THE FUTURE OF INTERFACES

T H E   F U T U R E   O F   I N T E R F A C E S

With the release of Google Glass fast approaching, and the wide variety of alternative input devices being invented by ambitious startups, I feel the need to share my thoughts on the future of human-computer interaction. Specifically, I want to disprove the notion that the days of the keyboard and mouse are over, and shed some light on why some people think new input devices are going to change the way we use computers forever.

Let’s start with an example of a good new kind of interface – the Touch interface. Touch screens, made popular by smartphones and tablets, present an intuitive way to interact with Graphical User Interfaces (GUIs). When something needs to be investigated or pressed, a user taps that section of the screen with the tip of their finger. This is received as a “click” and the system pulls the information up for the user. Gestures like finger “swipes” come naturally to us, since the human thumb is rather dexterous and can be used when holding a phone in one hand.

However, a good example like Touch interfaces is not without its share of  bad qualities as well. Touch interfaces in the wrong scenario (desktop monitors with touch controls enabled, or large “SmartBoards”) are clunky and unusable. This is because touch controls are designed for handheld devices like the smartphone or Nintendo 3DS. When the interactive screen is too large or far from the user’s hands, this method of interaction gets very tiring after a while, and the intuitive nature of the system degrades into a tedious chore. While Touch is good for fleeting interactions, menus, and games, it does not present users with an easy way to get "real work" - such as file management, extensive writing or programming, and detailed image manipulation - finished in an easy way. Most Touch users would probably agree that when real work needs to be done, the keyboard and the mouse provides the quickest route to finishing computer work.

Touch is our first example of how people’s need for convenient interfaces overshadow the novelty of a new kind of interactive medium. We will see this pattern repeat later on. Let us now move on to Voice commands.

Voice controls, a key component of the Google Glass system, present an interesting interactive dilemma. This is because Voice controls (if they are reliable) provide a quick way to command your computer without the need for nested menus. Simply say any command the computer recognizes, and your machine obeys. However, this practicality is somewhat of an illusion, because Voice controlled machines are vulnerable to outside interference as well as invasions of privacy. Again the issue comes down to “real work” – if you want to mess around with Google Glass and tell it to “take a picture, Glass!” that’s perfectly fine. You may not mind if other people overhear you. However, if you want to send a private message to a friend about something sensitive – even if it’s a short message – shouting it aloud to your machine is not the best way to keep your interaction secret. And as I have seen with some Google Glass demonstrations, you are not the only voice that commands your Glass. This means at the very least that using it exposes you to friend’s practical jokes, or perhaps other user’s nearby commands to their own machines!

Google Glass brings us to our next Human Computer Interactive component, which is Augmented Reality, related to Virtual Reality. While Virtual Reality is sort of a novelty of the simulation and gaming industry (something fun that will never become the “standard way” of experiencing the medium due to its high price point and built-in introversion), Augmented Reality is a bit different. Through smartphone apps and Google Glass, developers are trying to convince us we need Augmented Reality to receive our information. For those not familiar with the term, Augmented Reality attempts to overlay a layer of virtual information over images of our real world – thus, you may look through the lens of your Google Glass and see a sign over the Babbio Center that says “Babbio Center” – and some apps let you delver even deeper into the information they present. Again, we as users are presented with a method of receiving information that is novel, yet not necessarily any more helpful than standard menu interfaces that present information quickly. Developers of “revolutionary” technologies often fail to see that their users want information instantly and without any fluff – and they should not be surprised when their technologies are used for mere moment before they are discarded for more standard forms of information visualization.

The worst of all of these “innovative” input methods is the Gesture. Here I do not mean Touch Gestures like “swipes” or “pinches” that you can do on your tablet or smartphone. Instead I am referring to interfaces that sense your 3D skeleton and allow you to command your machine with hand movements, without touching the computer at all. The Microsoft Kinect, the LEAP Motion Sensor, and Panasonic's D-Imager are all examples of this. While the idea sounds cool, and seems to work well in science fiction, (Tony Stark’s version of gesture controls sparked many imaginations in the recent Iron Man series) it seems to belong purely to the realm of imagination. This is because Gestures go against every urge we have as humans to control computers. The first issue, one of depriving users of haptic feedback, is a serious problem. Without being able to touch our machines, we lose one of the key senses we use to interact with computers, which is touch. This leaves us with vision and hearing to determine if our command successfully was performed, leaving users with a sense of abandonment that develops into mistrust as time spent with the machine increases.

Secondly, there is a myth floating around the marketing community for these devices that the use of gesture is somehow natural or intuitive. This could not be further from the truth – for every software development kit has some unique way of controlling devices using their gestures, which varies based on the poor quality of the system and the lack of imagination within the development community. Indeed, most gesture controls are simply large-scale touch gestures (swipe with your arm, instead of your finger!) that do not come naturally to humans. Hand gestures are an anachronism of evolution, some kind of byproduct from a time when humans could not communicate with just their voices. They are unnecessary and are merely used for dramatic effect, especially by Italian Americans like myself. How this could possibly translate to interacting with machines is beyond my understanding – and if it did, there would still be the issues of getting “real work” done in private.

The last issue with gesture controls, and the final topic of this post, is the problem of non-immediate control mechanisms. To illustrate this, visualize this example – you are writing in a Word document, and you press the “M” key. As soon as the button is pressed, an “m” appears on the screen. The time between beginning the command and ending the command is so short that humans barely notice it – and thus, our interaction with the computer feels instant, as it should seem. Gestures fail this basic component of Human Computer Interaction because of how long they take to perform and be recognized. When you swipe your arm across the air in front of a screen, the system must wait until your command is finished to process it – lest it confuse it with something else and perform a separate function, leading to further frustration. This gap between call and response is what makes these interfaces so irritating to use, and will certainly lead to them never gaining much popularity in scenarios where “real work” must be done.

Anyone who thinks I am merely raging at new technology like an old man should look at history. New interfaces are nothing new at all. Nintendo released stupid things like the Virtual Boy and the Power Glove in its infancy – they were fun for a while until they became unwieldy or caused sickness in the user. And the only reason Tony Stark’s gestures work so well is because they are entirely animated by visual effects artists in postproduction who are reading what should happen in the scene from a script they have next to their desks.