If someone is trying to achieve a task and they are not provided with sufficient help or training to complete that task, does this increase or reduce their performance? Many would argue that not being handed everything on a silver platter is a great way to induce someone to progress, to find their own way, teach themselves and come out stronger. Others would say that it is unfair to expect someone to carry out a task when they do not know how to do it, or haven’t had sufficient training to do it well.

I would argue that whether it is good to challenge the person (by expecting them to do something they are not trained to do) or to help the person in completing the task (by providing training of some description), depends on two things:

  1. Will they be able to complete the task without help? (will they be able to train themselves, or seek help from some other source)
  2. Will they find the challenge challenging? (will they be spurred on to complete the task or just walk away)

Now, don’t worry, this isn’t going to devolve into a debate about the capitalist school of hard-knocks vs the soft-touch school of socialism. What I’m talking about here is interaction design in computer systems.

Sometimes it’s great to be challenged by a computer system. Games are a fantastic example … it wouldn’t be half as fun if the puzzles were all simple, the moves all explained and the enemies all ludicrously easy to kill. But challenging users in the wrong places, or the wrong ways, can lead to them walking away from your system … and possibly not even completing the task.

Often when designing human computer interaction for computer systems, we lose sight of there being both a place and a time to challenge the user. Reading the next unread email should NOT be a massive celebral challenge. Working out where the button to create a new document is shouldn’t either. When the task at hand is not the system itself, we should aim to make the system fade into the background. Using the system should not become the task itself.

So why are games different? Well, they’re not in all ways. Figuring out how to slam that opponent into a wall should only be challenging in terms of the game AI, not the controls. A recent OK/Cancel alighted on this particular issue, when comparing some seasoned gamers to their cartoon avatars. The skill should not just be in using the controls, but interpreting and predicting the game itself.

Why is it even necessary to talk about this? Go to the Interface Hall of Shame for some great examples. Too often we are producing great systems which challenge users in the wrong places, in the wrong ways. When they have an alternative to the system even if that is abandoning the task or *shudder* performing it hardcopy, the loser is the system … and so the developers