Paul McAleer

Windows 8: Inconsistency is a Feature

Paul McAleer

You have probably seen this floating about the web.

That’s been featured in a lovely piece from Microsoft’s Steven Sinofsky, writing about some major design decisions around the new Explorer in Windows 8. It’s the new ribbon in Explorer’s default window view.

There’s a lot of good stuff in Sinofsky’s article. I truly appreciate and geek out a bit over the stats he starts with. Notably, the new ribbon includes the top 10 commands Explorer users use, which covers 81% of their use cases. Pretty nice accomplishment.

While it’s easy to look at this ribbon and make a judgment on it - MG Siegler does a bang-up job of it at TechCrunch - this is a static image and doesn’t account for actually using the thing. After all, it takes care of most of what people use Explorer for. From a pure usability perspective that’s good.

Who’s it for?

My biggest concern about this overhaul other than the questionable visual design is the statement from Sinofsky that the ribbon is not for power users. That’s stunning to me, particularly if I consider the visuals first and foremost. It seems to be overwhelming: commands, sometimes with buttons and groups and labels, and sometimes icons.

For power users, Sinofsky notes that Explorer can shrink the ribbon into - ready? - an old-school toolbar with tabs. They call it a minimized ribbon, but that’s semantics. Look at it. In this case, the File/Home/Share/View tabs look like menu items. Those are not menus!

Maybe the potential confusion over menu items v. tabs and toolbar icons v. ribbon icons relegates this to power user mode; ie, some people can deal with this inconsistency. In any event I’m curious about the usability of the power user mode versus everyone-else mode - toolbar versus ribbon. I guess Microsoft is too, since they’re including both in the product.

This dual interface approach appears to be an integral part of the Windows 8 strategy. 

Repositioning Explorer and Metro

Consider this. If one positions Explorer solely for power users then this giant ribbon of commands makes a lot more sense. Unfortunately, this isn’t the reality; at the outset of his article, Sinofsky calls Explorer “the most widely used desktop tool… in Windows.” People interact with files. People interact with interfaces which mimic Explorer to a degree, like the Control Panel. File management is a very-well established metaphor in desktop computing, but the web and mobile devices have forced changes.

Let’s zoom out a bit, and we’ll see the same toolbar v. ribbon opposition happening at a much riskier layer: the UI. As Siegler argues, Microsoft is hedging its bets with the Windows 8 UI by combining both the attractive Metro and the legacy Windows interface into one product. If not executed well, it could lead to some questionable and jarring interfaces. Like, say, a Metro Start Menu with a legacy taskbar.

You Gotta Keep ‘Em Separated

Let’s cut to the chase: the legacy Windows interface, including Explorer 8, is something which needs to be fully separated from the Metro interface. As soon as a user needs to jump into legacy Windows after working with brightly-colored, typographically-smart information panels - a very different metaphor - it’s all over. It’s akin to jumping down to the DOS prompt.

Users should be able to remain fully in Metro and never see a legacy Windows window. Think about the power of that: a new experience, shaped entirely by the new direction.

In order to make it into the future, Metro’s interactions need to be radically different than legacy Windows. Otherwise it’s just a skin, and skins usually don’t account for the way people interact with their computers and devices.

The Risk

By combining the two interfaces, Microsoft is trying to have it both ways: “Here is our way forward, but we’re not throwing anything out! In fact, we’re still developing the old stuff.” Should Metro fail, Microsoft can continue evolving the legacy Windows interface. But if Metro succeeds, Microsoft can continue to slowly deprecate the old interface. Unfortunately it needs to be slow because of the bazillions of apps which are designed for that old interface. Stuff that’s out of their control, stuff that still expects IE5 to be around, stuff that looks like Windows 95.

So from a strategic perspective, it isn’t inherently bad. It seems to mitigate the risk of having long-time Windows users drop off or rush to another platform/device because of it. It’s a conservative way to introduce change.

But the risk to the user experience is significant: with this dual interface world, Microsoft is nearly flaunting the concept of inconsistency as a feature.