Tools Not Toys

Google's Project Ara has a wonderfully misleading sales pitch. It's a cell phone that lets you choose the parts you want in order to create your perfect phone. Interested in photography? Buy a high quality camera component and snap it on. Sounds simple. However, dig a little deeper and fundamental problems with this idea start to emerge.¹

What if the camera you buy has features that the camera app you use doesn't support? What if the camera's drivers are buggy? What if the camera hardware requires two component spaces on the phone frame instead of one? Is everyone meant to play a game of geometry to figure out what set of components fit together to allow the features they want?

How about a different scenario. What if you buy a higher resolution screen, only to find out that your phone's GPU is too slow to support it? What if you buy a faster GPU and it requires a faster interconnect than your cell phone frame uses? Why should anyone even need to know what a GPU is?

There's also the tradeoff of bulk, where making every component removable means losing significant space to external housing and interconnects. Even worse, each component is forced into a predetermined size whether it needs the space or not.

An optimist would say that Google is working on solutions to these challenges. On the other hand, you could look at 30 years of history with PCs and predict a bleaker outcome for Project Ara. In this more likely scenario, customers will need to figure out what components work together, which have stable drivers, where to buy them, and how to configure them.

It should be clear by now that the reality of piecing together technology components never lives up to the promise of a seamless product. Every time the tech industry tries to do this, it finds out that the whole is less than the sum of its parts. Exhibit A: Compare any laptop to Apple's MacBooks. Exhibit B: Compare any Android phone to Google's Nexus line. Exhibit C: Compare any Windows tablet to Microsoft's Surface.

Fraser Speirs once wrote:

The Real Work is not formatting the margins, installing the printer driver, uploading the document, finishing the PowerPoint slides, running the software update or reinstalling the OS.

The Real Work is teaching the child, healing the patient, selling the house, logging the road defects, fixing the car at the roadside, capturing the table's order, designing the house and organising the party.

As someone who works to make technology transparent, Project Ara strikes me as a wrongheaded concept. It rewinds the clock on the last decade of progress in the tech industry. Instead of ushering in a new generation of personalized, upgradeable phones, it will be a mess of barely functioning components from different companies, with limited compatibility across generations of hardware.

We should be making tools that people can use effortlessly instead of pretending tech products are toys that people want to spend time tinkering with.

1. Let's temporarily ignore the fact that Project Ara is vaporware, despite a puff piece in Time and the usual breathless coverage in the tech press. ↩

Reassurance

Thanks for the weirdly unsettling notification, Facebook:

Should I be feeling insecure about my social life? Does your data mining know something I don’t? Are my real life friends about to leave me? Is it me? Am I too anxious?? Oh God! It’s me isn’t it?!!

Number of Clicks Doesn't Equal Ease of Use

Jon Bell recently stated that counting the number of clicks needed to complete a task can be a misleading objective when designing software. I recently illustrated this exact point in the UX class I teach by way of an example from a real world usability study.

In the study, we tested two designs for a mobile checkout flow. One had most of the flow condensed into one page. The other split the flow into 8 or so separate pages, with each step asking a simple question such as “Are you a new customer? [Yes] [No]”.

Conventional wisdom would have predicted the first design to be more successful. Not only did it require fewer taps to complete the checkout task, but it was faster since there were fewer pageloads involved. However, users overwhelmingly preferred the second multi-step design because they had an easier time completing the task.

Rather than aiming to reduce the clicks needed to complete a task, aim to reduce the user's cognitive load. There is no one-size-fits-all solution for this goal. The only thing you can count on is that number of clicks isn't a valid metric to measure it by.

Neither First Nor Right

Apple usually takes the time they need to release products that work well. They are seldom first, but as John Gruber succinctly put it years ago, they are usually "first to do it right". With their move to minimalist design, they were neither first, nor, unfortunately, did they do it right. So while I'm griping about fundamental usability issues in iOS 7, I may as well let loose on some other problems:

Weather

• Have you ever tried reading the forecast when conditions are hazy? White text against a gray background sure looks pretty. Hope your eyes are as sharp as a 6-year-old's, instead of half-blurry as you wake from sleep and try to figure out how to dress for the day.

• The weather forecast in Notification Center is now written in conversational language rather than being represented with numbers and iconography. Good thing you weren't trying be quick by using the Notification Center shortcut to check the weather, right? Right?!

Music

• Do you play music wirelessly using AirPlay or Bluetooth? Great! Switching to wireless speakers now requires navigating a couple levels deep inside control center, rather than having a toggle directly in the music app. It's far less discoverable and far more cumbersome to use.

• Do you rotate your phone sideways to browse your music by album cover? No? Is that because you always see the first alphabetical albums no matter where you are in your music collection? Well, iOS 7 has done nothing to fix this, but the design is now flat! Oh, and Music.app likes to switch into this view much more easily than in previous versions of iOS. Lay your phone on a slightly uneven surface, and chances are Album Wall will take over.

Safari

• Did you forget the Reader feature exists? It probably doesn't help that the redesign gave the button the same look and location as a website's favicon. Oh, and it doesn't look like a button.

Reminders

• Switching between lists in Reminders requires a tap or swipe up from a narrow area at the bottom of the screen. Careful you don't invoke Control Center instead! You only have a few millimeters of space to get this action right.

Maps

• I almost feel bad for kicking Maps while it's still down, but not quite. If I look up a street address without specifying my city, don't show me a location in another state. My phone knows exactly where I'm standing. Use this basic piece of contextual info.

Control Center

• Do you use your phone as an alarm clock every day? That shortcut to the Clock app in Control Center sure is handy huh? Wouldn't it be nice if it didn't switch you into the countdown timer tab every time? Apparently you should be timing things more often. Maybe this was an internal shortcut the design team at Apple used to stick to the deadline that made them ship a major interface overhaul in too little time.
• Isn't it great how Apple overloaded the swipe-from-bottom-of-screen gesture on the lock screen? If you try to bring up Control Center you sometimes launch the camera instead. And vice versa.

• Speaking of swiping up from the bottom of the screen, turns out Control Center doesn't always like to pop up. Usually when there's a keyboard on screen. Ever tried invoking it from the Messages app and ended up with a string of gibberish letters in the message field instead?

App Switcher

• The redesign of the app switcher now lets you see just 3 apps at once on an iPhone, instead of 4. And best not to think about how little you can see on the iPad compared to the old design. The bigger problem is that navigating the app switcher requires far too delicate a touch. It used to be possible to fling that set of apps to the side to see the next set. No matter how much momentum you swiped with, the app switcher would always scroll and stop at the next discrete set of apps. Now there is no stop boundary, so a quick swipe will send 20 icons flying by before the scrolling slows to a stop. So you must slowly, delicately swipe, then tap the screen to stop too many apps from sliding by, then do the same. Tedious. Time consuming. Concentration-heavy.

Global Problems

• Animations. Slow, interaction-blocking animations. Turning them off is not a good solution. They provide important context. There is no good reason for them to take so long, or to stop you from interacting with your device until they've finished.

---

And if you're hungry for more, plenty of others have weighed in as well.