Interaction packed

The launch of digital TV towards the end of this year means the small screen is soon to go interactive. But how will it work and what will it look like? “A lot of interactive precedent isn’t televisual,” says Michael Crossman, managing director at Bates Interactive.

However, interactive TV won’t be like computers either – viewers, or navigators, will be 2.5m from a bad-resolution screen. They will also want to munch dinner while they interact: mouse and keyboard have been rejected as a messy interface solution. Designers face the intriguing task of producing a new interactive system that is both coherent and enjoyable to use in the front rooms of 14 million UK homes.

Functionality has to tackle three core issues: navigation, legibility and, indivisible from the success of the first two, saleability – arresting and holding on to an audience. The digital TV age will see competition between 200 or more channels. Facing up to this ruthless sink-or-swim culture, it is little surprise that silence surrounds companies evolving new interfaces, such as Teletext, British Interactive Broadcasting (BIB) and Cable and Wireless. Getting function right will mean staying in business.

At BBC Interactive, designer Andrew Kearney says that, although his team is working on a navigation paradigm for the immediate future: “Computer systems show that we’ll probably be using a similar interface in ten years’ time.” The race is on to design an interface which will ultimately be the industry standard. “It has to be relevant to all ages and needs to accept add-on levels,” he says.

Interactive TV will require minimal hardware: set-top digital-analogue decoders (about 200) and digital TVs will come with a remote control – the only point of physical contact between navigator and screen. With up, down, left and right cursors, red, yellow, green, blue and select buttons, this is the mouse of the future, says Deep End designer Andrew Donald. Navigating menus and information will be via horizontal or vertical hops and colour-coded options. BBC Interactive producer Tom Williams says the objective is to keep the system “simple, but not simplistic”. To deliver a comprehensible, user-friendly interface can be pretty complex, though.

In any windows-style screen environment viewers tend to become disoriented when they are taken deeper than three levels. An easy route “home” to the top of the hierarchy is crucial. BBC Interactive is looking at ways to show several layers of information simultaneously by dividing the screen into quadrants. The system provides textual and still picture information, menu options and a clear indication of where the navigator has come from at the same time as playing real-time programmes. While making good sense of the screen spatially, the system is challenged by text: the large quantities required by a news service, for instance, can result in frenetic scrolling – the wrong sort of interaction.

Turning underlying programming structure into easily apprehensible design imposes some basic rules. Crossman, whose company is working with Cable and Wireless, believes users shouldn’t have to backtrack: “It’s fine to ask people to explore, but you should never feel you’re stepping back. It implies you have made the wrong choice, that you weren’t clear about where you were going.” Ideas about navigation are emerging that circumvent retreat and make sense of the matrix-like navigation process. BBC Interactive employs a virtual carousel that is rotated by clicks of the left and right cursor buttons to reveal menu options.

The carousel taps into users’ memory banks: they have an intuitive grasp of how it works. At BBC Interactive, Deep End, Bates Interactive, Amaze and Icon Media Labs there is considerable interest in subliminal triggers that help make the interface transparent. Interactive TV needs to continue delivering the easy access that has underpinned the success of analogue telly. It can draw, to some extent on the protocol – given modes of operation – that has started to emerge in multimedia and Net design. But, by drawing more widely on collective experience of the real world, it is hoped that navigation itself can be made “invisible”. “People aren’t interested in technology, but what it can do,” says Deep End partner Simon Waterfall.

In the US the over-50s are the fastest growing group of Internet-users. They have time and money, but are coming at interactivity from a print, not an electronic culture. They are also more likely, physically, to have difficulty reading the screen. Dr John Gill, head of a small research department at the Royal National Institute for the Blind, believes good design for the partially-sighted is ultimately good design for everybody. Interactive TV will be far more text-heavy than television to date. Gill would like to see designers thinking about typeface, size, colour, contrast and users’ reaction times. His department is working on a true-type face which will be provided to manufacturers of set-top decoders as a default font.

With long risers and descenders, the RNIB font is being designed “using visual, not aesthetic criteria” to make individual character recognition as easy as possible. The brief is exacting: it must be read at 140 words per minute, on a screen 36 characters wide. No more than four lines of text can be shown at once, and they must occupy less than 30 per cent of the screen. Gill would like to see set-top decoders equipped with a function that enables individuals to size text according to their own vision requirements. In-built memory would enable each member of a family to select an optimum relationship between on-screen text and image. Not only would this be empowerment for the viewer; it has the potential to liberate the designer from a very exacting, perhaps constraining discipline of on-screen design.

Dealing with a different aspect of on-screen legibility, psychologist Mary Dyson at Reading University’s typography department has carried out research with Microsoft. She says: “We asked people what they thought they could read best: they were daunted by long lines.” However, testing showed lines of 70 characters gave optimum readability. Comprehension tests also showed people get as much from long lines as short, “even if they don’t like it”. Even if an ugly screen yields content better than a visually pleasing one, it is not in a broadcaster’s interest to use it if it doesn’t help win ratings.

It seems probable that interactive TV will diversify and target niche audiences. If this is the case, in due course designers will be given freedom to tailor design to particular, perhaps even esoteric tastes.

In the meantime, there are relationships between interaction, cognition and learning currently used in wackier website and CD-ROMs that may be useful far more widely. Crossman believes of interaction: “action should have reward. It’s about sending people messages.” Antirom designer Nicholas Roope says inter-action is “like a filing cabinet with sensual qualities”. That sensuality is provided by play, surprise and suspense. “Make navigation transparent but the content challenging,” says Roope.

Crossman thinks that initially people will forgive rudimentary design: as Teletext has shown, service is far more important than Lego-block type. Get the functionality right, and form will follow. In the meantime, there is a lot of experimenting to do, and a lot of testing that can only be done on a suck-it-and-see basis.