GUI Gets a Makeover

Sleeker, more contextual user interfaces with support for very large screens and gesture-based interaction are in the offing.

The graphical user interface — keyboard, mouse, windowing system — has dominated personal computing for the past 20 years. GUIs today represent the culmination of innovative work at organizations ranging from Xerox PARC in the ’70s to Apple Computer Inc. in the ’80s. But it was Microsoft Corp. that popularized it for mainstream business computing in the ’90s. Today, Windows and the Office productivity suite have become the GUI standard-bearers for the business desktop.

With major revisions of both products ready to roll out, the enterprise desktop GUI is getting its first significant face-lift in years. Most of those changes, such as the Aero 3-D features in Windows Vista, are small steps forward. Others, such as Office 2007’s ribbon bar, represent a bigger leap — both for the GUI and for end users.

More changes lie ahead. The GUI will get better at simplifying user options by filtering them down based on social or operational contexts — or where a mobile computing device is being used, researchers say. Tomorrow’s GUIs will adapt to bigger screens and multiple displays by rearranging the desktop and relegating different content to primary and secondary displays. Larger display acreage could also push gesture-based input devices such as touch screens, digitizing pads and the stylus into the mainstream.

The increased complexity of today’s computer systems is forcing change upon the GUI. As the number of features has exploded, users have been overwhelmed with layer after layer of icons, tool bars and menu options.

Both Office 2007 and Vista include user interface changes designed to the reduce visual clutter and help users find what they’re looking for. Vista’s Thumbnail Taskbar, for example, shows images of running applications in the task switcher or when the user moves the cursor over program icons in the taskbar. Live icons allow similar visual navigation for documents.A new Sidebar feature introduces a transparent pane of “Gadgets” — single-function applets such as a clock or calendar. And Aero’s 3-D effects make the desktop appear less cluttered by making windows transparent.

Office 2007 tweaks include previews that show how selected text will look when the cursor hovers over a style option, and font options that now show what the desired font looks like. But the biggest changes Microsoft has made involve the removal of traditional pull-down menus and the introduction of a contextual ribbon bar. Perhaps no productivity software has suffered more from features bloat than Microsoft Office, so it’s no surprise that Microsoft has attacked this problem in Office 2007.

“Creators of the paradigms of menus and tool bars never anticipated that software would be able to do as much as it does today,” says Jensen Harris, principal lead program manager for Office. While Word and other early programs had 10 to 30 features each, the word processing application now has some 1,500 and the Office suite has more than 10,000.

As the number of features exploded, the preponderance of options overwhelmed the menus, tool bars and other mechanisms for presenting them. This left users searching for key features and unaware that others even existed. Attempts to reduce the clutter in Office, such as cascading menus, only added to the confusion. The Intelli¿menu, a feature that removed menu options that weren’t used recently, was designed to “reduce the perception of bloat in Office,” says Harris. Instead, it annoyed users by making menus inconsistent and unpredictable, he explains.

So Microsoft has eliminated pull-down menus in the core Office 2007 applications, as well as in Vista’s Internet Explorer and Windows Explorer programs. In Office, the primary applications now use a ribbon bar, which presents only those features that pertain to the currently selected object. The ribbon bar changes as different items, such as a table or paragraph, are selected.

Jakob Nielsen, a principal at Nielsen Norman Group, views contextual interfaces such as the ribbon bar as the wave of the future. “The only way that this amount of complexity is in any way capable of being controlled by humans is by ... making it easier to use each feature,” he says.

One alternative to contextualization is to reduce complexity by reducing features. But that’s something that companies like Microsoft — which sell successive versions of software by adding features — are loathe to do. It’s also not practical, because users want those features to be available even if they don’t use them, says Steve Kleynhans, an analyst at Gartner Inc.

“People want more, yet they want it presented as less,” Nielsen says.

By making more commands visible in each context, vendors hope people will use more of those whiz-bang features — what Harris calls “exposing the richness” of Office. But will they?

“The vast majority of users never bother exploring more of the capabilities available to them because users’ motivation is to get things done,” says Nielsen. That’s a challenge not just because new features justify upgrades, but because users could also increase their productivity. “They’re not advancing in skill level, [and] there is no good solution to that,” Nielsen says.

Researchers are working on rolling up commonly used commands into higher-order functions. “The key of design is to try to figure out how you can cluster those low-level operations into a single higher-level operator that better fits your way of thinking,” says Bill Buxton, principal researcher at Microsoft Research.

A simple example is the pen or stylus, which lets a user select and delete text in a single stroke. “The functionality is the same as cut and paste, but the number of steps taken and ... the time taken are drastically reduced,” Buxton says.

Morphing for Mobility

Mobile computing has placed a new set of demands on the GUI. As functions formerly handled on the desktop migrate onto mobile devices, providing consistency across those devices and adapting the user interface to different contexts doesn’t always work well.

