We design and build software. Our software has user interfaces. These can be designed to work best for users who are novices or for users who are experts, rarely both at the same time. So for whom are we working? There is no single or simple answer.
As a matter of policy, we generally come down — at least at first — on the side of the expert user. You’re a new user for a while but after that you’re an expert for (we hope!) years to come.
What do expert users want that is different from what novices want? Experts want to do things quickly. They want interactions that are minimal rather than supportive. They know what the software can do and where to find the activators for those actions, whereas the novice wants more things exposed to make the choices visible.
The various computing platforms (in our case Apple) have their user interface standards from which it is risky to deviate. For a user who has learned the standard well, such a deviation might be frustrating. On the other hand, sometimes we really do have a better way. Apple’s — or anyone’s — standards can never address all situations. For example, if you need to choose a number, the familiar Apple method is the picker but, if the set of numbers is large, this can be slow. So we invented a “slider”. It was in some early releases of NurseMind but user feedback was negative (even though our own tests showed that it shaved a few seconds from certain tasks) so we went back to pickers. This was an example of a case in which novices won. Our policy is not inflexible.
The lesson is to test, test, test, and above all, listen to users. Their feedback is priceless. We work for them!