On Monday’s Keynote Apple showed off a much improved watchOS 3 operating system. When I saw the increased speed on the stage, I had my doubts.  As you may recall, when Apple originally announced the watch, we saw a lot of similar demos about how amazing and fast the watch was and we later discovered that most apps were simply too slow to use on the watch.

According to legend, when Apple first announced the original Macintosh, they had a souped-up version with extra RAM. Part of me wondered if there was some sort of similar magic trick at work on Monday. Then I got to take a look at the new software running on some developer friends’ watches here at WWDC the last few days and I’m relieved to report that the watch improvements are for real. Apps do launch fast and the watch is significantly snappier.

Apple is pulling this off by using some for the watch’s extra memory to keep apps in memory and allow background refreshes. Another reason for this significant improvement is that Apple has loosened up the watch’s battery usage. With the initial launch of the Apple Watch, Apple was very conservative about battery life. They did not want the story to be that their watch ran out of juice at 5pm. As a result, the watch is very stingy about power usage and I often end up with my watch battery only about half-used at the end of the day. With watchOS 3, they’ve loosened up the battery restrictions and push the processor in the watch a little harder.

The end result of all of this is that the existing hardware is much faster and more useful. Amen.

Another promising development is the watch user interface. There are only two buttons on the watch. One of them (the big one) was dedicated to the friends screen, which hardly anybody uses. I can sympathize with why Apple gave this feature such prominence in initial development. In Cupertino, all Apple employees have Apple Watches and this feature makes a lot more sense in those circumstances. That, however, isn’t true for everyone. So instead of doubling down on the friend button they’ve repurposed it as a dock button that lets you flip between apps.

So I’ve played with a few watches now using watchOS 3 (although I haven’t installed the beta on my watch … yet), I’m happy to report from WWDC that your Apple Watch is about to get a lot better.

I wrote yesterday about how Apple is responding to the question of how they provide interesting and useful services while preserving user privacy. My conclusion was that I like their explanation but in order to make this work, Apple has to not only write the check. They have to cash it too.

It’s still early days but yesterday I installed the iOS 10 beta on my iPad and spent time using iOS 10 on friends’ iPhones. I also tested macOS Sierra on my MacBook. This is just the first beta of a new operating system but I am initially impressed. I’ve been searching indexed photo libraries for pictures of dogs, beaches, buildings, and other objects and the device is finding them. (No luck on zip lines, however.) Overall, this local photo indexing solution is working better than I expected for the first iteration.

For some time now there has been an open question about Apple’s future. Specifically, with Apple’s interest in protecting user privacy, can they compete with companies like Google and its cloud-based user data whiz-bang features made possible by their storage of and access to user data? In other words, will protecting privacy cripple future Apple services?

Apple responded to this in yesterday’s WWDC Keynote. Multiple times speakers explained how they intend to bring great services and protect user privacy. There are two ways they intend to do this.

On the Silicon

If you had a drinking game where you took a shot every time someone said “on the silicon” yesterday, you’d have probably passed out by the end of the presentation. The best example of this was in the discussion of Photos. Apple Photos will now do face detection and search out other objects in your photos to create an index of images and their contents. In theory, I could search for any pictures of john that also has a cantaloupe in it and my iPad, iPhone, or Mac would find it for me. This is what Google is already doing on their servers. 

Will this work? I’m not sure yet. My guess is that my iPhone will pull this off but not as fast or as accurately as you’d get with the combined power of Google’s fully operational server farms. I’m not sure it has to work as good though.

Differential Privacy

The second component of Apple’s answer is “Differential Privacy”. This is a technology that allows Apple to anonymize user data as it passes through the Apple servers. Differential privacy parses a large data-set, using statistical science to learn about the sum total of the data without learning anything about an individual user. It sounds a little bit like voodoo but I spent an hour this morning reading articles about it and it seems like a real thing. Using differential privacy, Apple can learn from the sum total of our data but still not have any details on anyone. Because Apple doesn’t have user specific data, hackers and intelligence agencies also wouldn’t be able to access it. It inherently has limitations and I’m sure if they skipped all the differential privacy, they’d have better data, but this solution gives Apple something to work with while respecting user privacy.

Making Choices

After spending some time researching and thinking about all of this, I like Apple’s answer to the question of how they’ll continue to respect user privacy and move the ball forward. I don’t think their solution will match what Google is doing right now but as microprocessors continue to march forward, I think doing these tasks “on the silicon” is a real option. Right now all we have is words and we’ll need to see if Apple can actually cash the check they wrote yesterday morning but if they do, I’d be satisfied. For me, I think it comes down to a choice. I’d rather have 80% of Google’s features along with 100% of Apple’s interest in protecting my privacy than 100% of Google’s features with 0% of that privacy protection.

Jason Snell’s Six Colors website is hosting a Talkshow channel for the Keynote. It’s going to be fun. Tune in here.

This week MacSparky is sponsored by OmniGraffle, the best tool for creating precise, beautiful graphics. I use this application to create diagrams, logos, flowcharts, and everything in between. OmniGraffle is both powerful and easy to use. Using OmniGraffle’s built-in stencils, you can build diagrams without a lick of artistic talent. If you don’t find the stencil you want, you can download more in Stenciltown. Stenciltown is a curated collection of OmniGraffle assets made by people much more talented than me. Whenever I’m making a cute little image or diagram for a presentation or video, Stenciltown always bails me out.

OmniGraffle is designed by the OmniGroup’s crack team of programmers and UI designers so you don’t get lost in menubars. When you are done, you can export your graphic to numerous formats. Check out OmniGraffle for Mac and iPad today.

Yesterday Apple announced that there will be new pricing options in the App Store. The biggest change involves the addition of subscription plans for apps. We’ve been hearing for years how software as a service and subscription plans are the future. From the developer’s perspective, this gives them a steady stream of income and a basis from which to continue application development. I understand why developers want to move in this direction. However, in order for this to work there has to be participation from consumers. Frankly, I’m not so convinced that will happen.

I think, in general, it’s easier to pay $12 once then the thought of paying one dollar every month going forward. Now multiply that times the 20 or 30 apps that you really love and things just get crazy. I’ve already received several emails from readers and Mac Power Users listeners complaining about the idea of subscriptions for all of their favorite apps. Put simply, I’m not sure consumers will cooperate with this new model. I would like to be proven wrong but developers may find that subscribers are a lot harder to come by than they think.

The 800-hundred-pound gorilla on the sofa for me in all of this is the upgrade pricing model. Is Apple ever going to publicly explain why they refuse to implement upgrade pricing in the App Store? While that model has its problems, it has been a good solution for productivity software for many years and consumers are clearly comfortable with it. You buy an app and after a year or so, the developer updates it with additional features and, as an existing customer, you get a discount on the new version. Apple has never explained their resistance to allowing upgrade pricing in the App Store.

It is a tough time to be in the App business. While Microsoft can get away with a subscription plan, I suspect a lot of smaller applications will not. Perhaps because Apple makes all of its money off hardware sales, they don’t seem very sympathetic to app developers. Finally, this race to the bottom in pricing has led to a market where it’s difficult to make quality, sustainable productivity applications. I honestly don’t know what the solution is, but I have a lot of doubts about subscription pricing being the panacea that people seem to think it is.