As an iOS developer I’m excited about the arrival of the new watchOS since it will bring new capabilities and features to build upon. Considering this, I will bring you a wide panorama of what you can develop with watchOS2 and understand opportunities for your next killer (watch) App.
First of all, the main changes that watchOS 2 introduces are:
- Architecture: watchkit extension now runs in the watch, when in watchOS 1 it ran on the Phone. This brings two main considerations:
- WatchKit extension to be implemented using the frameworks of the watch SDK instead of iOS SDK.
- Files and data from the extension are stored in the Apple Watch; if different data is needed it must be taken from the companion iOS app.
- Clock kit: this new feature permits to use Complications within third party apps. As developers now we are able to present up-to date information of our app, updated throughout the days, in the watch’s faces. This features works entirely with the Digital Crown which gives the users the chance to go backwards and forward in time, to display the required information.
Added to this interesting changes, we will be analyzing two features which give sense to the Apple Watch - apart from time telling - and how to tackle them technically: Audio Recording & Location.
Definitely voice commands are one of the most important features within the Apple watch. Several apps are nowadays using audio commands to speed up users’ actions. Here are some examples:
- Amazon: search products from speech indications.
- Evernote: dictate notes/reminders.
- Shazam: identify music directly from your apple watch.
Within the new WatchKit, audio will be recorded through native apps which means less delay and no need to have your connected iPhone next to you when recording.
Added to this, the watch standalone may be able to store audio and later on send it to the iPhone app. Let’s say your next killing idea comes to you while you are jogging and you don’t have your iPhone around, is that a problem? Not anymore! With watchOS 2 you will be able to record this idea in an audio and later, when arriving home, synchronize it with -let’s say - Evernote.
Technically, once you get the audio on your watch you can do many things: attach to other information (mails, messages, etc), play it, share it on Twitter or send it to your phone to be processed. Also, you can perform speech recognition on it to fire commands. A compelling possibility is to send the audio to TLSphinx to detect special words and control your app from your wrist using your voice.
WatchOS 2 can also play audio and video. This means that a user can send audio or video directly from the phone to his watch. For example, an audio/video file sent via Whatsapp/Messenger can be reproduced directly from the watch (considering these technical specs) .
Another important feature when developing for the Apple Watch is the possibility to use location information provided from the user’s companion iPhone App.
Here we mention some Apps nowadays taking advantage of this feature:
- Uber: ask for cars directly from your watch and know when they will arrive.
- CityMapper: indicates the best public transport alternatives depending on your location.
- Babbel: learn foreign language words on the go depending where you are.
An opportunity arises: with the new ability to run apps in the watch you can now follow a predefined route getting directions on your wrist, without even looking at it. Let’s imagine an exercise app that permits you choose among many routes in Central Park NY - each one with different difficulty levels - you pick one, leave the phone in the car, and go for it. Because WatchOS 2 can run apps on it’s own, you can open the exercise app and check where you are on the route, the app also can tell you when to turn right or left using the Taptic engine (two taps: turn left, three taps: turn right) also available since watchOS 2. Translated, this means you can focus on the road and rest your busy mind. Of course all your progress can now be tracked using the HealthKit API added in the new version.
Myths & Truths
- We still need the companion iPhone App for speech recognition to be efficiently executed.
- Watch & iPhone linking can be done with wifi, accelerating information transition. We cannot connect the Watch to a wifi network and use it independently.
- Whatsapp nor Messenger have an Apple Watch app… opportunity?.
- Evernote does not have the mentioned feature about attaching audio from the not paired watch.
- No standalone location in the watch. Companion App needed for first location iteration.
Summarizing, there is a lot going on with the Apple Watch and watchOS2 and off course, there are other important features to focus in and tons of opportunities arise. It is relevant to understand that we are talking about a developing market (the wearables) which means that it is still a niche one. However, new technology implies new interfaces to build upon.. maybe you got into a great idea? Time to go for it and create the next killer Apple Watch App.
Any question? Let us know in the comments!
Like what you read?
Subscribe to our newsletter and get updates on Deep Learning, NLP, Computer Vision & Python.