes called – well-tested and understood standards of interaction design were being overthrown, ignored, and violated.
Violated? Sounds serious. SOMEONE CALL THE USER INTERACTION POLICE.
=====
INT: Nielsen Norman group. Donald A. Norman and Jakob Nielsen get into their superhero outfits and zoom towards the scene.
Super Norman: OH MY GOD, it’s worse than we thought, Jackob. It’s horrific.
Super Nielsen: Yes, new technologies require new methods, but the refusal to follow well-tested, well-established principles leads to usability disaster. I will KILL THE VIOLATORS WITH MY LASER VISION.
Super Normal: You don’t have laser vision, Jakob.
Super Nielsen: Bugger. How about moaning about the iPad in my bi-monthly column for ACM CHI magazine, then?
Super Normal: Sounds great!
END CREDITS
=====
OK, *serious face*, these guys do have some good points regarding visbility, consistency, scalability and reliability—all standard tenets of strong usability. Gestures aren’t necessarily easily discoverable in iOS and other touch-based systems, but that’s also largely because many of them are new. Guidelines are, through popularity, slowly being formed. Nielsen Norman group also don’t seem to note that the intuitive nature of gestural interfaces (rather than the abstraction seen in other forms of computing) means that things are more easily learned and less likely forgotten. My dad can happily do stuff on my iPhone, despite not owning any iOS device, yet his Mac still regularly flummoxes him.
Anyway, back to the article:
The first crop of iPad apps revived memories of Web designs from 1993, when Mosaic first introduced the image map that made it possible for any part of any picture to become a UI element. As a result, graphic designers went wild: anything they could draw could be a UI, whether it made sense or not. It’s the same with iPad apps: anything you can show and touch can be a UI on this device. There are no standards and no expectations.
No standards? Really? I’m pretty sure Apple has extensive guidelines on user interaction. But there are apparently other reasons people are having trouble.
The misguided insistence by companies (e.g., Apple and Google) to ignore established conventions and establish ill-conceived new ones.
Yes. Let’s stop innovating.
The developer community’s apparent ignorance of the long history and many findings of HCI research which results in their feeling of empowerment to unleash untested and unproven creative efforts upon the unwitting public.
JUST STOP TRYING NEW THINGS, IGNORANT DEVELOPERS!
In comments to Nielsen’s article about our iPad usability studies, some critics claimed that it is reasonable to experiment with radically new interaction techniques when given a new platform. We agree. But the place for such experimentation is in the lab.
ALTHOUGH IF YOU’RE RICH DEVELOPERS, WE PERMIT YOU TO EXPERIMENT IN YOUR ‘LAB’!
Most progress is made through sustained, small incremental steps. Bold explorations should remain inside the company and university research laboratories and not be inflicted on any customers until those recruited to participate in user research have validated the approach.
Bold explorations like the top-selling iPad and iPhone, you mean, rather than the sustained, small incremental steps we’d previously seen in smartphones and tablets? OK, sounds great. I’ll see you back before the turn of the century and we can party like it’s 1999 until we die of RSI through using our mice until our arms explode. I look forward to it.
Hat tip: Chris Brennan.
Looks like these guys never used a Xoom. Talk about old conventions in new hardware – that’s a mouse and keyboard away from being Windows.
Perhaps the problem is that they *have* used a Xoom….
Seriously, yes, we’re in a transitional period (as opposed to what, I don’t know), and yes, some conventions are going to be formed, and we’re all figuring out what’s working and what doesn’t together (and, uniquely, sharing feedback in a way we never could in 1994 when only some people like NN could create and manage a website to tell us that links must be underlined and blue because that’s what TBL decided and later admitted wasn’t a good decision [and which a decent designer could have told him up front]).
Here’s the bigger question, though: has anyone been *frustrated* by a touch interface that didn’t do what they expected, or *delighted* when it did something they didn’t anticipate? Gauged by my 3YO & 5YO’s experience, plus my own I’ll-pad-it-out-to 20+ years of design, it’s the latter. It’s the age of discovery, foax: go forth and conquer!
My mother-in-law (83, can’t/won’t operate a DVD player or a cash machine, fact) was looking at photos on my wife’s iPhone. Wife told her to swipe across to see the next one. She paged through a few dozen snaps like this but, HORROR, she was dragging instead of swiping! (That is, she didn’t let go.) And when you try to drag a whole screen, it stalls, because your finger can’t travel a full screen’s width within the screen. She was managing fine, but IT WAS INELEGANT!
Then I looked across and said “You don’t have to pull it right across, just flick it.” Immediately she KNEW WHAT I MEANT JUST BECAUSE I SAID IT IN PLAIN ENGLISH. Now try that with anything involving a mouse.
What Nielsen doesn’t get is that the world is the lab now. That’s the upside of rapid obsolescence. You get it wrong, everyone tells you you’re an idiot, then you get it right in the next iteration. As with UI, so with bigger issues. Google didn’t worry about copyright. Facebook didn’t worry about privacy. (Both launched with shit UI too, as it happens.) Everybody shouted about how this was the end of the world and what the hell did they think they were doing. Then the world adapted to them.
