decades of fun

The Microsoft Copy Machine


I’ve been following the Windows8/Metro debut online and it struck me that this is one of the largest examples of Microsoft being a “me-too” type of company I’ve seen in a while, complete with a fatal flaw in their copied version. My rambling thoughts:

First, the basics (if you haven’t watched the previews yourself): Windows 8 is coming, perhaps by the end of 2012. Unlike Apple, MS likes to show things off way before they are finished, and they are focusing on previewing their new touch-based Windows shell called “Metro”. I call it a shell because it runs inside of/on top of traditional windows and doesn’t appear to have it’s own boot routine, networking, filesystem, security, or hardware driver layer. Those sorts of things appear to still be handled by Windows. Of course it has a snazzy modern-looking interface and can be touch manipulated. Like Apple’s industry-leading iOS, it also has a curated application environment (aka a company AppStore).

As far as userspace, Metro looks far better than past MS offerings, but it is still playing catch up. Their guy actually had to explain to the developers why using the same font, font weight, and alignment makes apps look better (it’s 1983 again). He actually got applause when telling them they can use the whole screen for their app. Basic UI animations, scoffed at in OSX 10 years ago by these same Windows programmers, will be considered essential practice in Metro to make your app ‘not look like crap’. I guess he just called all past and present Windows software crap, haha!

This is a classic case of copying an industry leader with a ‘me-too’ mentality. On the surface Metro has most of what iOS has, just renamed and changed slightly. It looks nothing like iOS at first glance (although the apps will look similar) so Microsoft will ignore iOS, embrace it’s core fundamentals, then claim they did the ‘serious R&D’ to make this a ‘serious’ product. iOS 5 is shipping next month to 90+ million devices, and should make this Win8 preview (shipping 12+ months from now with no comparable hardware currently in existence) show how late to this game MS is.

This isn’t to say they won’t be successful, at least by their own standards. With Google around it’s now a 3 horse race, and MS owns all sorts of advantages with their other products (XBox, Back Office, SQL Server, Office, Outlook, dev tools, vertical market devs). Their installed base and distribution network is literally everywhere. But they are no longer the big dog in either size or profit, and in mobile their market share is smaller than the worst days of the mac. If MS got 20% share of Tablets with Metro by 2014 they’d be besides themselves. That’s their game though, slow and safe, full of screwups along the way. It has been that way for 30+ years.

While similar at first glance, where Metro and iOS differ is critical. Apple and Microsoft draw the fundamental user model completely different. iOS is essentially an executable sandbox that runs on the OSX kernel with many OS-type tools available within. It’s an pseudo-OS optimized for battery life, touch, and mobilily. Apple makes all of this available to devs through modern Libraries and API’s available in OSX to the iOS programmer. The app is compiled into iOS code and delivered to the devices through the app store model. The iOS devices themselves can boot into iOS but cannot load the full OSX. Apps need completely separate code for the PC (OSX) version and the mobile device (iPad/iPhone/iTouch/AppleTV).

Microsoft sees three classes of devices (much like Apple): phones, tablets, and PC’s. Microsoft also plans on having two interfaces to cover these classes, just like Apple. But Microsoft puts Tablets with the PC category, which to me is a fatal error in their design because it removes mobility as a focus. Mobility is the key and MS can’t or won’t focus on it.

If your CPU is plugged in and sitting under a desk (and was designed for it) the low-level software that is the OS should know that it has no limits on it’s physical resources (power in/heat out, data entry helpers, and ports/slots for add ons). Laptops add all sorts of management of this but they are still running full featured CPU’s, giving us battery life of only about an hour more per charge and more heat than laptops 15 years ago! Very slow progress I’d say, since the desktop OS, apps, and content have gotten bigger and more complex along the way.

However, if your CPU was designed to be battery powered and mobile, the OS has to be on board with that mission for maximum performance. Code intended for desktop-usage need not be carried around and executed on mobile devices. Ever wonder why a portable battery powered iPad2 with 10% charge can play movies, games & GarageBand as fast as a quad core 2ghz desktop? Because the OS, all the way down to the steel, has been designed to run on that kind of box. Unplug from the wall and that desktop would need a car battery to work in the backyard, and still be louder, heavier, and far less modern feeling than a touch tablet.

So by trying to make a tablet a full PC, Microsoft is shooting the whole idea in the head. If a tablet was just a PC we would have keyboardless PC’s running touch interfaces everywhere by now. PC’s have been out for 30+ years after all. But we all realize that a real PC needs a keyboard, pointing device, sturdy table, chair, and an outlet to get real work done. I don’t see that changing in a long time.

The tablet’s needs are different: UI speed, battery life, no heat, and long standby. These are all things that Microsoft is ignoring in their debut, and should keep them well behind the iPad (and the MacBook Air) in the marketplace for the next few years.