WWDC has wrapped up and so now it is up to us in the world of iOS to make sense of all these newly unveiled features. Now, being a developer who is often working in the enterprise realm, I tend to look at many of these features from a slightly different angle. When people hear about ARKit adding augmented reality to iOS, they probably think about the implications for games like Pokémon GO and interior decorating apps that allow you to see a fake end table super imposed next to their couch. My first thoughts are about it’s possible use in retail and warehousing to help highlight objects of interest or present additional information for the user. With this post, I’ll be going over some of the new features and SDKs in iOS 11, and how they relate to the enterprise and working world.
ARKit – Previously developers had to do all of this by hand, and truth be told, it’s not easy. Super imposing into an image isn’t very difficult, but however, super imposing graphics in a 3-dimensional space with depth, that is also moving, is extremely difficult. ARKit does the heavy lifting for developers by locating features and planes, and it allows you to create 2d or 3d objects via SceneKit or SpriteKit to super impose onto these features and planes. As stated earlier AR can be useful to locating and highlighting items, but when combined with some other technologies it can be so much more. For example, when paired with iOS 11’s Machine Learning and Vision frameworks, mechanic might be able to have their device see an unknown engine part, detect what part it is, and then super impose the documentation for that part on half of their screen. Obviously, this technology on a phone or tablet has some limitations, but as smart glasses become more mature they will be a fantastic candidate for augmented reality in the enterprise.
MLKit – iOS 11 also brings us a general-purpose machine learning framework that can work with models that the developer has trained. MLKit works on top of Accelerate and Metal, enabling it to leverage GPU processing. One advantage to this setup is that the model can be run directly on the device, meaning that unlike most current mobile Machine Learning setups it will not require an internet connection. This also allows for the user to have more privacy and security as well.
Vision – Vision is built on top of MLKit and handles image recognition. This includes recognition of bar codes, objects, faces, and text. Naturally the barcode recognition is a great fit in retail.
Business Chat – iOS 11 also introduces Business Chat, which I have had the opportunity to test drive. Business chat allows users to communicate with companies’ customer service or support departments within their Messages app. The advantage here is that the user doesn’t have to install any company specific app, and the companies’ can provide a solid customer service experience without having to invest in their own dedicated app. Business Chat supports making payments via Apple pay, presenting the user with menu options generated by the Business to select from such as available appointment times, sending of attachments and signatures. Lastly it supports integration to 3rd party Customer Service Platforms, making adding support for Business Chat very easy for businesses. Unless I am missing something, I can really see this being a popular feature for businesses given how much they gain for relatively little effort and expense.
NFC – This is something us developers have been waiting a long time for. NFC allows for a much greater level of security and access control for enterprise apps. For example, now an app that might have sensitive financial data can require the user to use their RFID enabled company badge to verify that they are indeed a member of the accounting team before showing any data. In warehousing this allows the device to instead of reading a barcode that’s printed on a box or pallet via the device’s camera, read the RFID tag on it instead. Unlike cameras, NFC readers are fast, they don’t care about lighting, and they don’t need to be a certain distance away.
SiriKit – Another item on many developers wish list is Siri support, and now it is here. With SiriKit a developer can make an extension that takes in commands from Siri. It can even act on those commands when the app isn’t even launched. For example, a retailer with a buy online pickup in store service could build an app that the associate can update order status with via Siri. The syntax could be something along the lines of “Set pickup order 523182 as ready for pickup”. The app extension would then receive that information and launch a request to mark order 523182 as ready for pickup on the server. This is great because it allows employees to not have to find the app, find the order, and then mark it themselves. SiriKit also can be very useful in environments where the user may not have 2 free hands to enter data. I also notice that you can use Siri from the lock screen on newer devices, making it truly hands free.
Overall iOS 11 looks like it could bring a lot of great feature to iOS in the enterprise and certainly makes me a little more optimistic about using iOS in the enterprise realm. Now if only they would make a ruggedized device that could take a hit…