On the possibility of using web-navigational schemes to communicate data rather than site-structure…

01/12/2003

I’m two-thirds of the way through my second Edward Tufte book, The Visual Display of Quantitative Information. Now, before your eyes glaze over and you start skipping over to one of those proper weblogs where they talk about sex and disastrous relationships and the movies they’ve been to see recently, I want to try and convince you that the books of Edward Tufte are fun and interesting. I have no obvious reason to read them – I’m not a statistician and I don’t work with graphs – and yet I find them endlessly pleasurable. I suppose there are several reasons for this , although some of them might be unexpected. Firstly, they are books which are intellectually stretching without being wordy or incomprehensible. They immediately open up realms or spheres of engagement with the world and with the information in it that are normally hidden from people. Secondly they are profoundly sensual experiences. Printed on high-quality paper, mixing textures between covers and cloth and sheet – these are quality publications. The typography is beautiful, the diagrams are never less than beautifully rendered – and occasionally they are simple beautiful diagrams, charts and the like. And finally they have the charm of the Schott’s Original Miscellany (in that they contain as asides blocks of utterly unexpected information and background) without the handicap of being about nothing… In fact all of Tufte’s books are satisfying because they exude the care of the artisan or the craftsman – someone with a profound expertise in, and love/respect for the process of creation of quality. In this respect his books really remind me of Robert Bringhurst’s Elements of Typographic Style, another book that beautifully expresses passion and expertise based around a respect for quality.

Which is probably why, when I was reading The Visual Display of Quantitative Information, I was so surprised to come upon a paragraph that was so completely alien to the practices of web clarity and usability. In it, Tufte states some beliefs that have long-since been rejected from web design – ostensibly because they are smack of elitism and are impractical – even costly. He states, in effect, that we shouldn’t forsake complexity just to communicate with our audience. Here is a brief quote from a much larger section of the book concerned with novelty and different paradigms of information-presentation:

Moreover, it is a frequent mistake in thinking about statistical graphics to underestimate the audience. Instead, why not assume that if you understand it, most other readers will, too? Graphics should be as intelligent and sophisticated as the accompanying text.

The section concerned is a lot longer and more detailed than the quote above, and is fundamentally about revealing complex and large amounts of information through graphical means – rather than about navigation or layout. But it’s still interesting to me. He talks about how most diagrams used in the national press (if compared with those used at schools and colleges) are of a pre-adult level, and that we shouldn’t allow this to happen. We should respect the intelligence of our audience. The comparison with the title of Steve Krug’s brilliant web usability book Don’t make me think! (much recommended) couldn’t be more striking.

So what is it about web navigation and site structure that means we should treat our users as idiots while infographics are allowed to treat their viewers like grown-ups? Unfortunately the answer is ultimately extremely simple – infographics are designed to transmit useful information and that information is the final goal of any interaction with them. This isn’t true of web-navigation. Web navigation is designed to structure and communicate information about how to find different information. Web-navigation (and my doctoral supervisor in Bristol is literally turning over in her tenure as I say this) is for the most part nothing but meta-information – no one (except web UI professionals) visits MSN.co.uk for the information communicated by the navigational scheme. It’s fundamentally uninteresting.

But this does not necessarily have to be the case – it seems to me eminently possible that we might be able to generate a hybridised system whereby web navigation actually itself holds valuable information – much like Tufte demonstrates the axes of a graph can be redeveloped to carry additional information about the data that they frame (I’ll see if I can find a visual representation of this as soon as I get a moment). I have at this time no sense of how one might go about doing this (except through a cascading series of inter-navigating graphs) but there seems to me to be a certain amount of potential in this line of investigation… More on this later (perhaps)…

Addenda: How to pronounce Tufte (after kottke.org).