My recent adventures in following the bibliographic paper trail have led me deep into the forest of interaction and design theory.  I am currently trying to make some connections.  The old "is it art or is it design?" question continues to come up.  These categories are quite inadequate and will never fully describe a field or a made object.  But somewhere at the root of the desire to categorize an object or field is the desire to categorize the maker's intent.  Anyway, much of the design theory I have been reading (M. McCullough, Digital Ground) is quite concerned with interaction.  As an self-classified artist pursuing a degree and Time and Interactivity, theories of interaction are quite interesting to me.  

In Digital Ground McCullough says "Embodiment is a property of interactions; latent embodied abilities exist [in the user]; and good interactive technology lets us exercise these abilities."  This, of course, seems to be a legitimate desire — design interactive technologies that are intuitive and don't require the user (participant/subject/spectator/interactivant/actor) to learn a new skill in order to access the art project.

But I am left wondering how much of these interaction theories should be "designed" into my art.  I suppose it is content driven.  I suppose there will be situations when the interaction should be intentionally difficult in order to prove a point.  In any case should have a clear sense of what role they are supposed to play, not matter how complicated or unnatural the interactive gesture is.

This seems to be a perpetual problem.  How much instruction do you give a participant and how much should be left for them to discover?  Will a project fail if the participant fails to discover the richness of an artists obscure interactive technology?

The answer to questions seeking an "always" seems to always be "sometimes".