I’ve spent what feels like most of the night trying out the first developer preview of Windows 8, using an Intel tablet PC loaned by Microsoft for that purpose. The early preview is frustrating, in that many of what will be standard apps like Mail and Contacts are missing, but it is already obvious that Microsoft has done a great job with what I am calling the “Metro” platform within Windows 8. Here is Control Panel in the new user interface:
This is the touch-optimized personality of tthe new operating system, featuring a Start menu with live tiles like an evolved Windows Phone 7, apps that run full-screen to create an "immersive user interface", and swipe control to show application menus, switch apps, or access standard features.
It is a delight to use; but this is Metro, with its own Windows Runtime (WinRT), a native code API which is wrapped for access by either HTML and JavaScript apps (which also use the IE 10 runtime), or by C/C++, VB or C# apps driving XAML-defined user interfaces – yes, kind of like Silverlight but not Silverlight.
What about all our Windows apps? For that we need the desktop personality in Windows 8. Tap the Desktop tile, or launch a "Desktop" app, and it suddenly appears, looking much like Windows 7.
The problem: while Windows 8 "Metro" looks great, there are currently zero apps for it, or at least only those supplied with the preview, because it is brand new.
In truth then, Microsoft has not quite done what would have been ideal, which is to make Windows touch-friendly. That would have been impossible. Instead, it has integrated Windows with a new touch-friendly platform.
The key question: will this new platform attract the support it needs from developers in order to become successful in its own right, so that we can do most of our work there and retreat to the desktop only for legacy apps, or apps which really need mouse and keyboard?
It is a big ask, and we have seen HP with WebOS, and probably RIM with PlayBook, fail at this task.
Of course it is still Windows; but I do have a concern that a proportion of users will try Windows 8, find the transitions between Desktop and Metro unsettling, and stick with version 7.0.
Let me add these are very much first impressions; and that Metro really does look good. Perhaps it will win; but a lot of momentum has to build behind it for that to be possible.
One question I have is what the preferred story will be for devs of “business” apps. WinRT apparently is all Win8-only, but most business apps will probably need to support Win7/Vista/XP too.
So maybe you’d write a lowest-common denominator version in SL/WPF that hits Win8-XP, but then also have a separate version that supports Metro/Win8/tablets.
Vic
It is not one or the other, it is both at the same time, I think that is important to realize. There will be touch devices out there, they will be supported nicely with Metro, there are desktop scenarios out there that just won’t have touch, they will be supported. The multi monitor support with metro on one monitor and standard on the other was really nice to see that they have included.
Steven has a good ability to focus on important details as well as the broad picture, and he seems to stand for the word “no compromise”, that is what made Windows 7 successful, I think it will carry over to Windows 8 if they can get devices out there fast enough.
They are releasing pre beta software(for everyone, seemed like it?), that is bold, but I believe they want the momentum around metro to start, so people build apps.
It was hard to discern but WinRT will end up on the phone as well.
So SL is the common denominator, ifdef:ed SL that is. To me WinRT is the new .Net as it should have been from the start, native in the windows stack, it is very exciting to see if they can deliver on their promise. Because a native .Net will have a huge impact on performance and foot print issues that we see with standard .Net apps. I wonder what they will announce for the server side running .Net apps.
We all saw it too, an app can be both Metro and standard at the same time and switch seemlessly between the modes, that was cool, differnet UIs, same running instance. I hope that will be available for 3rd party as well not only MS.
That developer preview slate was just an amazing piece of hardware, I wish they were up for order, even if it just ran android… I would love to know the battery time on those, they must be shorter than needed, 2-3 hours, or can it actually be good too?
Yes, focus seems to be consumer these days. As an business app developer, looking forward to see how this plays out.
Vic, do you want to write two versions of the app and still be locked to Windows? Will this push more apps to the browser (of course it will depend on the app itself as what can be done) or will the business world live in the Desktop “app” and things will carry on the way they have been?
That’s an interesting observation about missing apps. While I’ve been more interested in the possible backlash Microsoft is potentially going to suffer due to such a major change in their UI, the fact is all of the underlying tech appears to be changing. “Legacy” apps on Windows 8 which are not designed out of the box to adhere to the Metro interface are going to be in a separate environment. It will be hard enough for users to adjust to this, but then add to that the fact that developers are going to be scrambling to update their apps to the new look…
I’m a Mac fan, but I still like to see other platforms succeed. I think there’s room for everyone to succeed and I think it’s healthier that way. That said, I’m not yet convinced this isn’t going to end up being a big misstep on Microsoft’s part.
Lynn: Mac OS is moving into an app model as well, changing the UI for touch, Ubuntu is doing it. I still think they make 2 bets, one for the table/phone devices and all in one touch screen PCs and another one for desktop/business/professional. I do not think the switch between the model will confuse consumers anymore than multiple windows do today. Also I think metro will be alot more used in non desktop form factors, and they important aspect here is that mobile factors will out number the desktop/latop format, they need to bet on that somehow will maintaining compatibility.
I don’t want to sound disparaging, but Linux variants still don’t count for a significant portion of platforms, barring Android on mobile which is it’s own ecosystem. While MacOS does now offer an App Store, it’s still not the only means of deployment the way the iOS App Store is today for Apple’s mobile platform. Likewise, I haven’t seen any indication yet that Microsoft’s App Store will be the only means of deploying apps on Windows 8. Still, I agree, Metro is more likely to be used on mobile devices than on the desktop, if for no other reason than desktop users are going to fall back to the legacy modes they are already familiar with whereas Windows Phone 7 is already using what is essentially the Metro interface. Still, unless WP7 receives a HUGE boost in popularity, I don’t foresee its deployment overtaking desktop deployment anytime soon.