Applied to news, the Web doesn't suffer from one, but three flaws. Let's call these the Rectangle, the Bottleneck, and the Diversion. These flaws got built into the system from the very beginning and, now, their impact has become harder to deal with. For news sites, these unforeseen after-effects have grown to become real obstacles to reader engagement and monetization. Let's have a closer look.
The Rectangle issue
This is the main absurdity in Internet design. It goes back to the Web's early days: a text tool, brilliantly designed to organize scattered pieces of information - that was Tim Berners-Lee's 1989 original work. Understandably, he used the page as his metaphoric vector; that was the world he lived lived in. For news, since journalists controlled how the concept got implemented, the paper mentality, and even the vocabulary percolated down to the young generation of Web designers. They still refer to "the fold", pretty obvious for a newspaper (better have your story above the fold than below). On the Web, the fold points to the part of the screen below the first scroll. This brings us to the main contradiction.
I'm writing this on a 13 inches-wide MacBook. OK, I have attached to it a 24 inch Apple Cinema Display to allow multiple windows, but, for a simple measurement, let's stick to the realistic 13 inches. For a selected list of Websites, here are the numbers of scrolls required to go to the bottom of the home page:
Graphically, it looks like this.
There are many theories about the size of scrolls. Advocates of long pages pitch the following advantages: better referencing; more impressions for banners, more transverse navigation. Opponents retort: for referencing, you don't fool Google for too long by serving large chunks of your site over and over. For ads and for transversality, suffice it to say you lose readers real fast if they need to scroll down; the probability for a banner to be viewed decreases exponentially: 10 scrolls divide the probability by 100 or more. As discussed in an earlier Monday Note (see Measuring time spent on a Web page. ), a full 24% of all banners are not seen at all; a placement on the footer rates as low as 10% of what the header gets.
The page structure of Web news makes me think of the QWERTY keyboard: it sucks, but we are all used to it. Maybe it is time to move to the design equivalent of the Dvorak keyboard. It could look like this.
This is a capture of a New York Times' live prototype that present the news differently. Its home page is screen-based instead of being page-based. Less sexy perhaps but much more convenient. You browse faster, it adapts to the size of your window. You jump easily from one section to the other. This Times skimmer is a crude prototype; sections summaries still send the user to classical story pages. But it might be a path to a better browsing experience.
Screen-based structures will show a particularly good fit to tablet-like devices used in a landscape mode. Layers are loaded in the background and you "peel" them off as you read your e-newspapers. This would also work fine for off-line browsing.
There is a reason why Web designers rely on pages: the arrangement appears to multiply access to contents parts. At first glance, it makes sense but, in reality, it doesn't work. Fact is, leafing through a newspaper for three minutes is a great way for readers to find out what whets their appetite. Such leafing through is actually more efficient than squinting over a complicated home page which acts as a powerful choke point for content.
Such structure transforms most Web sites into a gigantic content sucking pit, burying URLs too fast way, shortening the lifespan of stories, portfolios, Web documentaries, especially if they are unfortunate enough to be disconnected from the frenetic news cycle. Solving this is not just a matter of page design. Rather, it entails embracing new modes of news consumption, with tools such as customizable pages, clever RSS readers and bookmarking features. Such new modes foster serendipity, "fortunate discovery," something not built into the original and, now, aging Web design structure.
Most news Web sites still make money from advertising. And that's the problem: the Web editor and the advertisers have exactly opposite goals. The former strives to capture and keep readers, thanks to interesting and carefully edited contents. The latter is equipped with a one-track mind: get the reader to click on the banner and, presto, transport her/him away from the site. The newspaper equivalent would be an ad space saying "leave this page and go right away to page 233", or "close this paper and turn on your TV."
In print, the ad is actually part of the design, not an incentive to go away.(And print designers take great care to use advertising space as a component of the grid page).
Do we see solutions? The best hope is a progressive switch away from banners designed to be clicked on, toward standalone image-centered spaces within the page, conveying efficient messages in Flash or video. Problem is, for the creative community, this is much more challenging than the facile crappy banners that infest the Internet today. And, still, for most part, advertisers keep wanting to transform a visit, i.e. get users to leave the media to land on their own site.
Another possibility involves micro sites, designed by the media to host the brand while keeping the user in the original environment. (For details, see Monetizing a social network, the Skyrock case.)
It is somewhat funny to see how, twenty years after the initial page-oriented coding of the Web, its structure, although massively adopted, is still in search of its own true nature, both from design and business perspectives.
By Jean-Louis Gassée and Frederic Filloux
Special to CBSNews.com