The Tale of The Three Tablets
This post was going to go up two days ago. Lucky I procrastinated, because Microsoft announced ARM support in Windows 8. This news makes my thoughts a lot clearer regarding tablets, slates, journals and phones. On the other hand, it’s gonna make the future of Embedded and Xbox a lot cloudier. Allow me to explain:
The 3 tablets—whatever might that mean, eh? Well, I was hinting at the three different dev pathways being pursued within the company to figure out how to properly approach the “new” market possibilities enabled by first the iPhone and now iPad.
The first of these was a combination of Windows 7 and Windows Phone 7. The idea was that you’d put an ARM SOC, a 8+ GB SSD and a small battery behind a 7 to 9 inch screen, and that screen would then dock into a netbook chassis, which had all the usual netbook parts (maybe swapping out the Oak Trail Atom for a 2011 Core i3 ULV or Fusion APU). The dock connector would allow the screen to be powered by the netbook bits, get juice from the larger battery/AC adapter and share data on the SSD & harddrive. When docked you’d run Windows 7 and as a standalone slate, you’d get Windows Phone 7.
Microsoft may yet announce this today at CES, depending on a lot of internal politics whose resolution I probably couldn’t ever guess. The first of these franken-PCs would be made by Samsung and Dell and be upgradeable to Windows 8 and Windows Phone 8 next year. They would not be cost-competitive with the iPad.
The second path, that I was particularly hoping for, but is now unlikely to happen was that of the Courier. Microsoft would, if it went down this road, not use either Windows 7 or Windows Phone 7 for their slate story. Instead they’d start afresh with an all new UI built on top of Windows Embedded 7, specifically for touch + pen operation. The first iteration would be built by Sharp, and be a single-screen 7 inch device. It would run on an ARM SOC, with battery life and price similar to the iPad’s. You’d be able to use everything just fine, using only your fat fingers, but if you added a stylus, you’d be able to do a lot more. The second iteration would then have dual 5 inch screens when that could be achieved at a competitive price.
The third, and now near-final path is one Microsoft hinted at today. Windows 8, as I have detailed previously will be the go to solution for screens as big as 27 inches to those as small as 7 inches. Anything between 5 and 3 inches will run Windows Phone 8 instead. (I know you can “run” Windows or Windows Phone on any screen-size technically speaking, but these are the optimal UX-based rules made by MS ). Windows 8 will rely on the Windows Marketplace to enforce UI guidelines on third-party apps. You buy an app once and it shows you which form-factors it supports. If you then run it on a “Windows Slate” you get a touch-UI, if you run it on a “Windows Tablet or Journal”, you get touch+stylus-UI. If you run it on “Windows Laptop” you get a mouse/keyboard-UI. And obviously Windows native bits do the same. Non-slate and all Enterprise SKUs will likely still be able to run non-marketplace apps.
Eventually, in the Windows 9 timeframe, the phone product moves from an Embedded core to the common Windows core. UI continues to be form-factor specific. This was the target with one of the Windows Phone 7 dev pathways before the “reboot” but Windows on ARM wasn’t doable in the Windows 7 timeframe and hardware was too weak on the ARM-side, while battery-life too poor on the x86 front. Metro design guidelines will have become pervasive enough by then.
What this third pathway does, however, is complicate the embedded and Xbox story. WE7 was getting close to Windows in capability and was going to be used in the TV set-top, DVR and software platform product categories; it may now be restricted to automotive and vertical niche markets over the years. The Xbox guys were told to go after integrated IPTV, but again, with ARM support Windows Media Center becomes relevant again.
This leads us too far beyond tablets though, so I’ll address it in my next post.