You can’t test that in a lab. Just do it, Jakob.
Oh wait, you don’t actually make anything, do you?
People can use an iPhone more easily than a Mac because it’s a lot simpler, not because its gestural user interface is anywhere close to being perfect. In fact, there’s a lot of stuff that’s just plain wrong with it (name the top three usability issues with shake to undo!).
So saying that people can use iPhones doesn’t really contradict what Nielsen and Norman write. It’s like saying “people can drive stick shift, so clearly, there’s nothing anyone could find wrong with it, and automatic transmission isn’t needed.”
Also, the bottom part of your article reads like one huge straw man.
For example: “JUST STOP TRYING NEW THINGS, IGNORANT DEVELOPERS!”
Not what they said. They said “try new things, but test them with users *internally* before shipping.”
Another example: “ALTHOUGH IF YOU’RE RICH DEVELOPERS, WE PERMIT YOU TO EXPERIMENT IN YOUR ‘LAB’!”
Again, clearly not what they meant. In fact, Nielsen is famous for advocating that designers and developers do their own cheap usability testing.
@LKM: Despite appearances, I have some time for NNG, and have interviewed its people a number of times. (They’re very nice, and they had great thoughts relating to mobile browsing when iOS first arrived.) However, the group increasingly pushes “don’t do new stuff, because older stuff is better and more well-known”.
I disagree that usage of devices isn’t directly related to gestural interfaces instead of ‘simplicity’—they’re largely one and the same. As Adam says, when interacting with a touchscreen, you’re using intuitive actions that people understand, free from the abstraction associated with pointing devices, menus and so on. That is a big part of the ease of use. I don’t disagree that there are lots of problems in these systems, but I’d rather have extremely rapid development and iteration with a few bumps than a sluggish rollout that sees us end up with something usable in 2020. (And, as I noted, my dad still has major issues even with relatively simple tasks on his Mac, but he’s taken to Kindle like a duck to water, and is fine with iOS, despite not even owning a device.)
As for cheap usability testing, I know a whole bunch of devs who do this, but they all tell me the same thing: while you can sometimes find real bloopers, it’s usually not until something’s in the wild and, as Adam said, using the world as your lab, that you can really hone your interfaces. And NNG comparing iOS to a web-page image-map? Really?
EDIT: It was also pretty interesting interviewing Fraser Speirs about his ongoing iPad-in-school experiment. My take-home from that was that iOS success in the classroom—from toddlers to teens—was extremely tightly reliant on intuitive gestural interfaces, but also many that didn’t necessarily conform to ‘norms’. People rapidly learn how each app behaves, with the iOS device becoming a new piece of hardware each time, rather than the Mac, which is a piece of hardware that runs software, hopefully in a consistent manner. With iPad apps, you use them individually, with fewer (but still some) global expectations, and a rapid learning curve.
>I disagree that usage of devices isn’t directly related
>to gestural interfaces instead of ‘simplicity’—they’re
>largely one and the same.
Compare Mail on your iPhone to Mail.app on the Mac. Is the one on the iPhone easier to use because you use your fingers, rather than a mouse, or because it offers less than a tenth of its big brother’s features? Or, to put it another way, if you could attach a mouse to your iPhone and control its Mail app that way, would it really be that much harder to use?
> As Adam says, when interacting with a touchscreen,
> you’re using intuitive actions that people understand
Unless we’re talking about something like a drawing application, where you’re replicating something that children do in the real world (painting with their fingers), you’re not. You still have buttons and list views and text fields. There’s nothing intuitive about this; it merely feels intuitive *to you* because you already know about these things.
In fact, a lot of the gestural stuff is the opposite of intuitive. Slide your finger over a mail message to delete it? Scroll all the way up to reveal a hidden search field? That doesn’t make any sense, and intuition won’t help you discover these things.
> As for cheap usability testing (…)
What you wrote doesn’t seem to have anything to do with what I said? My point was merely that you mischaracterized what Nielsen and Norman wrote.
Also, without knowing any of the details, it’s hard to guess what exactly you’re referring to, but it seems to me that the developers you’ve talked to maybe didn’t do the best possible job running their usability tests if all they found was some “bloopers.”
Now, of course there are things you’ll only find out once the application is released. I’m not denying that. But if you’re using this fact to argue that therefore, the best you can do is just release things and hope that they work, that’s a false dichotomy.
I’m not actually sure if you’re doing that, since I don’t quite understand specifically what you’re arguing for. Nielsen and Norman merely point out that you should test user interfaces internally before releasing them. You respond with “JUST STOP TRYING NEW THINGS, IGNORANT DEVELOPERS!”
I’m not sure if you’re saying that devs should *not* test this stuff before releasing it?
“Compare Mail on your iPhone to Mail.app on the Mac. Is the one on the iPhone easier to use because you use your fingers, rather than a mouse, or because it offers less than a tenth of its big brother’s features?”
