91Èȱ¬

The 91Èȱ¬ R&D Prototyping team has been investigating how multi-touch software could support television viewing in the future. This article, written by Dominic Tinley from R&D Prototyping describes the two prototypes we built and how we decided on the features for each of them.

Creating an easy way for viewers to decide what to watch is a problem which designers have tried tackling in many different ways since the dawn of radio and television. In our first prototype we wanted to explore how a multi-touch device could enable multiple users to collaborate on creating an evening's television viewing schedule in a simple and intuitive way.

The decision processes that even a single viewer undertakes when choosing what to watch are incredibly complex, and yet they are processes that millions of people take every day without giving them a second thought.

Deciding what to watch can be as simple as pressing the number for your favourite channel on your remote control, liking what you see, and watching it. The challenge was therefore to create something as simple and intuitive as pressing a single button but that would add a complexity and richness for viewers who wanted to broaden their interests and try something different.

Our first iterations imagined that each user would bring to the table a 'bucket' of programmes they wanted to watch that would be populated automatically based on personal preferences and past viewing habits. Users would then have to make conscious choices of which programmes to drag on to a timeline, and the software would assist in the negotiations between multiple people.

Initial testing of this idea failed our first criteria, ease of use. While it would be helpful when all viewers were in an active planning mode it wouldn't work in the more common situation of opportunistic viewing where people switch on the television to see what's on.

So we introduced an algorithm that takes the contents of each user's 'bucket' of programmes and prioritises these to slot them in to the available viewing time. A user can tweak these preferences at a macro or micro level with a single gesture. At a macro level a user can give more or less influence to a particular source of recommendations, and at a micro level a user can increase or decrease the weighting given to a particular show.

The result is an interface that presents multiple users with a television schedule that should meet their combined needs with almost no intervention but where a series of small tweaks will optimise this for their respective and combined moods on that occasion.

Having created a system for users to choose what to watch we turned our attention to how people might interact with a programmes using multi-touch. We considered the growing trend of viewers chatting along with television using social networking tools such as Twitter but rejected these on the basis of ergonomics. In short it's uncomfortable and impractical to constantly change your gaze between a vertical television screen and horizontal multi-touch screen and far more practical to chat along using a laptop or mobile phone.

The multi-touch table becomes far more effective for large gestural activities that don't involve constant change such as infrequent votes dotted throughout a programme. Using the 91Èȱ¬'s Strictly Social web tools and Apprentice predictor game as inspiration we developed a model for multiple users to play along with a programme and compare scores at the end.

In the next blog post we'll be looking at hardware and software behind the prototypes we've developed.

Links

Strictly Social Ìý /strictlycomedancing/play/strictly_social/about.shtml

Apprentice Predictor Ìý /apprentice/about/predictor.shtml