WWDC 2016 Wishlist for iOS Developers

[[164900]]

I don't usually write articles like this, but I felt it was important this year: Apple is becoming more open and receptive to feedback than ever before. iOS has made huge leaps on the iPad in the past year with iOS 9 and the iPad Pro. With that in mind, I wanted to document all the things that came to my mind that I'd really like to see that would make iOS as a computing platform even more comprehensive.

The following list is what I'd like to see on Apple platforms in the next few years, in no particular order, starting with the simple and obvious. Most of these have been mentioned before, but they're often forgotten. Most of the list is about iOS, but not all of it, and I'll be frank up front that I don't think OS X has a future at the current pace. Even if you disagree with me, there may be something you'd like to see below.

Split screen function---allows you to divide the same APP into several windows

This is an easy feature to implement. Currently, the split-screen feature on iOS only allows two different apps to be placed next to each other; the obvious next step is to allow the same app to be divided into two windows, so that you can have two web pages or two documents side by side in one window.

Windowed Open URL

Another minor improvement, but this allows apps to choose whether to open a link or application in two windows side by side on the same screen or in a new window to replace the current app. For example, when a link in an email or message is opened, it will automatically open in Safari on the same screen instead of kicking the current app away. Twitter's interaction logic on iOS is very smart---if a link is clicked on Twitter on the same screen, it will directly open a Safari window on the same screen.

Customizable window controller extension

Apple has established an ideal way to present a view controller from one app to another in terms of extensions. This is used throughout the OS, but most notably in Safari's view controller. Apple has identified a key area for this application - this is how photo editing extensions work today. I would love to see this feature registered by apps so that it can be used by more apps and features.

Quick View Builder

Likewise, the Quick Look feature seems like an obvious system extension to add. OS X apps can register a "Quick Look Generator" and generate thumbnail previews of document contents, which other apps can then render. As document processing becomes more common on iOS, it makes no sense not to add an extension for Quick Look.

System-wide painting and marking views

It's a simple wish - Apple has created one of the best drawing/markup views in Notes that is perfect for use with the Apple Pencil in any app, and it would be great to give developers a standardized customizable style that they can implement in their apps without having to reinvent the wheel - a drawing engine like this would be a very complex OpenGL/Metal renderer that would be difficult to reimplement.

Add a Command key to the iPad keyboard

Given iOS' support for UIKeyCommand and keyboard shortcuts, I'd love to see a command key added to the iPad's on-screen virtual keyboard. It should be easy to understand that it should only be useful when the virtual keyboard is visible (such as in a document editor), but it should be possible to modify app-specific shortcuts, and it will also help iOS developers realize that apps should have default shortcuts. In an era when most iPads support keyboard peripherals, I think this is an important step.

Drag

A lot of people have been talking about dragging on iOS since the addition of split-screen multitasking. On the surface it's an obvious feature to add, but when you think about it, there are many ways this is damaging to the OS. Finding a way to do dragging without damaging existing OS controls, while still making copy and paste faster, is not as easy as it might seem. Still, I feel it's worth pointing out that it makes a lot of sense to use it directly on a touchscreen.

Native apps on WatchOS

WatchKit is a great niche developer component; capable of being a WatchOS app platform. Unfortunately, third-party apps on WatchOs suck, and the original WatchKit doesn't help much with WatchOS 2.0. If watchOS is to succeed as an app platform, I think it should be capable of running real (read UIKit) apps. Since there are already people dabbling in it, I'm not sure this first generation of hardware is good enough to do that without compromises. So my hope is that better hardware will be able to do that in the future. Third-party apps need to perform as well as native apps from the manufacturers. WatchKit is a piece of shit.

AppleTV controller situation

A very easy problem to solve: currently a game running on tvOS cannot be played with a controller (unless, of course, Activision wants a hardware accessory, which is obviously hard). All games must support the Siri Remote: The problem is that the Siri Remote sucks for games, which means that every developer trying to make any complex remote control action on tvOS must include some ridiculous, self-righteous, and basically useless remote-only mode. This is a policy issue - I understand where Apple's intentions are, but it's a super frustrating experience caused by a terrible platform with no choice.

iBooks Author on iOS

iBooks Author is a derivative of the iWork suite and seems to be a perfect fit for iOS. For book writers, iBooks Author on iOS could mean a fully integrated writing and publishing channel that doesn't require a desktop computer. You can even create your own multi-touch to enhance the performance of ibooks on the device. To me, it feels like a new silly addition to the iOS camp.

All system-provided applications should support split-screen function

I'm surprised that there are still a bunch of system apps on iPad that don't support split-screen multitasking. I guess it's security related, for example, if you want to view iBooks in full screen, you need to use the iBookstore. Obviously the App Store, iTunes Store, and Apple Music should support split-screen. It's hard to imagine why a non-game app wouldn't support split-screen (although I know there are some third-party apps that don't have this option for some reason). Windowing shouldn't be optional, especially for system apps.

The simple ones have already been said...

A standard unified App platform between iOS and OS X

Now, I'm pretty sure OS X is a dead platform. It has lagged iOS for years, often taking a long time for features to be implemented on OS X after being available on iOS. Apple needs to create a unified app platform between the two operating systems.

This doesn’t mean the desktop will only run iOS apps, just like tvOS won’t only run iOS apps. The same ideas apply: a shared code base, few platform-specific elements, and a user interface optimized for the main interaction modes of the OS.

