As a user of WP7 who is looking forward to WP8, I really really want to give them the benefit of the doubt that there is a major ace up their sleeve they can't talk about for competitive reasons. Something like Nintendo announcing the Wii remote early enough for Sony to quickly switch to motion-sensing controllers as well. Something huge and disruptive that is so critical to their success that none of their competitors can know in time to ape.
I can't recall a single time Microsoft has ever done that, though...
The final OS is one thing (hold back whatever you want there), but where is the SDK? Developers need that if you want them to be ready (and you can always put your super-secret magic in the first post-launch SDK update). To me, the MIA SDK is the troubling sign.
Direct X and a shared kernel with Windows 8 proper is already pretty big IMO. I have a Nexus S(Running JB) and I'm going to be giving WP8 a serious look when it comes time to upgrade.
It's not a matter of if, it's a matter of when. When developers make Xbox games(Xbox uses Direct X), and Windows games(again Direct X), it will be trivial for them to also target WP8/WinRT.
I'm sure I had a video of the "Kinect" concept (from Microsoft Research) from 4 or 5 years ago! I could be mistaken, but I think they released a lot of "concept" videos like that through out the years, and then released the actual device.
But what was a little surprising is that there were no handsets for the press to play with. There were some demonstration units carefully attended by PR personnel, and while we were able to get kind of close to them, the general rule was "you can look but you can't touch."
Looking at this [1] video I would say that this article is just another BS from Ars.
So they ended up doing (what sounds like, or may as well be) a complete rewrite and it's taking them much longer than they expected? Shocker!
It sounds like they're running the WP7 apps in a compatibility box. Is that right? One issue developing for Android is the huge API sprawl. Android has this immense surface area of public API now, and they only add new methods and classes every release (I'd love to know how the framework engineers on Android feel about this, because it certainly detracts from the app development experience...). So a compatibility box sounds like a good idea in that case, though making everyone rewrite their apps isn't a good idea...
In order to use WP8-specific features, developers will need to use a new development model. This is more friction for developers when compared to using comparable new features in iOS or Android, where the development model is unchanged, but new APIs are utilized.
One can only imagine the sorts of bureaucratic infighting within Microsoft as various development tool factions seek to discard the work of factions falling out of favor. In contrast consider how long Apple stuck by Objective-C as its chief development framework.
> "In order to use WP8-specific features, developers will need to use a new development model. "
It's not only the WP8 model. Windows 8 has changed, Server 2012 has changed, and so on... I have so many new keys and downloads available at MSDN, I have to take time to plan what I want to load into a vitual machine and start testing against.
Apple hasn't really consistenly stuck by Objective-C; they tried to push people in the Java direction for OS X "Cocoa" development maybe about 10 years ago. Never really caught on, though.
I thought they didn't believe that people would actually use ObjC compared to newer managed languages (pretty safe bet IMO, but wrong in this case it turns out).
Actually now that I think about it the the Java stuff must date back to NeXT. Probably a strategy to sell into "the enterprise", which they were pretty hot on at one point.
Apple's engineering group is much smaller than Microsoft's (or probably anyone else's). What's the GDP per engineer at Apple vs any other product company (like MSFT, Google or Samsung)?
EDIT: Which is to say, I believe bureaucracy is more likely in larger organizations and that the scale of MSFT's engineering org makes it unmanageable.
IIRC I looked that up, and Google's revenue is about half of MS's for 1/3 the employees, making it about 3/2 the revenue per employee. Not sure about Apple/Samsung.
"Whatever the cause of the delays—whether they're because Microsoft has bitten off more than it can chew with the kernel transition, or due to some other reason—the situation is now growing critical. It's not just that it's annoying developers; the delays are undoubtedly hurting Redmond's hardware partners."
