Computer scientists from around the world will gather in Boston this week at Computer-Human Interaction 2009 to discuss the latest developments in computer interfaces. To coincide with the event, we present a roundup of the coolest computer interfaces past, present, and future.
The Command Line
The granddaddy of all computer interfaces is the command line, which surfaced as a more effective way to control computers in the 1950s. Previously, commands had to be fed into a computer in batches, usually via a punch card or paper tape. Teletype machines, which were normally used for telegraph transmissions, were adapted as a way for users to change commands partway through a process, and receive feedback from a computer in near real time.
Video display units allowed command line information to be displayed more rapidly. The VT100, a video terminal released by Digital Equipment Corporation (DEC) in 1978, is still emulated by some modern operating systems as a way to display the command line.
Graphical user interfaces, which emerged commercially in the 1980s, made computers much easier for most people to use, but the command line still offers substantial power and flexibility for expert users.
Nowadays, it’s hard to imagine a desktop computer without its iconic sidekick: the mouse.
Developed 41 years ago by Douglas Engelbart at the Stanford Research Institute, in California, the mouse is inextricably linked to the development of the modern computer and also played a crucial role in the rise of the graphic user interface. Engelbart demonstrated the mouse, along with several other key innovations, including hypertext and shared-screen collaboration, at an event in San Francisco in 1968.
Early computer mouses came in a variety of shapes and forms, many of which would be almost unrecognizable today. However, by the time mouses became commercially available in the 1980s, the mold was set. Three decades on and despite a few modifications (including the loss of its tail), the mouse remains relatively unchanged. That’s not to say that companies haven’t tried adding all manner of enhancements, including a mini joystick and an air ventilator to keep your hand sweat-free and cool.
Logitech alone has now sold more than a billion of these devices, but some believe that the mouse is on its last legs. The rise of other, more intuitive interfaces may finally loosen the mouse’s grip on us.
Despite stiff competition from track balls and button joysticks, the touchpad has emerged as the most popular interface for laptop computers.
With most touchpads, a user’s finger is sensed by detecting disruptions to an electric field caused by the finger’s natural capacitance. It’s a principle that was employed as far back as 1953 by Canadian pioneer of electronic music Hugh Le Caine, to control the timbre of the sounds produced by his early synthesizer, dubbed the Sackbut.
The touchpad is also important as a precursor to the touch-screen interface. And many touchpads now feature multitouch capabilities, expanding the range of possible uses. The first multitouch touchpad for a computer was demonstrated back in 1984, by Bill Buxton, then a professor of computer design and interaction at the University of Toronto and now also principle researcher at Microsoft.
The Multitouch Screen
Mention touch screen computers, and most people will think of Apple’s iPhone or Microsoft’s Surface. In truth, the technology is already a quarter of a century old, having debuted in the HP-150 computer in 1983. Long before desktop computers became common, basic touch screens were used in ATMs to allow customers, who were largely computer illiterate, to use computers without much training.
However, it’s fair to say that Apple’s iPhone has helped revive the potential of the approach with its multitouch screen. Several cell-phone manufacturers now offer multitouch devices, and both Windows 7 and future versions of Apple’s Macbook are expected to do the same. Various techniques can enable multitouch screens: capacitive sensing, infrared, surface acoustic waves, and, more recently, pressure sensing.
With this renaissance, we can expect a whole new lexicon of gestures designed to make it easier to manipulate data and call up commands. In fact, one challenge may be finding means to reproduce existing commands in an intuitive way, says August de los Reyes, a user-experience researcher who works on Microsoft’s Surface.
Compact magnetometers, accelerometers, and gyroscopes make it possible to track the movement of a device. Using both Nintendo’s Wii controller and the iPhone, users can control games and applications by physically maneuvering each device through the air. Similarly, it’s possible to pause and play music on Nokia’s 6600 cell phone simply by tapping the device twice.
New mobile applications are also starting to tap into this trend. Shut Up, for example, lets Nokia users silence their phone by simply turning it face down. Another app, called nAlertMe, uses a 3-D gestural passcode to prevent the device from being stolen. The handset will sound a shrill alarm if the user doesn’t move the device in a predefined pattern in midair to switch it on.
The next step in gesture recognition is to enable computers to better recognize hand and body movements visually. Sony’s Eye showed that simple movements can be recognized relatively easily. Tracking more complicated 3-D movements in irregular lighting is more difficult, however. Startups, including Xtr3D, based in Israel, and Soft Kinetic, based in Belgium, are developing computer vision software that uses infrared for whole-body-sensing gaming applications.
Oblong, a startup based in Los Angeles, has developed a “spatial operating system” that recognizes gestural commands, provided the user wears a pair of special gloves.