Both.
“Or, to put it another way, if you could attach a mouse to your iPhone and control its Mail app that way, would it really be that much harder to use?”
Yes. Because abstraction through pointing devices is _still_ something that flummoxes a whole bunch of people I know. Those same people have zero problem with gestural interfaces. Anecdotal, perhaps, but that really is my experience. Same goes for people with young kids—they’re all happily using apps and games on iOS devices that they cannot control using pointers. A friend of mine has a young son who was mortified at how tricky World of Goo is for him on Mac and Wii, but the same kid completed a huge chunk of the iPad version.
“You still have buttons and list views and text fields. There’s nothing intuitive about this; it merely feels intuitive *to you* because you already know about these things.”
Not really, and especially if you approach everything from year zero. Tap ‘x’ is better than: Hold the mouse… Yes, that puck-shaped thing. That moves the cursor—the arrow on the screen. OK, now click ‘x’. Oh, you click by pressing on the mouse. Why has a menu come up? Oh, you pressed the wrong bit of the mouse.
“In fact, a lot of the gestural stuff is the opposite of intuitive. Slide your finger over a mail message to delete it?”
That makes sense to me, in the sense of scrubbing something out.
“Scroll all the way up to reveal a hidden search field? That doesn’t make any sense, and intuition won’t help you discover these things.”
And, as I’ve said, I agree with you and even NNG that there are problems with gestural interfaces and discoverability, but in a global sense I still believe the touch systems are simpler and more intuitive.
“Now, of course there are things you’ll only find out once the application is released. I’m not denying that. But if you’re using this fact to argue that therefore, the best you can do is just release things and hope that they work, that’s a false dichotomy.”
That’s not what I’m suggesting, and I know full well my article was being in part facetious. But NNG has long held the belief that you should use tried-and-tested interface conventions forever, and only very slowly move to new ones. Hell, they banged on about blue links on web pages for a decade, when even the best and brightest in the industry were trying their hardest to yell no. My argument is that you should make things usable and also innovate through experimentation, but you shouldn’t hold yourself back just because you already know something works, and you shouldn’t be afraid of taking big leaps.
>Oh, you pressed the wrong bit of the mouse.
Mail.app only runs on Macs, which default to single-button mice. Just saying 🙂
> That makes sense to me
Of course it does. But that’s not really an argument; you’re not the kind of person who would have trouble learning about something like this. And yet, I’ve shown the “slide to delete” trick to people who have used iPhones for years, and most (if not all of them; not entirely sure) had never heard of it, and hadn’t found it on their own. Because they don’t read Daring FIreball, and don’t watch Apple’s video tutorials.
There are only two ways most people will discover that feature: either somebody tells them about it, or they happen upon it by accident. The second is very unlikely. The first is only likely for “sexy” features, like pinch-to-zoom. Everybody likes to show off pinch-to-zoom, so it doesn’t matter that it isn’t particularly discoverable; somebody will introduce it to you sooner or later.
> in a global sense I still believe the touch
> systems are simpler and more intuitive
I don’t necessarily disagree. But this likely contributes very little to the difference in usability between a Mac and an iPhone. Simplicity is a much larger contributor. A Chrome netbook is a lot easier to use than a Mac despite of the fact that both use a mouse.
And even if we assume that touch UIs are always way easier to use than mouse-driven UIs: it just doesn’t follow that there aren’t problems with touch UIs.
The stuff Nielsen & Norman point out seem mostly reasonable, and the solutions they offer seem appropriate.
“Mail.app only runs on Macs, which default to single-button mice. Just saying ”
Oooh. OK, fair enough. But I’m always like this when dealing with my parents, regardless of the simplicity of the application at hand. Far fewer problems on iOS.
“But that’s not really an argument; you’re not the kind of person who would have trouble learning about something like this.”
Fair enough (also re: discoverability); however, I meant it made logical sense to me, on the basis of rubbing things out. I wonder if one problem is that people are also afraid to try and explore. Lots of people I know have this fear of doing anything much on the PCs or Macs, in case they break them. And yet as James Baker says above, iOS can (not is, but can) be a system where experimentation reaps rewards. (Again, this happens a lot with my dad, who tries stuff, ‘discovers’ something, says that was obvious, and then goes back and swears some more at his MacBook.)
“And even if we assume that touch UIs are always way easier to use than mouse-driven UIs: it just doesn’t follow that there aren’t problems with touch UIs.”
Well, I’m agreed there. But I still believe, even if I said it in a facetious manner (and, well, this is Revert to Saved, after all) that NNG has for years now pushed the same old arguments about an extremely conservative approach that I don’t think is necessarily the way everyone should be thinking.
I’ll forgive them forever if they get superhero outfits though.
> I wonder if one problem is that people are also
> afraid to try and explore.
Yeah, definitely. This seems to be especially (but not exclusively) true for people who are used to Windows computers. Years of using Windows have trained them to avoid doing unneccessary or untested things, because it’s really easy to accidentally mess things up irreversibly.