WWDC 2015 Wish List & Predictions, the Developer's Perspective
Mutual Mobile has been very lucky this year to have a number of our devs awarded tickets to attend WWDC. As the time grows close, speculation about what we'll be seeing this year is reaching a fever pitch. As developers, we have a lot of stake in what Apple reveals in a couple of short weeks, and we've had some great discussions about what we expect to see and what we are hoping for. Here are a number of the things we're wishing for and estimates of how likely we think it is that our dreams will come true.
New Apple TV
The last time Apple mentioned the Apple TV, they lowered the price: "starting at $69". The "starting at" makes it a near certainty that there's an updated Apple TV in the works, and rumors seem to indicate that an SDK for doing development for the updated box is a very likely as well. An SDK would allow video providers to create new apps to share video content from a variety of new sources, and to enrich that video content with stats, multiples points-of-view, social viewing, etc. Likelihood: 95%.
Even more interestingly, however, Apple is well-positioned to become a dominant force in console gaming here. Assuming that the platform will build on Apple's established technologies like Metal, SpriteKit, SceneKit, and the Game Controller framework, it would be very easy for developers to port their existing iOS games to the new box. This would give Apple a huge software library very quickly, easily eclipsing the number of titles available for Playstation, Xbox, or Wii. Likelihood: 80%.
In addition, the existing Apple TV can already invisibly act as a hub for HomeKit, allowing one to control one's HomeKit devices even when away from home. It's possible that Apple may expand the device's role, adding a microphone for Siri support, a UI of some kind for HomeKit functionality, and additional support for automating entertainment hardware. (Putting an infrared emitter in the Apple TV box would allow it to control cable tuners, televisions, and satellite boxes by acting as a remote control for them; alternately, defining common Bluetooth LE profiles for those devices would allow them to be controlled by any device with BTLE support.) Likelihood: 50%.
Finally, Apple could add much of the functionality that iOS devices already enjoy; with the iMessage infrastructure they've already built up and use of on-screen overlays, it would be easy to receive and send texts, place and receive phone calls, provide on-screen notifications, and with the addition of a USB port to support webcams, do FaceTime video conferencing. Likelihood: 25%.
When the subject of what we most want from iOS 9 came up, there was an initial rousing chorus of "Bug fixes! Nothing else!" from our devs. While Apple has gotten most of the obvious, visible bugs sorted out in iOS 8 at this point, there are still quirks, inconsistencies, and things that are just plain broken in the SDK that we still have to work around on a regular basis, and which cause some developers to look back on Snow Leopard, with its promise of bug fixes and performance improvements, with fondness. Unfortunately, bug fixes don't make good advertising copy. Likelihood: 10%.
Other improvements we'd love to see to the iOS SDK: improvements to indoor location tracking, opening up CarPlay so that any app developer can support it, and an API for Siri. As users, we would love to be able to remove Apple-supplied apps altogether or replace them with ones we select, get directions to our friends' current locations in Find My Friends, be able to name locations without having to create contact book entries, improved assistant functionality (a la Google's just-announced and excellent looking Now on Tap or Microsoft's Cortana), and the ability to split the screen between two apps. Likelihood of these is all over the map, but I would guess we'll see at least a few of these appear this year.
OS X Whatever
Our prediction: we'll hear about a new version of OS X. Other than a font change for the new OS, we've got nuthin'. Likelihood: 95%.
As developers, we also have strong feelings about Xcode. While its UI is a little long in the tooth and could benefit from a refresh, stability improvements top out our wish list here too. For most developers, Xcode still crashes many times a day, and we'd love to limit our coding interruptions to bathroom breaks and caffeine runs. Likelihood: 50%.
Given its enthusiastic adoption by the community, we also expect to see support for Swift continue to deepen and mature, bringing refactoring tools, better performance, and further refinements to the language. Likelihood: 80%.
We've done a lot of work for our clients on Apple Watch, so this is a big one for us. Our own Conrad Stoll did a presentation at this past weekend's CocoaConf titled "Wish List for WatchKit"; per today's rumors, it sounds like he'll be very happy in a couple of weeks.
Jeff Williams mentioned plans for the native WatchKit SDK to be revealed at WWDC, which will finally give developers access to health sensors, the digital crown, and tools for making games. In addition, we'd love to see better support for gestures, access to the taptic engine, the ability to take advantage of the device's audio functionality, and better performance. Likelihood: for various parts: 50%-99%.
HomeKit is another area we've been particularly interested in. With the first version of HomeKit, Apple provided a good foundation for a connected home infrastructure. Unfortunately, rollout has been really slow, and there are still currently no HomeKit devices available for purchase. (This should change within a few weeks.) But while the foundation is solid, there are still a number of pieces missing that are needed to help fulfill HomeKit's promise.
The first of these is client software. HomeKit provides all of the necessary plumbing to connect up various devices, but doesn't provide any User Interface at all. Many of the hardware vendors creating HomeKit devices are also working on software to support their gadgets. Unfortunately, being an expert in hardware design doesn't make one an expert in software design. In addition, hardware vendors have little incentive to create a client that will work with any HomeKit device -- they're generally only interested in making sure that their own devices are well-supported.
There's nothing that would prevent an enterprising third party from making a universal HomeKit client that is both beautiful and usable. With the dearth of available devices, however, there has been little incentive to do so. Rumors have begun to swirl that Apple will take this task on themselves and release a HomeKit client soon. This has pros and cons; on the upside, everyone would have access to a solid control panel for their connected home. On the downside, it would reduce the incentive for third parties to develop something of the sort themselves in the same way that having a built-in Calendar app does; third party clients would still exist, but they'd be a niche. Still, I suspect Apple may view this as a worthwhile way to kick-start things in the HomeKit space. Likelihood: 60%.
HomeKit has also established a good foundation for connecting various devices, but has done very little to allow automation of those devices once they're connected. There's a clue in the HomeKit API, however, that Apple has plans to fix this. Devices can respond to various triggers, but currently the only trigger available in HomeKit is HMTimerTrigger -- the single subclass of HMTrigger. In order to move from a connected home platform to a home automation platform, Apple needs only to flesh out this class cluster to include more triggers for occupancy, sunrise/sunset, weather, other sensors, etc. Likelihood: 50%. (Though I'm totally convinced it will happen eventually.)
Other Stuff and Surprises
There will doubtless be other things announced that are less interesting to developers: a new streaming music service, a new television service, etc. And of course, Apple is notorious for their "one more thing" presentations. Items that keep coming up in the rumor mill include an iPad Pro, an Apple Car, and more. Last year, Apple sprung Swift on us with no warning. Likelihood we'll see something unexpected: 90%.
We're expecting a lot from WWDC this year, and are as excited as can be to have many folks from our team attending. You can follow along with our WWDC adventures on our Twitter account, where we'll be sharing what we learn. Come and join the fun!