So I was thinking about user-experience design, specifically watching my partner squinting at the web-based paper-marking interface built into the British universities’ joint monster Turnitin plagiarist killer. And I’d said to Oliver Rivers that few things fascinate me more than human-computer interaction, but then again most things seem to fascinate me in the end.
OK, real quick, what does great UX feel like and why should anyone care? Well, I would say it should feel toolish, ha ha ha. Like good tools. Anyway, here’s a now-classic talk that speaks to the point:
Check out the daringly minimal use of his slides! So, why are we going on about this, and isn’t this just more Harrowell machismo and planespotting? Well, I saw this quote and it instantly reminded me of the Children of the Magenta Line:
maybe we could get the interface down to an iPhone app that would superimpose a bright white line over the camera’s view of the surrounding street just telling us where to walk and what to do and buy all day long. Wouldn’t that be a bit of a relief?
And, of course, it also made me wonder what I’d do #ifihadglass. Perhaps I’d make as big a monkey of myself as Scobleizer, boasting about how addicted to Google services I was and how I never wanted to take ’em off.
Anyway, the first major insight from Children of Magenta is that automatic can be more work. Depending on what you’re trying to do, solving your problem by punching more instructions into the autopilot can involve more work, more effort, and more stress than just grabbing the yoke and the throttles and pointing the aeroplane in the direction you want it to go.
More generally, a supposedly easier and friendlier UX can be more painful, less efficient, and more stupid to work with, especially for use cases that are more than trivial or when efficiency and speed are important.
The second major insight is that automation required a change of cognitive style that wasn’t helpful. They had to burn the limited resources of Kahneman’s System Two to get around the FMS in order to do things you could do with the intuitive, tireless skills of System One. As a result, they also tended to lose their strategic overview of the situation, the sort of thing System Two is good at.
So, people tended to do what seemed to make the automation happy, although it didn’t have any intelligence that made that meaningful and still doesn’t and probably won’t for a hell of a long time. This was the outcome of relying on it for command functions, when the original design concept was meant to let the machine steer and the pilot command, and of trying to steer through the management interface.
We have three antipatterns here; wondering what it’s doing now, not wondering at all, and responding to problems by making a PowerPoint presentation, roughly. I think this is relevant. I see people doing this with computers all the damn time. Technology is not liberating them to reason, it’s forcing them to guess badly at how it works, and it seems to encourage them to adopt pathological solutions.
What’s the alternative? The alternative is the aesthetic of tools. Think of those things you pick up and feel tempted to work with. A good kitchen knife, the long-handled pliers I used to carry around Australia in my belt being a really terrible apprentice, the Nokia 6230i, the Nokia E71, the iPhones, the ThinkPad keyboard. I could write a small book about that; IBM’s buckling-spring technology, the indisputable physical click, especially the speed with which you can tab through a web form or knock in commands in a terminal. The beauty of it is that doing so feels like not much more work than this post.
OK, enough aestheticising. Henri Cartier-Bresson said the best camera is the one you’ve got with you. This has been taken as a rebuke to materialism. If you lack the Artistic Soul, young man, no equipment will help…you know the sort of thing. But the camera you’ll have with you is a potential design objective. Dimensions: easy to carry. Material and structure: as tough as physically possible. Features: general purpose capability above all. Today we’d probably say that as much of it as possible should be in software, as nothing is easier to carry around than software.
What are the tool-like qualities? What you can do with it – the affordances – should be obvious or at least learnable. The tactile experience should be satisfying. It should be easier to use it right than wrong. On the other hand, it should reward discovery and experiential learning. These two points are contradictory, like most things that are important, and the challenge is to cope with them. In many ways, things that are good to work with lead us from system 2 to system 1. They are also notoriously beautiful.