Peter Cochrane's Uncommon Sense: Screen tests

"Computer, show me that spreadsheet..."

Something as simple as the size of a user interface governs much of what we can do with computers. It's led Peter Cochrane to envy the 'tell, find and show' machines used in Star Trek... At the age of 56 I have to say the mobile phone and I are not the best of friends. First, I have very large fingers and as phones have gotten smaller it has become increasingly likely I hit more than one key at a time. Second, as I get older, I find it increasingly difficult to read anything on a very small screen without changing my glasses, especially in poor lighting conditions. I purposely purchased a laptop computer with a 15-inch screen and large keyboard. And I habitually type using 18 point so I can read with ease in any environment. One of my key discoveries in terms of operating environments has been that the use of a laptop computer in physically unstable conditions - automobiles, trains and sometimes aircraft - can be overcome by selecting all, triple spacing and increasing the font to 22 point. Large letters, well-spaced words and characters, with lots of white blank spaces, make it much easier to work and obviates any symptom of motion sickness, which rapidly onsets when working with single spacing and 10 point. My thinking on screens has been that there will come a point where we move away from the very small to the very large. We will be driven by application and the Captain Kirk Condition - the need to address any machine, any screen at will, and in any location. In Star Trek, no one really uses or wears small screens. They mostly approach any screen, on any deck of any ship, and boldly ask a computer to tell, find and show. I think a similar regime is about to be engineered for you and me. Most mobile phones have a screen measuring a few square centimetres where a limited amount of text and very small pictures can be accommodated. While this may be adequate for making and receiving telephone calls, receiving text messages and playing very simple computer games - or perhaps even watching some moving images - it's wholly inadequate for looking at spreadsheets, timetables and most other complex graphical forms. Next up the scale is the PDA with a screen area that is roughly 10 times bigger than a mobile phone's but still less accessible than the laptop computer. Each of these screen formats has yet to find its true place in our IT future. We now have screens spanning centimetres to metres for mobile and fixed devices but what we don't have are the necessary interfaces to make the accessing of information intuitively simple. In my ideal world a very large plasma screen located in any shopping mall, railway station or airport should have a telephone number across its bottom. By using my mobile phone as an entry device, I would be able to dial into that screen and talk to the computer using natural spoken language - and keys. I want to be able to call up any form of information or interaction that is inaccessible on the phone itself. Such a regime has several advantages that those in the business will immediately recognise. For companies installing kiosks in public places it is frustrating to see the occupancy largely dominated by non-customers, people who don't really use them for the purpose they're intended. This includes those who just want to play and those intent on vandalism. Having to use your mobile phone to pay for access would switch off idle curiosity and vandals would have to be incredibly stupid to smash up their own handsets! It would be also be almost impossible to damage a plasma/LCD screen behind bullet-proof glass. The benefits could be huge. Those carrying mobile phones would have instantaneous access to all information on a screen that could be easily read and would be far more powerful than a mobile or PDA alone. Suddenly the mobile phone would become the passport to the information world. The biggest hurdle is the voice interface. Put bluntly, there has been little real progress in this arena over the last decade. Silicon speech recognisers have been more efficient than humans on individual words for a long time. The problem is not speech recognition per se but the contextualisation and cognition capacities of machines. Natural conversation between man and machine as illustrated in the movie 2001 is still a long way off. We have speech synthesis and recognition that is adequate for command and control but wholly inadequate for a conversational enquiry as demonstrated by Captain Kirk. And the reason for this is our inability to fully understand intelligence and cognitive systems. This seems to be the next big challenge. This column was dictated onto an Olympus recorder, typed on an Apple iBook by my secretary, emailed to me via a cable modem connection, edited by me in the back of my car while waiting for my son to complete a music lesson, and despatched to my editor via my Motorola280TimePort mobile phone at 9.6Kbits.