Joe Clark: Accessibility | Design | Writing

Deep Sites:
Intelligent Innovation in Contemporary Web Design

What is the purpose of writing dead-tree books about Web design? What is the point of using an old medium to discuss a new one? Isn’t the new medium perfectly able to discuss itself? In a recursive twist, Weblogging librarian Jenny Levine ( explains that, through books, “the concept of blogs will officially be catalogued and indexed in our collective memory (not just the memory of those of us who live online). And as a librarian, I feel it’s important to get some historical perspective and preservation of the early years” (emphasis added).

We can expand Levine’s analysis to Web sites in general: We write books about the Web because the Web is evanescent, while books are less so. Max Bruinsma’s Deep Sites inadvertently fulfills this archival function, while adding an overtone of curator or editor. That’s typical of survey or overview books on the topic of the Web: They talk about sites the author likes, and such is Bruinsma’s strength and weakness. His criticisms originate in a philosophy that’s tried-and-true but actually outdated.

There is, of course, the issue of rendering an ever-mutating medium, often equipped with multimedia and animation, on the printed page, but what’s little-discussed is how nice Web pages tend to look in the compacted screenshots found in computer books. Bruinsma and his designer Annelys de Vet pack an enormous mosaic of screen captures onto each page, and they work well as pure documentation. But an oddball anachronism of a book concerning the Web is the use of catchwords – the first words of the next page printed at the bottom of the current one. Rather Victorian, you might say. (Catch the catchword flub on page 61.) Franklin Gothic, however, is at root a letterpress face, and looks brilliant and overprecise even in the ITC variants printed on the book’s coated stock.

Since designers all use the Web and tend to evaluate sites through the graphic-arts prism, designers tend to be ignorant of trends in Web development that are unrelated to appearance. Deep Sites’ subtitle, Intelligent Innovation in Contemporary Web Design, makes Bruinsma’s deficiency in this respect all the more ironic. By far the most intelligent innovation in contemporary Web design is standards compliance – designing sites in accordance with published standards for HTML and accessibility for people with disabilities. (That means valid HTML and adherence to the Web Content Accessibility Guidelines, among other things. See

By dividing Deep Sites into topic areas that pertain mostly to looks and behaviour in his own browsers (including the notoriously-noncompliant Netscape 4), Bruinsma recapitulates the typical error of print designers approaching the Web: That the guts under the hood, all the HTML and other coding, serve no goal other than producing a site that looks good and works for them. That isn’t intelligent, innovative, or contemporary at all.

Bruinsma himself nails the argument for standards compliance in his foreword: “For designers, few things are as annoying as seeing a carefully-orchestrated arrangement of texts and images fall apart on every other computer screen than their own.... the interface and its visual representation do not hover generically over a site’s subject-matter; rather, they are part and parcel of it.” I just wish the book gave us proof. It would be an interesting exercise to republish this book with the addition of an assessment of each site’s standards compliance.

Still, even if one resignedly accepts that Bruinsma’s conception of the Web is sort of ancien régime nowadays, in all fairness he has a good solid eye for his topic areas – Interface, Typography, Animation, Community, and the only theme unrelated to æsthetics, Authoring. But Bruinsma’s widely-varying tone and concision are frustrating. This former editor of Eye can spot the genuine innovation in a site, as with, and sum it up nicely: “[Erik] Natzke experiments a great deal with Flash to make transitions from one page to another less awkward than the usual ‘hard cuts’ or empty screens that say ‘please wait.’ ” But he’ll also noodle along with descriptions so plain they belong in a courtroom deposition, as with “In a parody of the Web’s most popular content... the soundtrack behind this page is a metallic female computer voice, frigidly reciting the most obscene proposals.”

At what point do sites bearing little or no innovative graphic design become interesting? Peter Bilak’s Web site,, is supposedly of interest because “any hint of frivolity... is relegated to the contents... [with] a strict typographic austerity, both in the choice of typeface (Courier) and of coding (simple HTML).” Cornel Windlin and Stephan Müller’s “rigidly poses as being ‘undesigned’ ”; who says that’s a pose, and how is it intelligent, innovative, or contemporary? (Does Bruinsma like their sites because he likes them and their other work?) How does undesign square with Bruinsma’s praise of, whose tricky speed-sensitive scrolling method “should be followed more widely”?

Bruinsma can be wowed by small things: At, visitors can hear a site owner talk about his creation “and at the same time load the site’s URL into this application – media at their best.” Nope – it’s a simple use of frames, actually, not that anybody uses frames much these days. On, “navigation devices... are elegantly revealed on mouseover”: Well, what else do you expect to happen on mouseover?

In survey books on Web design, graphic knowledge isn’t enough. You need good solid technical qualifications to address – and record for posterity, Shifted Librarian–style – “the interface and its visual representation.” What designers, and critics, seem to need is greater training in Web structure and development. Give us some science with the art.

Updated 2006.09.10

You were here: joeclark.orgDesignDeep Sites review