They become available to registered Apple developers via the dev site (developer.apple.com/wwdc/videos). This year for the first time they will also be streamed live.
It's probably premature, but I'm hoping they will make good on this statement:
> When the binary interface stabilizes in a year or two, the Swift runtime will become part of the host OS and this limitation will no longer exist.
The current ~10Mb overhead is a hard sell to your manager when it will double your application size and slow down the development team in the short-term.
I absolutely agree with you. But as a startup moving at full speed and burning through investment, it's not just the size you need to take into account. It's the maturity of the tools, especially around unit testing (and automated testing), the stability of the IDE, and a whole host of issues where you don't have the luxury of wrestling with workarounds for an immature environment. You speak of program correctness but we have a whole suite of tests, analysis and coverage that would have to be migrated to work with Swift and obj-c and it's an investment we can't take lightly.
Having said that, I think it is about time to move to Swift and it's going to be in the spotlight for us soon.
If your target audience is countries with those limitations (and no wifi) and the 10 MB overhead is too much, you'd probably opt to stick with objective-C for a while longer, then.
Mobile over the air download limits. Nowadays with the limit at 100mb it's not as bad as it used to be when it was 50mb, or even 20mb back in the day. Still, you really don't want to go over that threshold if you can possibly avoid it
I don't get the hate for Objective-C. I think it does a better job with the "object-oriented semantics bolted on to C" requirement than C++, having extensively used all three. Something about brackets and semicolons speaks to me. Who knows. Maybe it's just me.
Now, having said that, maybe Swift is better for the use case, but it's not a reflection upon ObjC IMO.
I think newer versions of C++ are going in the direction of Haskell: a lot of additions that make functional programming much more pleasant to use such as lambdas, variable templates, etc, but I couldn't remember any new C++ feature designed for traditional object-oriented programming. So to me it's entirely understandable that object-oriented aficionados would find C++ less and less likeable.
Better OOP is definitely not the direction the language is going. C++11 added override and final. C++14 had no OO features. C++17 may get UFCS (i.e. f(x, y) == x.f(y)), which is either an OO feature or another step away from OOP depending on how you view it.
I know a lot of people hate Objective-C… But I'm a bit sad we aren't gonna get any updates to it anymore. I loved it, loved it more than C++. It was a very nice approach to C with classes.
Can you explain why it will slow down your development team in the short term? Because you have to find another 10Mb?
That's interesting. I remember the days of counting every byte, but I've never experienced it on iOS, just cause my apps were naturally pretty small even when being extravagant.
I'm quite scared, that they will in fact introduce so called "Rootless" kernel-level security - not beeing able to do everything with my computer would be a problem for me and I will definitelly consider switching to ie. Dell XPS with some linux on board.
When I read the rumors about it (which are mostly speculation), I essentially expect them to do the following (on OS X at least):
A) Make the Apple-provided system files read-only, maybe even put them on another read-only partition like iOS does.
They explicitly said[1] they were locking down system files 2 years ago at WWDC 2013: "in the future as we start to lock down the /System folder, you might actually get write errors.". As /System is already owned by root, write errors at /System means elevating to root won't be enough.
B) Assuming "rootless" is in fact a superset of what I mentioned in A, and going by the rumored name, they could be restricting what the 'root' user can do by default on OS X, matching some configurations of Linux where simply gaining root access doesn't automatically gain you total access. In other words, root can do things like load signed kexts, install system-wide software, etc, but not overwrite critical system files or perform various other sensitive actions that shouldn't be possible just by entering an admin password (which a LOT of software packages require during installation).
edit: also, they did include Hypervisor.Framework in 10.10 which is a significant engineering investment but doesn't seem to have a purpose at the moment (major virtualization vendors aren't likely to use it, neither are app developers. That leaves Apple...). I honestly wonder if they're planning to use it for security purposes in 10.11+. They did something similar with XPC, introducing it for use by developers first, and then in 10.10 they started using it extensively to isolate things in the OS (including Extensions).
This will be configurable, if only because Apple need developers in their ecosystem, and many developers won't be able to do their jobs on a machine they don't have full access to.
Supposedly that possible, but the real question is, if this still will be the case in 10.12? This concern was posted by someone somewhere few versions back, when they introduced code signing explicite allowance for apps not signed properly. This is next step. And as long, as I agree - security is important, judging on their past decisions, they may try to close their ecosystem (but hey, it's "their" hardware and software, their game and their rules - one can always decide not to buy Apple anymoe).
It would be very out of character of Apple if it weren't configurable. Yes, iOS is secure, but OS X has always been more open in this regard and there's really no reason to believe that they won't continue this trend.
After all, when they introduced App signing they could have made it so the mac would only run signed apps.
It's not a product show -- though they do showcase some products from time to time, they do so mostly for major new stuff (e.g. new iPhone model etc). And they usually don't show minor updates (e.g. laptop speed bumps etc).
What we'll certainly see is new iOS and OS X versions announced.
A new Apple TV might be shown (though there's rumors about delays with content deals etc). A new MacBook Pro not so much, especially since they were recently updated and they wait for Skylake.
Hopefully Apple will follow last year by focusing on making iOS more open to 3rd party products. I know iOS Finder would be too much to ask for, but I will anyway. iOS Finder, please.
For OSX, I would love it if they just left it alone (features wise) and put the entirety of the next year into making it stable and responsive.
