My Brain Has No Space For Your User Interface
joshtimonen.com
joshtimonen.com
Does it take some figuring out when I jump from one android to another? Or to an iPhone? Or an old style phone? Sure, but so what? Do I want to do it daily? Not really. However, doing so every now and then is hugely beneficial to seeing how others see the world.
This is no different than why you should try different programming languages. Sometimes, it just helps to see things differently. Even if it is the same old todo list you have seen countless times. Maybe something will "click" this time.
Seriously, consider that when you are using a new UI, you are seeing something as envisioned by someone else. It may be that the someone else is a committee that couldn't agree on anything. In this case the learning experience will be mostly frustration.
However, do you really think that things that you know and use daily are truly easy? Watch a child try to operate a door sometime. Any kind of door. It does not matter.
Maybe I'm just projecting and hoping that by instilling a sense of "keep trying" and "learn what you can do, do not get upset with what you can't" to my kids. Surely I am not alone in this, though.
And this is far from new. The devices we have to learn nowdays are downright easy compared to the stuff from years past. Have you seen some of the heavy machinery that people operate? There are more levers on a standard worksite tractor than I can really make sense of.
Consider the physical world. Cars, toilets, old-school dial telephones, doors, blenders, paper books, hammers, bulldozers, can openers... have vastly disparate user interfaces, but no one complains about having to learn them all, nor do they expect some kind of chimeric "common user interface" and complain when the hammer and the table saw don't work the same way.
Hammers that required you to adjust screen sliders to adjust the force and angle of the blow would suck. Hammers and table saws that were given some sort of half-assed "common user interface" would suck bad.
So, yeah, there are a lot of bad UIs. That doesn't make (current favorite UI) a good one, though. The way that this gets better in the long term is for people to keep making new UIs (most of which will inevitably be bad). Trying to stop that process just means that it'll take that much longer before we get good ones.
This also reminds me of something Neil Gaiman said, which is really a second hand quote from Douglas Adams:
From: http://readingagency.org.uk/news/blog/neil-gaiman-lecture-in...
"I do not believe that all books will or should migrate onto screens: as Douglas Adams once pointed out to me, over twenty years before the kindle turned up, a physical book is like a shark. Sharks are old: there were sharks in the ocean before the dinosaurs. And the reason there are still sharks around is that sharks are better at being sharks than anything else is. Physical books are tough, hard to destroy, bath-resistant, solar operated, feel good in your hand: they are good at being books, and there will always be a place for them. They belong in libraries, just as libraries have already become places you can go to get access to ebooks, and audiobooks and DVDs and web content."
When we arrive at good design, it can be surprisingly resilient.
Some people do complain. Many versions of the common objects you mention have really stupid interfaces, which cause delays, mistakes, and lots of frustration.
I highly recommend Don Norman’s book The Psychology of Everyday Things (later editions retitled The Design of Everyday Things) which analyzes such objects (doors, telephones, cars, refrigerators, faucets, stoves, etc.) and concludes that many of their interfaces are horribly broken.
But even before reading the book I observed people running into trouble with household objects all the time, found great frustration in some of their designs, and was constantly puzzled by the many features on appliances and gizmos that no one ever seemed to use.
The book just gives a nice framework for thinking about such problems, and a great set of concrete examples.
There are plenty of things with great user interfaces, and the way you can recognize them is if the interface "disappears" when you use the thing. If you can use the thing without thinking, without having to read a sign, without having a cheat-sheet, when you're very sleepy, and you still do the right thing, then the interface is good.
My pet peeve is household appliances that are just in the way, that do the wrong thing, that make you do unnecessary steps to get what you want.
I still miss my old microwave with its huge digital knob for setting time and power. I developed muscle-memory for using that thing, because it just felt right, my body knew exactly how much to turn the knob to get a certain time.
My current one is atrocious, it has a number pad. No.
And it doesn't have to be that way, I'm always concious about user interfaces of things I buy, and go out of my way to buy stuff that has good interfaces. I hope that by supporting companies that put in the effort, they'll win in the end. (It's a slim hope, though.)
I'm not sure any of this applies to table saws. You might be setting the standard a little high.
The ideal car user interface (for car-as-a-tool, not car-as-a-toy) is "set destination, go." Or some sort of browse mode where you just tell it when you want to make a turn.
There are similar advances available for some of those other tools: consider the table saw/router/jigsaw/etc. If you just want to get something built, do you want to learn the finer points of using and controlling all those separate tools, or would you rather just be able to say "cut it to look like this" and let a computer do the driving? (I'd love to have a nice CNC setup in a workshop someday...)
I'm sorry, but this has got to be the most ridiculous UI idea I've ever heard suggested as an improvement.
Yes, one day soon we'll have self-driving cars. But until then, the steering wheel is damn good UI. You turn right, you go right. You turn left you go left. You turn more left you go more left. It just works.
I agree the other parts of a car UI suck as of yet. Bit the steering wheel, brake and gas pedal are examples of good design, they do whT they are supposed to and get out of the way.
There are a lot of positive things to be said about the car's UI.
[Edit: grammar]
Windows are again usually pretty good. However, in my mini cooper, the window switches are located in the center console, unbelievably bad design in an otherwise pretty well designed car.
I mentioned steering, gas and brakes primarily because they are the same for almost all cars.
In some ways I think Accessibility is a misnomer -- it's the way Apple says, hey, you 20% of pro users, here's something you can try. Hence Accessibility on Mac getting a security center so apps can use those hooks to resize and read the screen the way "most" shouldn't.
I treat it as a toy box, and frankly still miss most of its settings on Android. Have a look at WWDC videos on accessibility, unlike the counterparts from Google and Microsoft which often require special extensions or apps to enable, Apple delivers the whole thing, and it really works.
