The bigger problem with the Windows 8 UI


Late yesterday my twitter stream lit up with news of a MSDN blog post revealing the new direction for the Windows 8 Explorer UI. It’s always great to see companies being open about their process and sharing pre-release product. Especially when they take the time to open a discussion around it and (hopefully) taken on any useful or pertinent feedback. Now, it’s become all too easy to berate companies for poor design and/or execution and particularly so when its everyone’s favourite corporate giant Microsoft too. But, beyond the inevitable slew of snarky comments about how cramped and ugly the new ribbon was, no-one appeared to have spotted the crux of the problem.

To me it boils down to three fundamentals that display an inherent problem in the Windows design team’s process.

Data is only half the story

It’s becoming common for design decisions to be made off the back of analytics data. Don’t get me wrong analytics are great - set up correctly they give us a great picture of what people are doing on a website or in software. That’s peachy. But what data on its own doesn’t provide is context. Or as i’m always banging on about - it tells us what people did, not what they didn’t do. That’s a bit of a mouthful but the basic premise is that data gives us the what - not the why. As a designer i’m more interested in the why. Why didn’t they use that button? Why did they give up on that purchase? Why are the majority of your users, using keyboard shortcuts or contextual menus over and above toolbars or similar? Did anyone stop to think why that is? There’s a pretty good chance that maybe they just like it that way.

What’s the problem you’re actually trying to solve?

Understanding the ‘why’ leads onto what problems actually need resolving. Even sometimes if there are any at all. The desire to develop, to evolve is so powerful in software companies, quite often the baby gets thrown out with the bath water. Not once in that MSDN blog post does the author mention what specific problems they’re trying to solve by introducing the new ribbon. The inference is that it was introduced because data said that people weren’t using the toolbar. Is that a real problem? or a self-made one? Maybe people are voting with their feet and it’s really not a desirable UI metaphor so why keep banging your head against that same problem?

Don’t ask. Watch

It’s another common UX myth that flat out asking people what they want provides a better user experience. That kind of insight can be useful (if you get a large enough sample) to gauge desirability of given features but not on it’s own enough to warrant designing a solution around. In general people don’t know what they want or how they want it, they just need that thing to do what they need it to do. The Simpson’s episode where Homer designs a car is the obvious reference there but I digress… The vocal minority in software design are passionate users, often power users if you must use that awful term. Their needs are based on an agenda quite specific to them. The article even references the validation for adding ‘power user’ features based on an minute fraction of users installing extra software to customise their file system browser. Why on earth would that warrant adding those features in for all users? People use Windows for a number of reasons, Work, Play, Leisure. Of course there will be a wide range of comprehension, ability and needs within that user base. Understanding those needs along with those levels of comprehension are what makes great software. Not once in that article did they mention observing users behaviour as they undertake common tasks across the many contexts in which they use the software. Sitting down, watching and talking to those people would probably reveal a very different product to the one that appears to have been frankensteined from a mixture of telemetry, noisy individuals and this odd reference to heritage that keeps popping up.

The whole design process and outcome smacks of a committee based approach which inevitably puts users second at the behest of an individual groups stamp on a product. My own twitter response pretty much sums up how I think it went down.

“The windows 8 file system UI looks like everyone round that particular table got their own way that day”