From a recent episode of the Under the Radar podcast, discussing new Apple technologies that are announced at, eg, WWDC:
I think increasingly I've come to more of the conclusion that, if I don't get [the new technology Apple's releasing], if I don't think it's going to be ... if I don't really see the immediate utility of it or think it's really cool, then there's a good chance that I'm probably more typical than not. There's a chance that I'm missing the next big wave, but it's also just as likely that I'm not, and that they're actually not going to be this big thing. Because the nature of these new announcements is so often they are solutions going in search of problems, and they're often very cool and technically very capable. But it's difficult. [emphasis mine --mfn]
This is exactly the current issue at Apple when comes to software (well, this and a fundamental mishandling of QA, though they're closely related): They dream up use cases, target very specifically for those use cases, and don't spend enough time dealing with edge cases, even useful ones.
The problem here is twofold:
- Often Apple's use case misses. Exhibit Ping, but even the Apple Watch to some point & the current MacBook Pro's crappy keyboard, and then the engineering time is wasted.
- People use the software in secondary ways Apple didn't intend, and d/w/on't support, meaning those secondary use cases don't work well.
For the second point, I'm reminded of this horrible experience Jason Snell wrote about, or, perhaps better, this case where I couldn't update a Google OAuth password in macOS preferences.
For Apple, there's the Happy Path or there's no path at all. That works a lot better when designing hardware than software.