I wonder if the Windows 8 kernel will also hurt Microsoft's chances for ever competing on the mass market. So far all of the announced WP8 phones are using hardware like Qualcomm's latest gen S4 chip, which could be a way for them to streamline the launch process, but it could also be a performance issue with WP8 (based on Windows 8 kernel), which I doubt is as lean and nimble as WP7 (based on Windows Embedded).
Well, targeting current high-end hardware (and tomorrow's merely average hardware) may be a better strategy than Microsoft's attempt with WP Tango, which targeted low-end hardware on a platform that was already behind the competition.
I would assume that Microsoft is smart enough to avoid pushing two separate ecosystems (WP7 and WP8) since they don't have a real foothold with one.
The core operating system swap. Taking a desktop operating system and shrinking it down to run on a phone is no easy task. In fact, it's a gargantuan undertaking. Add a bunch of new end user features and you get that the amount of technical work that is going into WP8 is phenomenal. However, the windows phone engineering team is superb. I have no doubt's that both consumers and developers will be thoroughly impressed come holiday season.
Taking a desktop operating system and shrinking it down to run on a phone is no easy task. In fact, it's a gargantuan undertaking.
iOS is a shrunken version of OSX, while Android is a barely shrunken version of Linux. Aren't all modern mobile platforms shrunken versions of desktop operating systems?
Who is to say those weren't gargantuan undertakings as well?
The relationship between Windows Phone 8 and Windows 8 is closer than that, though. My understanding is that both will make calls to the same WinRT APIs, meaning a significant amount of code (and the XAML defining the UI) can be shared between both platforms. That's a very different relationship than Mac OS has with iOS (where the UI differences are significant) or Android has with Linux (where there is almost nothing in common between X and Android's UI).
If there's one thing Apple has executed on, it's the synthesis of hardware and software.
That's an often overlooked key to a successful customer experience. Android has some decent hardware - Galaxy S III apparently but I haven't tried it yet - but unless you luck out and a carrier throws you an Ice Cream Sandwich or Jelly Bean, what's the point?
Microsoft knows they need something, but if they don't have some pretty awesome software they'll be in the same place Android is. (Sans patent lawsuits)
Microsoft has been crushed by Android in mobile. From a once dominating position, Microsoft phone OS's now get lumped in the 'other' category when market share pie charts are illustrated.
Despite patent licensing fees, Google's Android is free to use - or to fork as was done by Amazon. This is a resounding win for Linux. On the other hand Apple effectively gives away iOS when you buy or upgrade one of their mobile products. What niche is left for Microsoft - even if they do create some pretty awesome software?
Oh I wouldn't write Microsoft off entirely yet. Nokia is probably a dodo at this point, but it would be a bit hasty to forget Microsoft's $60 billion in cash [1]
I do think Microsoft is trying to apply PC-era tactics to embedded devices. It's like they're chatting up that hot girl who, it turns out, is a ninja assassin on a psycho killing spree. The only thing the embedded world produces with any regularity is bankrupt companies; what I'm saying is, companies like HTC and Samsung are used to an intensity that Nokia may never figure out.
Microsoft is capable of surviving in the embedded world, but they probably won't "wake up" for another year at least.
Agree that technically mobile is embedded, but mobile is more popularly considered a separate market from embedded. For example, while a phone and a washing machine may have the same sort of CPU, most engineers consider these separate markets.
Do you think that Steve Ballmer will be the one that wakes up, or does the board find a new chief executive?
I don't know if Ballmer will stay or not as CEO, but remember that he and his good friend Gates still have a controlling share of the company.
What separates mobile from embedded won't shield mobile, or Microsoft, from the race-to-the-bottom, cloning, and short product lifespans that make embedded such a hard business to be in.
Until Apple figures out annual upgrade cycles, Samsung will probably be number 1 again by spring 2013 as Apple consumers postpone purchases until the next annual Apple product release.
Samsung and others are free to attack the seasonal gaps in the Apple annual - holiday season - upgrade cycle.
I can't recall a single time Microsoft has ever done that, though...