A field of research called haptics explores ways that technology can manipulate our sense of touch. Some game controllers already vibrate with an on-screen impact, and similarly, some cell phones shake when switched to silent.
More specialized haptic controllers include the PHANTOM, made by SensAble, based in Woburn, MA. These devices are already used for 3-D design and medical training–for example, allowing a surgeon to practice a complex procedure using a simulation that not only looks, but also feels, realistic.
Haptics could soon add another dimension to touch screens too: by better simulating the feeling of clicking a button when an icon is touched. Vincent Hayward, a leading expert in the field, at McGill University, in Montreal, Canada, has demonstrated how to generate different sensations associated with different icons on a “haptic button”. In the long term, Hayward believes that it will even be possible to use haptics to simulate the sensation of textures on a screen.
Speech recognition has always struggled to shake off a reputation for being sluggish, awkward, and, all too often, inaccurate. The technology has only really taken off in specialist areas where a constrained and narrow subset of language is employed or where users are willing to invest the time needed to train a system to recognize their voice.
This is now changing. As computers become more powerful and parsing algorithms smarter, speech recognition will continue to improve, says Robert Weidmen, VP of marketing for Nuance, the firm that makes Dragon Naturally Speaking.
Last year, Google launched a voice search app for the iPhone, allowing users to search without pressing any buttons. Another iPhone application, called Vlingo, can be used to control the device in other ways: in addition to searching, a user can dictate text messages and e-mails, or update his or her status on Facebook with a few simple commands. In the past, the challenge has been adding enough processing power for a cell phone. Now, however, faster data-transfer speeds mean that it’s possible to use remote servers to seamlessly handle the number crunching required.
An exciting emerging interface is augmented reality, an approach that fuses virtual information with the real world.
The earliest augmented-reality interfaces required complex and bulky motion-sensing and computer-graphics equipment. More recently, cell phones featuring powerful processing chips and sensors have to bring the technology within the reach of ordinary users.
Examples of mobile augmented reality include Nokia’s Mobile Augmented Reality Application (MARA) and Wikitude, an application developed for Google’s Android phone operating system. Both allow a user to view the real world through a camera screen with virtual annotations and tags overlaid on top. With MARA, this virtual data is harvested from the points of interest stored in the NavTeq satellite navigation application. Wikitude, as the name implies, gleans its data from Wikipedia.
These applications work by monitoring data from an arsenal of sensors: GPS receivers provide precise positioning information, digital compasses determine which way the device is pointing, and magnetometers or accelerometers calculate its orientation. A project called Nokia Image Space takes this a step further by allowing people to store experiences–images, video, sounds–in a particular place so that other people can retrieve them at the same spot.
In addition to enabling augmented reality, the GPS receivers now found in many phones can track people geographically. This is spawning a range of new games and applications that let you use your location as a form of input.
Google’s Latitude, for example, lets users show their position on a map by installing software on a GPS-enabled cell phone. As of October 2008, some 3,000 iPhone apps were already location aware. One such iPhone application is iNap, which is designed to monitor a person’s position and wake her up before she misses her train or bus stop. The idea for it came after Jelle Prins, of Dutch software development company Moop, was worried about missing his stop on the way to the airport. The app can connect to a popular train-scheduling program used in the Netherlands and automatically identify your stops based on your previous travel routines.
SafetyNet, a location-aware application developed for Google’s Android platform, lets user define parts of town that they deem to be generally unsafe. If they accidentally wander into one of these no-go areas, the program becomes active and will sound an alarm and automatically call 911 on speakerphone in response to a quick shake.
Perhaps the ultimate computer interface, and one that remains some way off, is mind control.
Surgical implants or electroencephalogram (EEG) sensors can be used to monitor the brain activity of people with severe forms of paralysis. With training, this technology can allow “locked in” patients to control a computer cursor to spell out messages or steer a wheelchair.
Some companies hope to bring the same kind of brain-computer interface (BCI) technology to the mainstream. Last month, Neurosky, based in San Jose, CA, announced the launch of its Bluetooth gaming headset designed to monitor simple EEG activity. The idea is that gamers can gain extra powers depending on how calm they are.
Beyond gaming, BCI technology could perhaps be used to help relieve stress and information overload. A BCI project called the Cognitive Cockpit (CogPit) uses EEG information in an attempt to reduce the information overload experienced by jet pilots.
The project, which was formerly funded by the U.S. government’s Defense Advanced Research Projects Agency (DARPA), is designed to discern when the pilot is being overloaded and manage the way that information is fed to him. For example, if he is already verbally communicating with base, it may be more appropriate to warn him of an incoming threat using visual means rather than through an audible alert. “By estimating their cognitive state from one moment to the next, we should be able to optimize the flow of information to them,” says Blair Dickson, a researcher on the project with U.K. defense-technology company Qinetiq.