After a pretty crazy 2013, I was expecting something a bit tamer from iOS 8. 7 was huge: it was the most radical design change that iOS had ever seen. 8 should’ve been like 6 — smaller features like Passbook and OS-wide sharing that are handy, but not game-changing.
Instead, Apple came to WWDC 2014 with new software that might be the most exciting thing we’ve seen on both the mobile and hardware front in years.
iOS 8
There’s a couple things up front that seemed obvious. Photo Stream was version 1.0 to something bigger: Dropbox has been doing pretty killer photo storing up until now, but Apple has the server capabilities now that they’re willing to handle all the photos. The new changes to the Messages app are great, and are worthy additions to a massively competitive space. Snapchat’s recent shift to a “chat” app is pretty similar, and is a pretty good sense of what the bare minimum is to get land share in the chat space right now.
Things like the QuickType keyboard and the Spotlight changes are great too, but they’re really the icing on the cake.
Where iOS 8 is really shining is the SDK changes. (Warning: developer mode is being activated — though you should expect that considering this is WWDC, after all.)
iOS has always had a pretty killer SDK, and it’s tough to compete with the tooling around Xcode, Interface Builder, and the iOS Simulator for rapidly building apps. The release of Android Studio last year makes it pretty apparent that the IDE environment, while not everyone’s favorite solution, is the course app development is going to take for at least the next few years.
My complaint with iOS for a long time has been Apple’s reluctance to open things up. On the Android side, things like changing keyboards and adding actions (Intents in Android nomenclature) have been going on for years; Apple has quite simply just been behind. Places like MacStories.net have been great resources for pushing the boundaries of inter-app connectivity on iOS, but the fact is that if the system isn’t designed to support those connections, it’s always going to be a hastle.
Extensions
iOS 8 lets you extend select areas of the system by supplying an app extension — code that enables custom functionality within the context of a user task.
Hey, that sure sounds like Intents, doesn’t it? Things like Pocket have been attempting to work around not having this functionality for years: instead, we’ve made use of things like ShareKit for developers to share access with their apps between each other. The App Extension Programming Guide documents an example of the Action plugin as such:
An Action extension helps users view or transform content within the context of another app. For example, an extension might help users edit an image in a document that they’re viewing in a text-editing app. Another type of Action extension might let users view a selected item in a different way, such as viewing an image in a different format or seeing text in a different language.
Things in the iOS development landscape like x-callback-url, an open spec for sharing information between apps, have been trying to do something like this for years. I’m excited to see how the space can evolve now that it’s more or less been sanctioned by Apple.
Extensions doesn’t just cover things like sharing and inter-app sharing. Custom keyboards are now possible, as well. An extension like this is a global change, not just implemented for one app. It’s pretty exciting to see the Apple of 2014 writing an SDK that supports that.
*Kit
As someone who isn’t developing photo-intensive apps, PhotoKit has primarily one appeal to me: editing photos in tools like VSCO Cam without having to re-save or duplicate photos. Photos in an Instagram world have a strangely mutable quality to them — it’s rare to see a photo posted online somewhere that hasn’t been cropped or edited with filters. PhotoKit should make that a bit “safer”, and a bit more supported by the actual OS.
HealthKit and HomeKit are the most exciting prospects here. After grabbing the developer market share with killer tools, we’re starting to see Apple realize that they can pretty much write whatever damn tools they want. With a solid OS, can they branch out and build a toolkit for interfacing with an entire burgeoning market of home automation tools? Sure! Why not, right? In the technology space, there’s probably no two spaces growing faster than home automation and personal fitness/quantified self tools. In one iOS release, we’ve seen serious and committed support for both of them.
I’m torn on CloudKit. I haven’t had time to read through the particulars yet, but offering something that is essentially going to take the place of a backend engineer seems significantly more complex than Apple is letting on. Things like authentication and database design aren’t easy, and it’s a little concerning to see that responsibility just be handed off. CloudKit seems pretty similar to Parse, which is a pretty well-established backend-as-a-service, but I haven’t had enough experience with either to feel super comfortable with them yet. Of course, this is coming from someone who spends a lot of time writing backend code, so YMMV. ☺
Some other things: Metal looks great, but I’m not qualified or experienced enough in graphics rendering to know the difference, really. All I know is that seeing the Frostbite engine run on iOS is pretty amazing. A Touch ID API should’ve arrived with the initial release of the 5S, but regardless, it’ll be great to see it implemented in finance and journal apps.
