Somewhat a downer this year. Two big side-projects (where one was due to get released this fall), more or less canned by new features in iOS 9. I'm not sure this is actually something bad. It is great that these new features are added to the system, and now it will get much better than what a third party could accomplish. The sad part is that I've been working on and off on these for several years, so it is a lot of work that now is just a waste. The lesson here is you must focus and prioritize, you can't let a project linger for years in the fast-paced mobile world. Either do it now, or move on to something else. Who knows, they might get a new life in the future, in some other form. And maybe I'm judging this a bit too fast - we'll see.
Also newsstand goes away, turning newsstand apps into normal apps again. I don't know if this change is necessarily bad, the newsstand folder has been quite awkward since iOS 7 (it made at least some sense when it was see-through). I just hope all my newsstand customers agree.
Internet over http from apps will now fail, there is a workaround you can supply in the apps plist. I guess every app will contain this entry when iOS 9 is released. Then Apple will put restrictions on this, next year or so, and everyone will need to update their servers to https anyway. In the long run, this is a push for a good thing, but it will add so much to my inbox... I will just try to hold it off for as long as possible.
Swift's developers finally gave in and built their version of exceptions, the "goto of the 90's", because "that is what developers are used to". There is perhaps no other positive wording available. This is a hard problem, and I guess they had to do something, but I just wish they had come up with something smarter.
On the other hand, swift is coming to Linux. Which means I can move all my server code to swift, and finally get a good server language. I was almost about to switch to Node, but now I will wait for a bit longer. Swift did also get a host of good improvements, making it a viable alternative to objective-c. I haven't been able to use Swift yet, since I still support iOS 6 for most apps, so I can't really say that it is finally good enough and time to switch, but it really seems like it.
Metal did also become a first class citizen, which I was waiting for (did not think they would do it this year though). I don't make any games (yet) so it has no effect on me, but I think it's a very positive signal that they are focusing on game/graphics API:s. It's not Apples strongest card, so I like that they give it a boost.