I think Metro is great for a phone UI and as good as possible for a tablet UI. In my experience that's an inherent limitation of touch devices generally and tablets in particular.
And there are a few fundamental reasons why touch--and by extension, Metro--aren't ready for heavy desktop use. From a workflow perspective, I shouldn't be taking my hands off of the input device. If touchscreens made excellent keyboards, that
wouldn't be a concern--but watch a few third-party walkthroughs and tutorials on Windows 8 and an external keyboard ALWAYS makes an intro once the host starts flubbing the touchscreen keyboard. First, touching the screen means not seeing what's below
your hand. You have to put your hand between yourself and the UI to interact. This is often worse in multitouch uses when you might need to put both hands on the screen surface to perform an action, or "itsy-bitsy spider" your fingers across the
screen. If the mouse cursor was the size of a human hand, or even the size of a human finger, we would certainly call that a detriment to workflow.
Second, touch means you have to be within touching distance of the screen to interact. Using touch to interact with a wall-mounted or other fixed-orientation screen means that there is no "one size fits all" ergonomic answer to that problem.
I have seen touchscreens used in healthcare environments and when used for data entry, there are always complaints about placement. A traditional desk monitor setup isn't comfortable because you must almost fully extend your arm to reach the monitor,
and do so well above normal elevation for extended periods. The problem exists through standing or seated arrangements, because reading position and interacting position aren't the same thing for most people. If the screen isn't mounted at all--if
it is a tablet--you are either craning your hands into your lap or using some sort of stand to tilt it in your general direction. Tilting, of course, brings typing angle issues and so on, through the manifold pretzel of ideas of ways to simultaneously
get at and view a single input/output panel.
Of course, most monitors in the world currently aren't touch screens. And what gestures are best for touch aren't necessarily best for the mouse. I find that using a mouse in Metro feels like attacking the UI rather than interacting with it.
Just passing from the initial Windows 8 screen to the login page requires a sort of savage dagger thrust, and moving icons in the Metro start menu doppelganger isn't entirely unlike fishing in a congealing oil slick. So it seems that for the average
desktop user, Metro will be somewhat ungainly without introducing new features. I'm saying "no new features" because we've had desktop widgets for some time, and I'd argue that they didn't take off for a reason.
But what's really wrong with Metro, in my experience, is the fact that it takes over your screen. Historically, let's think of things that take over your entire screen without being resizeable: ancient software installation packages (that most
people learned to alt-tab out of), video games, Media Center applications, and Party Modes on some media players. Of course many things--perhaps most things--can take up your entire screen. But Metro and the listed examples
must. What do the listed examples have in common? They purposely preclude you from doing something else. Party Modes, video games, the Media Center, and vintage application packages are meant to be the only thing you're doing when
you use them. In fact that's the entire point of windows (and, dare I say,
Windows.)
That's a very fundamental statement, and to clarify, I am saying that
the Windows desktop embodies multitasking. And to build the long tail of that statement, I am saying that Metro hurts multitasking. It's obscuring the entire desktop. It's akin to locking the machine when you walk away. Why
can't I search for something while I'm looking at something else? Why does engaging the start menu preclude viewing two things side by side? I'm not in Party Mode. I don't want to just do one thing, there's nobody else to keep out of critical
functions sitting at the keyboard.
Consider that if Metro were really the right direction to go in, wouldn't most people be auto-hiding their taskbar? Right now, looking at my right monitor, Firefox is fullscreen. However, within Firefox I have 9 tabs open and can click on any
of those at any time. I can see them. And below Firefox, I have seven icons showing open applications on my taskbar. I can pull any of them up at any time. Windows 7 doesn't do two monitors terribly well, so I've installed DisplayFusion
so I can have a taskbar and start menu over there as well, that reflects which programs are open on that screen; and yes, there are applications open over there as well. At no point do I want to not be seeing these things. Emails keep coming in,
clients connect to me remotely, server consoles show alerts, and anyone I know could toss me a message through Live Messenger. Covering up the screen means I won't see those things when they happen, but that's my job.
Taken together, it's obvious why Microsoft is doing what it is doing. Microsoft is trying to leverage its desktop dominance into support for the Metro experience and attendant app market, by tying Metro into the everyday workflow of the average desktop
user. In doing so they are hoping to unify platforms and user experience. Unfortunately, the current desktop environment isn't ready for touch, and touch isn't ready for traditional desktop workflow. Metro proves that desktop and touch-tablet/phone
interfaces shouldn't be unified, and doing so will come at the expense of what makes the Windows desktop experience what it is.
