Donald Clark observes
Knowledge is not held in our minds alphabetically or in a linear or hierarchical menu structure. Knowledge is held in different ways procedural, episodic semantic, and called up into working memory, but it is fundamentally a neural network, physically and representationally. A hyperlinked representation of knowledge is therefore a much more useful learning tool as it reflects this structure and allows us to learn new knowledge structures that fit into our existing pre-requisite networks. These networks are personal and hyperlinked networks allow us to move through knowledge in a way that fits our existing structures, expectations and intentions. The brain is hyperlinked and so knowledge needs to be for efficient learning.
Having laboured to produce a book with Viv I am very aware of how frustratingly linear the format is.
The first time we used that book, in an early draft, we stood in front of a group and ripped the spine off. Then we gave out different pages to different people. We invited them to read their one page and then find someone else to discuss what they thought of it. We got them to repeat that with several different people.
I liked that activity for all sorts of reasons. Not least because it lowered the status of our content and made it more of a jumping off point for conversation. That was the last time we really used the book in a week long workshop, and rightly so.