To me, interaction is indeed a feedback loop between two or more people/things that involves them taking turns sensing, thinking, and responding to each other (and, alternatively as Chris Crawford says: inputting, processing, and outputting).
However, interactions come in all shapes and sizes, and unlike Crawford, I feel that the opening of a refrigeration door fully counts as an interaction, just not a very ‘rich’ one.
Some interactions are more ‘rich’ than others, and it seems to be based on the quality with which something performs the sensing, thinking, and responding phase in the cycle. But I don’t think scoring low in one of those categories should disqualify the interaction.
If we evaluated interactiveness on such a basis, I would hesitate to call anything with negative feedback (output) a good interaction; yet a lot of our most basic senses are designed to detect the ‘bad’ just as well as the ‘good’. Some of them even save your life!
Everyone draws a different richness (or lack thereof) from certain interactions, and not everyone is sensitive to the same things. A not-so-subtle note for all you cat people:
So its tough for me to say that any interaction can be objectively ‘good’. But for me, good physical interaction is definitely based on a synchronization of visual, audial, and tactile feedback. #teamTactile
It’s in our primal instincts to touch, so I strongly agree with Bret Victor that our hands are extremely important for navigating our interactions. There is definitely a dearth of this type of interaction in the screen based word we live in. This is one of the reasons that we have fidget spinners and god-damn fidget cubes.
I applaud efforts in the haptic feedback realm that close this gap. For example, force touch in general, but also the recent addition of haptic feedback to the iPhone in places such as the timer/clock function. This was a small change that helped the device enhance the feedback (output) it gave, making for nicer experience, and perhaps a more efficient one as well (less likely to over-scroll).
With that being said, interaction isn’t everything, and there are lots of digital technologies that should not be made interactive. Abstracting from the frame of one interaction, it is hard to say where one interaction stops, and another begins. I think there are a lot of great technologies that can be built to remove interactivity and better achieve their purpose. Some examples include automated lights, payment systems, data processing systems, and other back end operations. Sometimes it’s better to be minimally stimulated or not bothered at all. To me, this is still interaction because two things take turns (or one turn, or one half turn) interacting with each other for a certain purpose; but it is not enriched by heightened output or extended input.
In conclusion, I hate a lot of kickstarter ideas. They often miss the point of interaction by complicating it or “smart”enizing something either in the wrong way, or that might not need to be improved at all. Some examples. Down with Kickstarter. Up with purpose driven interaction models!