Why watchOS 2 will make your Apple Watch insanely great

When Wellbeyond released a meditation timer, it was designed from the ground up for the Apple Watch — with a “Watch First” mindset. The user interactions on the Watch were designed first, and the iOS app was almost an afterthought.

While this fostered a very simple and elegant user interface, most development time was consumed by working around the limitations of the first version of the Apple Watch SDK. For example, the opening screen lets you simply select how long you want to meditate:

The app makes the user feel comfortable by presenting a UI that is very similar to Apple’s built-in Workout app. Unfortunately there were two major — and unavoidable — UI problems with the implementation:

Adjusting the time is very slow. The response time of the buttons is terrible. When the user clicks the plus or minus buttons, what happens under the hood takes a ton of overhead:

The watchOS sends a Bluetooth notification to the iPhone, iOS calls a callback to let the app know that the button was clicked, the code then increments the number of minutes to meditate and changes the value of the label. In order to do that, iOS has to notify the Watch via Bluetooth to update the label.

The whole round-trip takes about a quarter of a second, which is extremely slow if the user wants to press this button multiple times to change the minute counter. As a workaround, the number of minutes changes in increments of five for any number above five, and in steps of one under five — 1, 2, 3, 4, 5, 10, 15, 20, etc.

The Digital Crown doesn’t work. Users expect to be able to turn the crown to change the number of minutes. This is how Apple’s Workout app works, and it’s what users rightfully expect. In the original version of WatchKit there’s absolutely no way to get access to the Digital Crown.

And when a user went on to meditate, there are more challenges:

The app can vanish in the middle of a meditation. Almost by definition, people don’t use their phone while meditating. This means that iOS can yank the app out of memory because it’s apparently “not being used.” This is preventable if your app plays sound (such as a background sound or a narration), but the App Store guidelines are very strict in not allowing any of this trickery for apps that stay silent. We worked around this problem by scheduling a local push notification at the end of the meditation time: if all else fails, the user will still be notified when the meditation is over. There would be nothing worse than a meditation app that makes you wonder whether you’re done.

Looking at the Watch during meditation only shows the Watch face. While it’s possible to adjust the settings to show the current app, most users will only see the standard Watch face when they look at it while meditating. They’re not interested in the time, they want to know how far along they are.

You can’t play a sound on the Watch. While it’s possible to play sound on the iPhone from the Watch, WatchKit 1 did not let you play a specific sound on the watch itself. There is no workaround.

The taptic engine is not accessible. It would be great to notify people they’re halfway through their meditation using a small vibration on their wrist. While Apple’s built-in apps fully take advantage of this, third-party apps don’t have access to the taptic engine. The push-notification described above notifies people of the end of their meditation, which is better than nothing.

The trouble with glances. The “Glance” screen isn’t always up to date either:

To update the glance screen, WatchKit 1 has you start your app extension on the iPhone and call a function that would generate the graph and send it back to the Watch over Bluetooth. The Watch then caches the graph from the last time it was shown, but takes about 8 seconds to fully update. That’s way too long for a “glance.”

Ready set… … … … … … … … … … … … go!

Finally, like other Watch apps, the meditation app often takes several seconds to start. While meditation is supposed to teach you patience, this is not exactly what we had in mind.