I think this should be UIKit-based. After tvOS, we can no longer say UIKit can't run without touch, we know that's not the case. It's all built on the same Core Animation foundation, so at least for now you'll be able to interleave AppKit views/layers in hybrid UIKit Mac apps. AppKit itself should have a lag transition period, just like Carbon before it, as new features are gradually built into UIKit-based frameworks. AppKit should remain a desktop app, not iOS, and then gradually fade out and be replaced by compatible apps bit by bit.

In this way, iOS (mainly iPad) and OS X can grow together, and features in one can be more easily compiled for use in the other. iPad apps will have a way to transfer to desktop computers, and desktop apps will also transfer to iPads. The two platforms will evolve and grow as a whole, rather than one growing at the expense of the other.

Xcode for iPad

I've been asking for this for a long time. I said I wanted an Xcode for iPad development, which meant "writing code, debugging, and deploying Cocoa Touch apps on the iPad, without having to use a Mac." Most likely, such a project would work nothing like Xcode on a desktop system. It's likely that it would only work with Swift (which would make me sad), so I guess Swift Playgrounds would be included. Still, a complete software developer toolchain is a big missing piece for the iOS software ecosystem.

There are some great apps on iOS that have successfully done something similar, Pythonista is a good example: by bridging C to OC code, you can fully run the Python IDE on the device. It is powerful enough to let you interact and rewrite your own UI using Cocoa (in Python), but it is a little disturbing that Apple is said to no longer allow such apps and will remove them from the App Store. That will be a bad situation, and people have to consider where the boundaries of what iOS can do are.

File and disk management system on iOS

In the beginning, iOS tried to do the right thing with a file management system, but for nine years, this imaginary physical file management system didn't exist or actually happen. Finally, we have the iCloud Drive app and third-party file provider apps, but we still can't interact with external storage except for importing photos. I think it's time to implement this functionality at the system level: allow document pickers to open files in external storage appropriately, and allow applications to copy files to external storage. On OS X, the file picker provides a sandbox that you, as the user, can choose an app to access. In this model, the file system is built-in to ensure security, but not virtual non-existence.

Terminal environment for iOS

Worth a try, right? I'd be happy to see a terminal/BSD environment on iOS, even if it was confined to its own sandbox, not the entire OS filesystem. Let a techie create something a techie needs that can't otherwise be handled by GUI iOS. The only way I can see Apple doing it is if it's completely separate from the rest of the OS, and even that would be a nice step forward (or step back, depending on how you look at it)

iOS takes over OS X apps

Since the iPad is increasingly replacing things that the Mac used to do, it seems to me that Apple should also close the gap between the apps on the two platforms. I would love to see TextEdit, Automator, Font Book, Keychain Access, and external storage support. Disk Utility and TextEdit may seem trivial on the Mac, but it's kind of crazy that there's no built-in text editor on iOS that can access TextEdit files on iCloud Drive. Automator is almost unused, but those who do know how powerful and useful it is. In fact, Workflow, one of the best third-party apps on iOS, is very much like an extended Automator. Font management and keychain support are other apps that don't have third-party equivalents on iOS, even though they're important to certain users.

Without iTunes, iOS devices can install the latest OS in Recovery mode

Now, one of the only remaining reasons to connect to a desktop computer from an iOS app is to install an OS, and patching this is undoubtedly difficult, but NetBoot & Internet Recovery have been on the Mac for too long. Either I'm wrong, but I remember that the first generation (x86) Apple TV could re-download its OS from the Internet in Recovery mode if it failed. So ultimately I think iOS needs an extended Recovery environment to allow it to recover itself.

'AppleScript' for iOS

Perhaps it is a forgotten tradition, but not long ago AppleScript gained support for the new Java scripting language, and perhaps AppleScript also has a place in the iOS ecosystem. The existence of AppleScript itself is a great thing - it was launched with System 7 and it was one of the first programming environments I learned.

iOS supports more USB expansion devices

A hard sell, especially for apps made for the iPhone is a big deal for Apple, but there are all sorts of devices that can be used with iOS via USB adapters, in addition to audio, keyboards, and mass storage. I want to allow developers to write user-mode drivers to talk to existing hardware - for me, that means things like capture cards, TV tuners, serial adapters, external cameras, and input devices. Every single USB device needs an MFi-certified chip and certificate, which puts a big dent in accessory apps. You can buy a pre-existing Apple MFi RS232 modem adapter and use their approved SDK to develop apps to talk to a Newton, Raspberry Pi, or Arduino. You can't use Apple's USB adapters to do the same thing as non-MFi-certified adapters. I don't think Apple will change this, but I hope so.

Repair Mac's App Store

Finally, here's the big wish: the Mac App Store is clearly no longer viable. Simply put, sandbox restrictions and monetization issues have driven away too many long-standing and well-respected Mac developers. Those who stick around often create both MAS and non-MAS versions of the same app, and the non-MAS version is often fully functional. Third-party developers have forced MAS users to use their non-MAS versions of their apps. This shouldn't be allowed to happen - it's bad for developers, and bad for users. All Mac software should be readily available from the Mac Store. Microsoft's Office suite, creative cloud services, etc. Apple should do everything in its power to make the App Store the only way to sell Mac apps. What's worse is that now Apple's software available through MAS doesn't follow their own sandbox rules, and the same restrictions drove everyone away. I'm sure of this, and probably still will be. It's a joke now, and no longer has leadership.