Research, Design and Windows Explorer

by on September 14, 2011

Quantitative research and design make uneasy bedfellows at the best of times, but a recent Microsoft blog post shows just how uneasy this relationship can become. Trying to do design for a massive corporation in which design comes a distant third behind the business model and engineering is plainly maddening.

Note first they need to defend themselves against the engineers: “…telemetry data indicates these add-ons and alternatives are mostly used by us power-users and we represent a small but influential group of people” and other weird statements that really would not need to be said by anyone other than a Microsoft designer.

They then show a chart that indicates from “hundreds of millions of data points” that 10 commands are used over 80% of the time, despite there being over 200 commands in the application. Moreover, that over 50% of them are invoked using a right-click context menu (wow – that’s a lot!). Menu bar use on the other hand, is tiny – it looks like maybe 2%.

So what is their response to this? They say they should make commonly-used functions easier to get to. Their prescription for is the Ribbon UI, which we have to just assume is better than a context menu in this case. At least, they offer nothing much about this beyond the almost throw-away “…to make it more effective—more visible and uniformly accessible”. That’s fair enough. But then they go bonkers and proceed also to put a bunch of commands into that ribbon that their own research tells them aren’t being used!

Given the size of the sample, it would seem to me to make a little more sense to work out why the lesser-used commands are used less and – oh, I dunno – not shove them in people’s faces? Or do Microsoft think that people just want to make more new folders, or do more “editing” or making “new items” (whatever that command does!) and if they do it will somehow make for a better experience?

This is perhaps why the Ribbon appears to me to be such a pain to use. The UI’s weird heterogeneous visual design (some things are icons, some are not, some require different interactions to use, etc.) makes me have to rely far too much on visual processing to sift through commands I’m never interested in just to find stuff I want. Stop designing and start thinking is what I say, but I am clearly not the sort of person who would do well at Microsoft.

Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>