People Suck at Computers.

Most people are afraid of computers. Even something as gloriously inviting as the Microsoft Surface, which I worked on during my internship at Microsoft Research, has people interacting with it cautiously. At least this is what I saw in my user studies. My participants were mostly professional adults (read: the people who actually make the decisions to buy things).

Much has been said about “the kids” being different than the adults. Watching younger people use anything new, you can see that they are much more destructive and risk-taking than adults are. This type of behaviour is, undoubtedly, the best way to learn and explore. For the sake of argument, let’s call this behaviour improvisational behaviour.

Computers have evolved to afford this type of behaviour, good examples of which are Undo and the Recycle Bin. Computers that allow improvisational behaviour are good computers. But most people do not behave this way normally, and you could even argue that it is slightly “unnatural”. One could point a finger at “society”, or simply point it at age, or even say that some people are inexplicably “better” at things. But these are all defeatist attitudes and don’t address the problem.

I am personally very biased towards this “improv” model, but I really do believe that people who have had improv experience are faster learners of new things, and feel better in unfamiliar situations. And the faster you can learn and be comfortable, the faster you can start doing Things That Matter. A complaint I frequently hear about improv from non-improvisors is “I could never do that” or even the notion that some people are good at it and some people are bad, and that’s it. High improv skill is treated like skill in any creative endeavour – it’s a magic thing that cannot be understood, much like falling apples and the movement of planets used to be. That is wrong. While Science doesn’t have a complete model for creativity, improv trainers and practitioners have a large vocabulary of terminology to refer to when describing their thought process, and what actually happened. Becoming a good improvisor requires a lot of work, but the point I’m trying to make is that, no matter what, it can be done.

So what really interests me is not just interaction experiences that afford this type of exploratory, improvisational behaviour, but those that somehow turn you into a more exploratory, improvisational person. It’s not just the computer that gets intentionally changed by you, but it intentionally changes you back! (like in Soviet Russia).

Computers-becoming-intelligent-and-thus-evil nay-saying aside, this is actually very important for the future of the Computational Experience. Anyone who has been to a good acting class should remember moments of extreme discomfort and self-consciousness. These are the awkward moments where you begin to truly understand yourself and what you are doing and can change. It’s social pressure that stops you from running terrified from the room. This applies to any activity where you are putting yourself out there and are responsible for it. You can’t accomplish something real unless you’re getting a frequent reality check.

This social pressure is pretty important. We’re terrified of other people much more than we are of anything else. I am afraid that it’s hard for a computer to put much pressure on the user, because they can just stop or walk away. We still think of computers as tools, rather than as a sort of being that can actual contribute back and you can feel relaxed around (no computer today is good enough for me to feel relaxed around it). In order for computers to really be useful, we need to figure out how they can change us back to make us better too.

And until we figure out how to do it, all people will suck at computers.


2 responses to “People Suck at Computers.”

  1. I suspect the truth is simpler than you suggest. I suspect the fear of breaking things, rather than the fear of improv (i.e. of looking silly before others) is more behind the tentative approach people take to new things. But there’s an easy test: leave a person alone with the object. If they’re inhibited by fear of looking bad in front of others, being alone should remove those inhibitions. If they’re inhibited by fear of causing damage, they’ll still be tentative.

  2. Yeah John, I think you have something there. Maybe it’s just that kids are not financially responsible when something breaks, so unless you tell them to ‘stop’ they’ll just go nuts. There are definitely different pressures for different people. In lots of usability labs there are giant one-ways mirrors, and now that I know people are watching I wouldn’t be able to be in one of those again without feeling self-conscious.

    Part of what I was trying to get at (I realize now) was that there is a certain optimal “outlook” to have for a certain tool. I certainly don’t want to say it’s a good idea to behave “improvisationally” at the controls of a nuclear power plant. However, the nuclear power plant engineer who spends all day maintaining safety at the plant can’t go home and interface in the same style with his/her family. Different situations require different outlooks. This is the difference between professional engineering software such as AutoCad, and .

    Computers should be able to be aware of these attitudes and steer them towards the “right” one.