Every day I wish I could downgrade to OSX 10.9, but I need the latest xCode. Unlike most people who complain, I love the new UI, like the look, but hate the slow graphics. Pleas, Apple, stop with the features, finish fixing all the bugs!
I'd just like to be able to open links in apps, eg so a Twitter profile opens the Twitter native app, rather than a web browser. Android has done this for years.
It's been widely rumored that iOS 9 / OSX 10.11 will be Snow Leopard releases i.e. they will focus on stability and performance. Hardly unsurprising since given they share so many components any improvements will translate to better battery life for the Apple Watch (which will be needed when Native SDK is released).
Last update was pretty good especially in case of 15" MBP, increase of SSD read to 2GB/s and write to 1.5GB/s is, in my opinion, significant. Current CPUs are still very good.
In terms of software: more of Android/Win features being borrowed (e.f. split screen "multitasking"). Possibly better icons and GUI overall (but only if Ive realised that he should stick to hardware).
Updated iPad Pro with Stylus and Kickstand,
connect a keyboard / keyboard cover and mouse/trackpad, able to run full OS X, but also touch optimized iOS Apps.
This is definitely the missing piece in Apple's development story. Microsoft has already moved to a one app, run anywhere model which is very compelling from a development perspective. Apple as yet hasn't provided their take on it.
I expect at some point soon they will deprecate AppKit, move everything into UIKit and use the iPad Pro as the vehicle to drive it i.e. it will only support iOS and Universal apps.
This is unlikely. Apple's stance on this matter is to make a clear division between professional computer (OS X) and entertainment consumer devices (iPhone, iPad) and corresponding interfaces (touch vs pointer). I don't see this changing. They even afraid to make iFinder or iCloud Drive for iOS.
Very unlikely just because Apple said so? I don't trust what they say because in the past they have talked down about certain tech only to turn around and implement it later. The biggest example of this is when they switched to Intel CPUs.
Apple isn't going to turn into Microsoft. And there's a reason why this is the wrong product. iOS is already OS X enough. Running a desktop operating system on a tablet doesn't make sense.
If you want a laptop that's as small and lightweight as the iPad you can get a Macbook Air.
You're sort of right, but laptops with dumb screens that are incapable of understanding touch events also don't make sense in 2015.
Even if it only gets used one in a while, every laptop (and desktop) should be able to let you scroll and tap buttons and UI elements with your finger in a pinch.
Software here is more expensive than hardware. Apple wouldn't do it unless they can provide a good experience for it, and they really don't know how to do that yet. It isn't just a few gestures, but the whole guerrilla arm thing, which is why you don't see touch being used often on PC laptops that don't have flat tablet configurations.
I totally used to agree with that, but it only took one hour with a touchscreen windows PC to make my MacBooks all feel broken.
I still mainly use my MacBook Pro, because overall, Windows just doesn't cut it for what I need to do. And I personally don't really need touch that often -- but every single time I do (on the phone, holding a baby, jerking off with right hand, whatever), it is maddening that it doesn't work.
Even if limited to just basic scrolling and tapping, basic touch screen support is still way better than nothing. I think you are right about why Apple doesn't do it yet, but I think Apple is utterly wrong not to do it for that reason.
It's a classic case of making the perfect the enemy of the good.
Truly awesome touch (and stylus!) support would be great; no touch support at all just feels incorrect.
Apple is going to talk down about it right up until the day that they do a 180 and finally catch up or surpass Microsoft in this arena.
The nay-saying Apple apologists around here just can't wrap their heads around the fact that Apple are masters of bullshitting people into thinking that they know best, even when they leave out obvious features like a second mouse button. I remember back in the 90's, Apple lovers would actually argue that a single mouse button was somehow better because you had to "think about it less". What a crock of fucking shit! :)
And they don't even talk down on them. These days, Apple's silence on the issue is taken implicitly as a dis, when in reality they could be working on it right now, having yet to find something that lives up to their rep.
> Running a desktop operating system on a tablet doesn't make sense.
Having a different OS for desktops vs. tablets doesn't make much sense. There may be some things you want to do different from a UI perspective, but that should be alterable within the same OS (and, anyway, with touch increasinging available on desktop and laptop displays and people frequently using tablets with, at least, a keyboard if not a mouse, you probably want desktop and tablet OSs to be able to use the same range of UI capabilities, just with different defaults.)
> Running a desktop operating system on a tablet doesn't make sense.
Of course it does or else people wouldn't be buying them. Microsoft is doing 4 billion dollars a year in Surface sales.
I've been using Windows on tablets for over a year now and nothing beats it. I get way more control over my tablet with Windows than I do with any of the mobile systems.
That would be too awesome. This is in fact the main reason why Surface Pro looks so tempting. iPad is gread device, but not beeing able to code on it is it's main problem (although I still love my device)
I don't know, I would bet on a new Magic Trackpad based on the new "taptic" trackpad in the 15" MBP. I'm really enjoying clicking anywhere and having it register properly and I can see it as a direct upgrade to the existing Magic Trackpad.
* millions worldwide will be able to watch sessions streamed live.*
I'm guessing that's hyperbole, but it's rare for Apple to exaggerate on anything. Does the keynote (presumably the highest volume streamed session) actually hit two-million viewers?
Many non-developers treat the keynote as the source of information for future products. Many proto-developers start with the iOS platform. Sounds a lot, but non unreasonable to me.
Are there estimates of how many people have registered a developper account?
