One of the never-ending challenges associated with designing the Office 12 UI is managing screen real-estate.  One of the tenets of our design is to leave as much room as possible to work with the document.  On the other hand, there are always more and more features competing for space, trying to infringe on the document space from all sides.

Some days I feel like our main job is playing "defender of the pixels."  Trying to find ways to avoid having features take up unnecessary space that you'll never get back.  Everyone wants their feature to be more prominent, but if every feature is prominent you end up with a pile of undifferentiated junk.  On the other hand, if you find the right home for each feature, you end up satisfying your discoverability criteria without upsetting the balance between UI and document.

Here's an example.  I keep my car keys in a prominent place in my house, near the door.  If I don't, I risk losing them and not being able to drive the car.  I use the keys multiple times every day, so having them out in the open is crucial and efficient.

Tucked away in a drawer in my kitchen are a bunch of new AA batteries.  Whenever a device runs out of batteries, I go to the drawer and get the batteries I need.  The batteries are far less prominent than the car keys, but I can find and use them just as easily.  Drawers let me keep objects organized so that they are always there when I need them.

Now, let's assume that I promoted every object in my house to the level of prominence of my car keys.  My living room would now be piled to the ceiling with holiday decorations, playing cards, vegetable peelers, floppy disks, magazines, remote controls, taco seasoning packets, trombone mutes, and a thousand other things.  Everything in my house would now be "more prominent", yet nothing would be easy to find.  And worse, there would be nowhere to walk around anymore.

We've been given an opportunity to start over in Office 12 and correct some of the priority inversions that have been inflicted on the product in the past.  Yet, a key design challenge remains convincing people that every feature doesn't have to be front-and-center in order to be discoverable or usable.  And it's a vicious cycle: once you artificially inflate the prominence of one command, it contributes to the clutter which requires you to promote another command to compete with it.  Pretty soon, your design is out of control.

Screen real-estate should be the most highly-prized commodity; as a user interface designer, it's the only building material you have to work with.  And once it's gone, it's gone.

Fight interface squalor.