On geometric algebra: Let $I, X, Y, Z$ be the Pauli matrices with identity. Let us consider them as spanning the space of Hermitian 2 by 2 matrices. The multiplication rule for such matrices is: \begin{eqnarray} (aI + \vec b \cdot \vec \sigma)(cI + \vec d \cdot \vec \sigma) = (ac+\vec b \cdot \vec d)I + (c\vec b + a\vec d + i \vec b \times \vec d)\cdot \vec \sigma \end{eqnarray}
This looks a lot like an equation of geometric algebra. Is there a connection?
On flow version self-interruption: When learning to behave in new ways we constantly interrupt ourselves. "That's not the right way to do it" we think. And we think about how to do it in another way.
By contrast, when in a state of flow we are not trying to get better. Rather, we are simply operating.
There is a tension between these two ways of doing things. My guess is that to get really good at something we need to spend a fair amount of time in both states. Per Timothy Gallwey, I rather suspect that we tend to spend too much time in the self-interruption mode, and not enough time just operating.
What would an out-of-control explanation look like? Usually when we think of explanations we think of something that has been designed: a documentary; a book; a verbal monologue. But much of the trend around networked media has been for things to get less under our control. What would an out-of-control explanation look like? Does such a thing even make sense?
A partial answer is that ordinary conversation has something of this flavour. A good explainer responds to the person they're explaining too; it becomes like a dance. Exceptionally good explainers do this well even in a group. And great explainers really are, sometimes, out of control: they do not know what will happen.
Can we do this technically? I think we perhaps can. The point is to set up circumstances and processes where this is likely.