Why else would Apple offer 8-9 different font sizes in roughly four font weights (normal, bold, and fonts overwritten normal and bold). Yes, some options could be highlighted during the welcome screen, in the same way VoiceOver is on a Mac at first launch. But like right-click, some things you have to turn on yourself until they sync over the cloud. But that's life. At least their settings apply across the board to all apps.
The default is just tuned for pretty to look at and show to your friends, not accessible. Fitting for a status item.
Makes sense to me.
1. I have a lot of really cool brand new stuff.
2. Instead of having 100 different UI's, let's have less.
Is there any value in either? Half bragging and half common sense does not an engaging essay make. And while I appreciate what I understand to be his second point, I don't think it's necessarily right.
Sometimes it makes sense to have more or less buttons on something, depending on its functions. A TV remote with anything less than 20 buttons is probably useless. A UI which does not take advantage of the wide-ranging input is equally useless. A car's GPS can't really work well with just 2 or 3 physical buttons, and vanilla Android is obviously ill-fitting, even with a touch screen. The author, a serious Mac user, shouldn't even be satisfied with OSX. Has he read the UNIX Hater's Handbook? It criticizes OSX's underlying operating system with the same points he makes against everything but his beloved iToys.
I have a simple solution to the problem of too many UIs. Stop buying so much shit. Keep the growth of technology in check by voting with your wallet, and keep your mind clear by not filling it up with anything other than what is absolutely necessary. Use your phone as your GPS and TV remote. Watch TV on your computer. You can really reduce a lot of your bulk, both in terms of hardware and software, by getting more out of what you already have.
[Obligatory comment against software patents.]
The web isn't a current solution, although it has the potential to be it.
And there aren't any good desktop solutions because there are so many UI patterns and elements that are required.
I wish I could define or remap buttons on physical appliances, granted it's an advanced mode common users would not be burdened with.
There... now you are UI-defragged and you make better coffee.
This sounds as much about having too much stuff in your life than it does about user interfaces.
If you are complaining about your cars touch-screen, you should perhaps be considering if you really need a car with a touch-screen in it at all.
And for the record, my coffee machines user interface is roughly as complicated as the user interface to a light-bulb.
I can sincerely say this about all the graphic designers and ui/ux designers I've ever worked with!
There were differences between applications running within the same ecosystem, of course, and not as wide of a range of device types. But because of the greater consistency, it was generally much easier to learn how to use new applications. A lot of existing knowledge carried over, and was immediately applicable when using a new application.
This uniformity has unfortunately never really developed well on the various mobile platforms used today. It has deteriorated quite badly under Windows, with Microsoft themselves being somewhat responsible for this. The situation is perhaps the best under OS X. But even here we've still seen applications like Chrome and Firefox use non-standard UIs.
Hah. Haha. Hahahahaha.
If you pulled similar samples of UIs from the 90s and from today, I bet you’d find about the same proportion of fucked up designs in both (and that proportion would be shockingly high).
Some applications (especially on Mac OS) tried to stick to Apple’s human interface guidelines, and some of those were successful, but even on the Mac there have always been heaps and heaps of horrible broken user interfaces. And don’t get me started on Windows applications from the 1990s. What a steaming pile...
I think what’s changed today is that (a) with ubiquitous internet and a very easy install process, people end up interacting with more applications (at least to download, play with once or twice, and then forget about),
It would have worked, too, if Apple had ever been willing to follow suit.
I never claimed that all UIs back then were "perfect", or that there weren't "badly designed" UIs. I was very clearly talking about consistency between different applications running on the same system.
When using applications from that period of time, one could expect to find a menu bar and menus that were quite similar to those of other applications. Of course there were application-specific menus and items, but there was at least a common subset shared by nearly all applications, especially any that were designed with even the minimal level of care.
The same held true for toolbars, keyboard shortcuts, common dialogs, and so forth. There were also general conventions for how the application-specific parts of UIs looked and behaved. While not everyone followed these conventions, of course, most of the major and seriously-designed applications did.
We have lost much of that today. It's particularly bad when it comes to mobile apps, and getting worse when it comes to desktop applications. Yes, Microsoft and Apple themselves are to blame, to some extent. But it goes much beyond that these days. Inconsistency is the norm, and that does lead to inefficiency.
There were horrible UIs with giant software manuals, weeks of training, etc.
You likely just grew up on the stuff and didn't realize how bad it was because learning was easy for you then.
And it's not just devices, but the web, too! Every web application must, practically by necessity, reinvent some or most of the basic UI controls.¹ Most professional websites do, and each is different. In many ways, the web makes me think that many of the concepts discussed by HIGs (simple, non-graphical stuff like, "put the 'accept action' button in the same place) have been lost.
Even Chrome disobeys my system preference for UI, and rolls it's own. Then the websites it displays roll their own.
Some of the comments below you remark that many applications disobeyed (or ignored) good UI principals, but nothing's changed with the web; there's still plenty of bad examples². If anything, its worse, as there is no system HIG to fall back on as "correct": it's not what a bunch of people decided was a good idea, it's not what I want, it's just whatever the designer liked.
HTML is still playing catch up in UI: we just got flexbox for layouts (which you need, imho), but some things are still just plain hard. (Pinning a header row and header column on a table, for example.)
This doesn't even begin to scrape the "flat UI" fad, which I think is bad because you can no longer discern actionable areas like buttons. With desktops, I could choose my UI theme. With the web, I'm again stuck with whatever some designer liked, instead of what I want.
¹HTML provides the simple stuff like checkboxes and buttons, but a tree-list-view? Good luck!
²We will have reached UI nirvana when I never see another checkbox used as a radio button.
And for this reason, an app has to be significantly better for it to be worth the bother to switch.