Johnson's book, Interface Culture, is about the growing culture of the interface, the way we interact with the world around us. It is based on the nearly invisible premise that we interface with much of the world, and have been for most of our time on this planet. I found this immediately intriguing because some f the hardest things to observe are the interfaces that we sue to connect and interact with the world. Johnson frames his discussion of interface with the elements of computer interface; the desktop, windows, links, text, and agents, all common to those people coming from a computer literate society.
Where Johnson really shines (and I admit a personal bias for the topic) is in his discussion about hypertext and the poor job that silicon valley has done in really pushing it to the limits of it possibility. He presents a picture of an industry that continues to try to bring television to the web (real video, real audio, flash) all attempts to bring movement and animation to a naturally solid state-dynamic environment. The real power of the web is in the link, in the ability of authors and users to "create their own story" - to navigate through the content as they wish, not necessarily how the author intended. Johnson uses Dicken's stories as examples of thinking that incorporates the sense of disparate ideas - all connected into one story - the kind of thinking that Johnson thinks needs to be used to harness the power of the link.
Johnson also takes time to explore the differences between "surfing the web" and "channel surfing", arguing that the two are fundamentally different. He argues that the passive, almost lazy activity of channel surfing actually works against our ability to conceive of the web differently. People who have this mentality will not be able to clearly see other possibilities for the web.
Johnson spends quite some time bitterly complaining about the lack of real innovation in hypertext environments, and in the end suggests that his own online magazine "FEED" is at the forefront of hypertext theory, pushing the limits of use. I was less impressed that I though I would be. Johnson is so very eloquent and keenly aware of the need to use hypertext as storytelling environment, to really push out lazy use of it, and to exploit the full potential of this tool. I feel that Johnson fails to acheive the goal that he so clearly lays out in his book. While FEED does use hypertext in new ways, it didn't strike me as particularly clean. By this I mean that the *interface* was clogged with too many links, the user while given many options was not given any clear or clean sense of direction. Burrowing into the site, the linking grew in scope and complexity, but instead of making my interaction more pleasant, I found I was more confused, and really had to try to find order. Perhaps this is just a natural reflexive response to the new use of a familiar thing, but I didn't to stay at FEED. I can see what FEED is trying to do, and I agree with the goal - to provide a dynamic interactive hypertext environment... but the interface was too hard to use. From a design perspective it is always easier to add a bunch of bells and whistles, the hard part is to take away everything that distracts from the message, that interferes with the usability. It seems like the producers of feed became excited about the possibilities of hyperlinking and no one ever stopped to ask when was a good time to stop. While all links are relevant to the content, the sheer volume of linking distracts the user - taking away from their ability to smoothly interact with the environment.