The GUI was originally designed for a stationary PC on a desktop. Initial attempts to extend it to mobile devices with Windows CE proved challenging. “With mobile, I don’t want everything. I want what I need,” says Jerome Nadel, vice president at usability consulting firm Human Factors International Inc. in Fairfield, Iowa. Newer versions have done a better job of offering the user shortcuts, he adds.

As the GUI evolves, it must adapt not just to the device but also to the user’s situation. “The interface has to have the dexterity to let me work within the device in the context of where I am,” Buxton says. For example, choosing voice commands may be preferable when using a PDA in a moving car, but the keypad is preferable in another context — a crowded airplane. Devices in the future could switch automatically, adapting to situations such as when the user is in a noisy area where voice commands won’t work.

In PDA phones, the user interface can take advantage of location-based services to reduce complexity, filtering down search lists based on where the user is and what he’s doing, says Nadel.

Some cross-pollination between the desktop and mobile GUI attributes is also occurring. Dual-mode user interfaces that include voice controls will go mainstream first on PDA phones but could eventually percolate up to laptops and desktops, Buxton says. A dual-mode interface is already in the works for Outlook/Exchange users, who will soon be able to access and change calendar appointments and e-mail on the road using either the PDA’s graphical e-mail client or by calling in to an interactive voice response system.

Big-Screen Computing

“The best way to improve productivity is to give everyone a larger screen,” says Ben Schneiderman, professor and founding director of the Human-Computer Interaction Lab at the University of Maryland. That’s just around the corner, thanks to dropping LCD prices. But the desktop GUI doesn’t scale up to larger screens very well — at least not yet.

The use of multiple displays will go mainstream by 2009, according to a Gartner report. Display sizes will increase to the point where 1920 x 1080 dpi resolution and 25-in. or larger screens will be common. By the end of the decade, computers may also incorporate secondary, lower-resolution e-paper displays that can maintain an image even when a laptop or desktop computer is turned off.

“In five to seven years, it will be cheaper per square foot to put up a 100-dpi electronic display than to have a whiteboard,” says Buxton. That will lead to wider adoption of gesture-based input devices such as touch screens. “You’ll be able to work on free-form, larger surfaces in the way a football coach does [on a whiteboard],” he says.

The Windows GUI is already evolving to support secondary displays. SideShow, a new feature in Vista, will enable users to view items such as calendars or to-do lists on secondary e-paper laptop displays that might be mounted on the top of the case.

“As [displays] get bigger and bigger, you can get more information to the user,” says Mary Czerwinski, principal researcher at Microsoft Research. But the current desktop GUI, which simply extends the same desktop across multiple screens, doesn’t scale well. With more screen real estate available, computers will begin monitoring and presenting more information to the user.

“You’ll have a focus area that’s easy to grok,” Czerwinski says. “But then you will have all of these peripheral things you’re monitoring. These need to be glanceable.” Peripheral objects that might be automatically exported to secondary screens include things like the single-function Gadget applets in Vista’s Sidebar.

“The screens themselves will wind up being contextual,” says Gartner’s Kleynhans, and the ability to arrange objects spatially will become more important. Multimedia elements such as video will be integrated as just another window or information feed, like the stock ticker programs available for Windows.

“People need to be able to bring up a video and still have other information presented in a transparent or translucent way on top of that video,” says Kleynhans. He thinks GUI designs could do more with 3-D technology as a navigational aid. That work needs to go beyond current efforts like Aero’s 3-D effects in Vista, which he dismisses as “eye candy.”

Buxton says that will happen. “We can take the skills that we’ve learned from living in the everyday 3-D world and use it to help us navigate through the information space of computers,” he says. One challenge with 3-D is the complexity and processing overhead required. But, says Buxton, “we can get 80% of the benefits for 20% of the complexity if we design it right.”

For example, a 3-D metaphor could be used as a navigation tool — even if the objects are two-dimensional. In this context, a camera phone would be used to navigate through a 2-D spreadsheet on-screen by moving the camera to pan back and forth or zoom in or out.

The technology to do so already exists in today’s PDA phones, Buxton says. “If you do a real-time analysis of the signal from a smart phone’s camera, you can determine all you need to drive the content of the digital information on the screen,” he says.

As personal computing changes and fragments, the GUI is evolving with it. “There is a different way to think of interaction as we move to different devices and different places [for] using them,” Buxton says. Researchers have plenty of ideas for enhancing the user interface to meet that challenge. The hard part for Microsoft and other vendors will be deciding which ones will be a hit with users.

Sometimes GUI initiatives fall flat. See design experts' take on major GUI gaffes.

Copyright © 2006 IDG Communications, Inc.

7 inconvenient truths about the hybrid work trend
Shop Tech Products at Amazon