All in all, I’d consider iOS 8 a slam dunk. iOS 7 was a little rough around the edges (and we still don’t know whether we’re capitalizing our letters or not), but it sure seems like now it was just set-up for what we’re seeing now.
Totally unscientific score: One devil horns out of one
Mac OS 10.10 (seriously?) Yosemite
Sure. 10.10. That’s a real thing. Is it that hard to just move up to 11 by now?
I was initially scared when I heard that the new version of OS X was going to have a design update. I like iOS 7, but it’s not my main computer. I’m a programmer. I don’t want my file system hidden, and I want to feel like the Unix parts of OS X are given just as much care as terrible, terrible things like Launchpad (don’t even get me started).
It looks nice. I don’t know, I’m not a designer! In fact, I’m colorblind. So my opinion on a design refresh to OS X is about as useful as it would be to a painter. But it is something I would (and will) use.
Continuity
Continuity is the specific term that Apple used in the WWDC keynote to refer to the new integration between OS X and iOS — strangely, it doesn’t seem to show up in either the consumer-facing marketing or the developer documentation, instead referring to the “Handoff” framework or enhanced iCloud support.
Regardless, it’s exciting to see that this approach, which was inevitable, was handled with equal care for both OSes. I wasn’t the only one that thought OS X would get the bad end of the straw here: I was concerned when 10.9.3 hid the “/Users” folder, but it turned out to be a glitch that was fixed within days. The file system is still there!
There’s not much to say here because this is a pretty big change that I think we’ll see more and more of as the years go on. The initial changes to the Messages app (sharing phone calls, SMS and iMessages between devices) is great and has been pretty much the only point of contention as an iMessage user. It’ll be great to fade from Spotify on iOS to the Mac. Make it so!
Spotlight
Spotlight has long been underused, and for beginner to intermediate Mac users, it could fix a lot of the issues they have with things like finding files and preference panes. Making it a first-class citizen on the OS by putting it directly in the center of the screen is a good idea, regardless of its dubious similarily to Alfred, one of my favorite Mac apps. While it won’t replace Alfred for any power-users, it’s definitely something that’s been long overdue.
iCloud Drive
I don’t trust iCloud. You probably shouldn’t either. Dropbox has been doing this for a loooong time, and they seem to have it down pretty well. I’ll give it a shot, but I’m cautious at best about this feature.
Totally unscientific score: OS X 10.8 out of OS X 10.10: Rancho Cucamonga
XCODE OMG YES
APPLE MADE A PROGRAMMING LANGUAGE.
It’s closed-source, looks like Ruby and Scala had a baby, and has some pretty gutsy performance claims (though the keynote mentioned that it is “93x faster than Objective-C apps”, I can’t find a screenshot of it — watch here).
Swift looks great! I’ve published an iOS app and have a couple custom OS X apps sitting around — Objective-C is not that fun. It’s not as bad as people say (really? The syntax is your complaint? Is anyone aware of how much less code we are writing on average?), but the Cocoa APIs are very verbose. Swift is syntactically similar to a scripting language, but covers all the Objective-C APIs. It should be a really great way for new developers or those scared of Objective-C to jump on the platform: I know I’m excited to start finishing some of my apps!
Some things that you may have missed that weren’t announced about Xcode in the keynote:
- A REPL thanks to Swift
- Live rendering of Interface Builder elements (as in no more build => simulate => quit => repeat)
- Storyboards for OS X
Along with that, the entire documentation library has been updated, with support for code examples in both Objective-C and Swift. If you ask me, it looks like the introduction of Swift has given Apple’s developers a bit of pep in their step: that benefits all of us, developers and users alike.
Email me when Kristian Freeman publishes or recommends stories