Physical Interfaces

The main focus of this course has been on system interfaces with respect to how the software is presented or on "new" technologies like touchscreens or motion tracking. I also think that the design of a complete system includes the physical devices used as input. In my last post, I discussed how the mouse radically changed the computer world. The keyboard, on the other hand, has been around as long as processors themselves and haven't changed much. This is exactly why it is so important not to try to change the keyboard layout.

I looked into the evolution of keyboards and the majority of them look almost identical. The Ctrl, Alt and proprietary keys (Windows, Option, etc) are the only ones that seem to move. The biggest shift seems to be the release of the IBM 101-key "Enhanced" keyboard in 1986. This keyboard moved the CapsLock next to the A, swapping it with the Ctrl key, and moved the function keys to a new row above the number keys. It became the standard for all common computers and is still used to this day, despite research showing that other layouts, such as the Dvorak layout, are faster and more accurate.

Once typists and casual users learned the specific layout of the 101-key design, they became accustomed to it to the point that any other keyboard would require a complete re-learning. Even if one decided to learn the Dvorak layout, they would still be forced to use QWERTY present on nearly every other machine they used.

The main point is that at this point in history, it would be nearly impossible to replace or redesign the keyboard. When we design interfaces, it is critical to evaluate how old systems worked and how willing users would be to change to a new design.

The Old Macintosh

I was browsing my Instant Netflix queue the other day when I came across a documentary titled "Welcome to Macintosh". I watched to movie keeping the objectives and concepts of system interface design at the forefront of my mind.

I knew that the development of a GUI was a huge step in the technology world, but I never realized how long it took and how novel the concept was. The movie featured interviews from several long-serving or Apple employees that shared the process of developing many of the technologies that went into the major systems such as the Apple II, Macintosh, iMac and iPod/iPhone.

After the success of the Apple II, a small team of about a dozen designed the first Macintosh, the first GUI and the first successful point and click environment. Most of the original artwork was done on graph paper that could easily be translated into bitmaps because the team could not afford to build a prototype for each contributer. The text editor was capable of several different fonts of different sizes.

Today GUIs and point-and-click have evolved, but many of the principles remain the same. We still use images of folders and disks to give a metaphorical representation of file systems. We still use windows to manage viewing different applications. One clip in the movie reference an IBM quote saying, "No computer that can fit on a desk could ever be powerful enough to do worthwhile computing." The Macintosh didn't have the speed or memory of a contemporary IBM mainframe, but certainly left a much greater impact on the world.

I think that it is more important to focus on interface design than it is to focus on the “smarts” of a system. If users are able to look at and interface and understand how to use it properly, then smarts are not required. With a well designed interface, users typically have more options, especially advanced options. One example of an easy interface but little AI is the automated checkout system in supermarkets.

Another example of an intuitive interface without much smarts is the Apple iPod. Spinning my finger around the wheel is an easy motion that relates well to what I am trying to do. I spin clockwise to turn up the volume. Click the middle button to select a menu item. Easily scroll up and down through menus. Skip songs by pressing forward. My old MP3 player (the original Nomad Jukebox) had several more buttons with specific tasks. The iPod consolidated the buttons to just the scroll wheel. This is a huge advantage because I can control the player with one hand while walking or doing some other task.

to

This opinion does, of course, come from an engineering-oriented individual. I enjoy experimenting with new technologies and learn from trying things out on my own. Other people who might not be so tech savvy can often be deterred by the newest advancements. This is where some would argue the system intelligence would be able to figure out what they are trying to do. I would argue that unless the system has an approachable interface, the new users would never try the system to start with.

Minority Report Interface via Wiimote

This week we will be working on an introduction to using and tracking the movement of the Wiimote. We will measure and analyze the accelerometer readings from a variety of hand motions. I looked into some other interesting uses for the remote and found this video that demonstrates one potential use.

Essentially, the Wiimote is serving as an infra-red camera (none of the accelerometer features are being used) to detect the reflection of infra-red off of fingers or reflective tape. Using the common Bluetooth protocol, it is easy to have the Wiimote communicate with a variety of devices, like the PC in the demo. Johnny easily moves, twists, and zooms a simple image around the screen using intuitive hand motions. He does mention that continually moving hands around in the air, while fun, does wear the user out.

Due to the physical demands this type of interface requires, I thought of two good uses. The first would be in a demonstration environment. Sales people at trade shows could interactively move computer models of their products rather than show a pre-rendered animation. If a customer wanted to see a specific angle, the sales person would spin the product to that angle and adjust the zoom. The second practical use would be physical therapy. Patients would be instructed to move their arms around to "touch" objects on the screen. The level of therapy would be easy to adjust based on each patient's needs. Doctors could design sequences that would target specific joint movements by placing objects in patterns that would intuitively lead to a wrist twist, elbow bend or shoulder stretch.