Only shuffle for you

The shuffle all button in iTunes

A confusing UI element in Apple’s iTunes.

  • Clicking the button will result in an action that works as advertised, it shuffles the playlist. This button includes 2 other actions as well which may lead to confusion for many. It also plays the song list, without any indication that this is an included action. The other action is skip. While the list is playing clicking the button again will result in the player skipping the currently playing song.
  • There is no indication that the shuffle function is active and no way to turn it off (no stop), except by going into a nested action inside the menu bar.

These little details matter when delivering a coherent user experience.


Bluetooth Pairing UI Fail

Logitech K380 bluetooth keyboard pairing UI is likely difficult for but the most advanced users.

I gave this keyboard, a temporary replacement to my Apple wireless, to my wife to use with her iPad – it’s a neat keyboard in theory but I never was comfortable typing on it. She now seems pretty happy with it, at least compared to typing on glass.

But for normal people this keyboard, and Apple’s for that matter, presents a UI problem. My wife wants to connect the keyboard to the iPad – at first glance how does she accomplish this task? There is nothing in this picture that supports that task, that supports her mental model of connecting 2 devices together.

She knows it’s a bluetooth keyboard, so she looks through the settings app. on the iPad and finds Bluetooth settings within which shows which devices are connected. No luck there.

So she asks.

I tell her she has to pair the devices. There is no UI to support that label. Because I have previously learned the UI from numerous other bluetooth devices I come over try long pressing the bluetooth icon and eventually we are in action. Known of that is at all obvious or learnable without outside guidance. Most bluetooth devices, especially the Apple BT keyboard I have, which has you long press the power button (!), allow users to fail in this basic task.

If we look to minimalistic or very simple UI the task should be automatic, like with Apple’s new EarPods. If we are unable to accomplish that then we need more obvious UI that directs the user to task completion – like a button with a label that has connect/pair or an icon with a universally accepted connect/pair meaning. The bluetooth icon is slowly becoming that symbol but I bet most people would not recognize it as such in tests.


Because every person knows what he likes, every person thinks he is an expert on user interfaces.
Paul Heckel


Something found from my notes. How much closer are we today (this was from 2008)?

By 2020 the terms “interface” and “user” will be obsolete as computers merge ever closer with humans. It is one prediction in a Microsoft-backed report drawn from the discussions of 45 academics from the fields of computing, science, sociology and psychology.
It predicts fundamental changes in the field of so-called Human-Computer Interaction (HCI).

By 2020 humans will increasingly interrogate machines, the report said.

In turn computers will be able to anticipate what we want from them, which will require new rules about our relationship with machines.
Computers to merge with humans


Dialog boxes

Perhaps due to the influence of Windows based software poor UI design, I often come across the common mistakes superfluous and poorly thought out dialog boxes. In addition to the maxim below, I believe we should avoid creating error dialogs when an undo will do. Unfortunately the essential undo function is still often forgotten.

Dialog boxes should be action-oriented; they should help guide users towards what their next step is likely to be and it should provide them with the information that they need in order to be able to accomplish that next step.


He doesn’t state it strongly enough. Unfortunately, no matter how strongly or loudly you state this fact, often it goes unheard.

Typically, the burden is on the user to learn how a software application works. The burden should be increasingly on the system designers to analyze and capture the user’s expectations and build that into the system design. Norman, 1988


So often in my new experiences complexity is the selling point, the starting point, and/or the proof of your value. People (customers) don’t share this vision. People are intelligent but must be set free to construct the level of complexity they are comfortable with, or need.

Complexity isn’t designed but rather rises spontaneously through self-organisation. Start with basic or simple interactions and allow more complex behaviours or patterns to emerge.
From an old project proposal, source is likely from theory of emergence.


No medium has managed to reach the status of genuine artistry without offending some of its audience some of the time. Even under the user-friendly dictates of interface design, you can’t make art without a good measure of alienation.
Steven Johnson, Interface Culture (HarperEdge, 1997)


More choice equals more choice

My current projects all involve dealing with issues of featureitis, software with simple uses, but with a monstrous amount of controls and options. It’s well designed software created by brilliant nice people, but many have fallen into the belief that more UI controls, more options, more visible data, somehow makes software more desirable. This is of course a long held problem, routed not just in software (ala. Microsoft Windows) but in Western Society herself. As far as interface design is concerned, I know from experience, more choice as a feature seldom works, as complexity leads to more complexity, more choice leads to dissatisfaction.

Neil Turner summarises some of Barry Schwartz’s thinking in answering the question:

Is lots of choice a good thing?

You see it turns out whilst people will invariably ask for more choice, lots of choice is not really a good thing for the following reasons:

  • More choice means more options for people to consider, and a greater cognitive workload to do so, as all the different options are weighed up and evaluated.
  • With lots of choice the burden of responsibility is placed on the person making the choice, rather than those drawing up the choices. If a bad choice is made it’s because someone chose the wrong option, not because a poor set of options were made available.
  • More choice means greater expectations, and a greater probably of not meeting those expectations. With so many options available, people will expect there to be one that is exactly what is need, and will no doubt be disappointed when they don’t choose it.
  • More choice means less engagement.Sometimes people would rather not take part, than have to go through a million and one different options. For example, an interesting study showed that for every 10 investment funds that an employer offered for their pension scheme (e.g. 10, 20, 30, 40 different funds, and so on), uptake fell by 2%. Employees were put off participating because they didn’t want to have to select from so many different options.
  • The paradox of choice and why I stopped using Spotify

Ideally, we would focus entirely on those features or controls that users need to accomplish their goals while deleting our through perhaps progressive disclosure keep all the complexity hidden from all but the most advanced user. My experience is that that is far more difficult than it ought to be but its a challenge worth engaging in.

Below is a ted talk where Psychologist Barry Schwartz takes aim at a central tenet of western societies: freedom of choice. In Schwartz’s estimation, choice has made us not freer but more paralyzed, not happier but more dissatisfied.


A redesign of the iOS notification center

Some nice ideas but in practice I doubt I would use them. People prefer to do most of their interaction within the app itself. Notifications are simply alerts and as such don’t need to be interacted with. Launching the app that sent the alert is only a swipe away, that should be efficient enough.


OmniGraffle for User Interface Design

A video detailing how Omni designs its own apps, “starting with a quick iPad sketch and ending up at a pixel-perfect, interactive design”. I may revisit using OmniGraffle but for the short term at least I’m invested in using Sketch.

This is not seat-of-your-pants level of exciting but I always enjoy listening to how another teams approach interface design.