Play Incorporated's goal is to democratize the medium of television by creating tools that allow anyone to create and distribute broadcast television-quality video productions. As an example, Play recently launched PlayTV.com, an Internet-only television station that broadcasts worldwide 12 hours of original programming every weekday. Play's headquarters are in Sacramento, California.
At Play, we believe human interface design is still in its infancy, and as an industry, much needs to be done to allow people to derive maximum benefit from the digital tools available. Today's software is simply too difficult to use. We strive to go beyond the accepted normal interface and to innovate wherever possible. Our development teams consist of designers, programmers, and artists who are passionate about the products they create, and whenever possible, are actually creating the product for their own use. As anyone who has shipped commercial products knows, creating great products is incredibly difficult. It requires an equal mix of art and science, and passion and discipline. We feel that without the care of the creators, adherence to all the interface rules in the world will not make anything other than an emotionally flat and unsatisfying end-user experience.
Most products are purchased to fulfill a particular goal or need. The actual goals are quite varied, but nonetheless, they can all be generally classified as something the user wants to achieve. When purchasing a game, people want to be entertained. When purchasing a graphical modeling package, people want to be able to create shapes. It seems simple, but all too often designers get caught up in the details and lose track of the primary purpose of the product. Good CHI design always keeps the end result of the product in mind.
In particular, the first time someone uses a product is a very delicate time, when most users experience some level of apprehension. Novice computer users are often concerned about their ability to understand how a new product works. Advanced users may be doubtful of the product's ability to perform the task at hand, or have preconceived notions from using similar products. In any case, the sooner users can achieve goals, or at least see some part of a goal tangibly fulfilled, the better. Time is an incredibly important factor in these early stages. Many people become frustrated if the product requires too much learning, preparation, or reading before results can be achieved. Any early frustration with a product can easily lead to the user abandoning the product.
Instant gratification is a great way of providing the user with an timely sense of accomplishment. For instance, in our 3-D modeling product Amorphium, the user is immediately presented with a shaded 3-D sphere. The user can interactively push, pull, stretch, and twist the shaded sphere as if it were virtual Silly Putty, quickly creating 3-D shapes. The initial experience is rewarding because upon startup there is no need to create a "new project," define the "workspace size," define "views," create "polygons" and "surfaces," or any other stops to accomplishing the primary goal: Making three-dimensional objects. In this case, the mathematics behind the three-dimensional tools are kept neatly hidden from the user, who is free to mold a shape that behaves much like a real-world object rather than a geometric abstraction.
Many software products require extensive preparatory work before the actual task at hand can be started. This often takes the form of configuration dialog boxes that must be completed before the product can be used. Mandatory user registration procedures are often another culprit. We try to eliminate as many steps as possible before the user actually starts working on, and seeing tangible progress toward a goal.
Much is made in the industry about the definition and adherence to interface conventions and standards. The basic idea is that if software applications behave similarly, users only have to learn these behaviors once and that knowledge can be applied to all future software products. This idea is made concrete in most modern operating systems, which provide programmers with relatively easy methods to make programs that have all the standard conventions, such as documents, views on those documents, toolbars, and menus.
At Play, we believe that strict adherence to these conventions is creatively stifling at best and incorrectly simplifies product design. In particular, the Microsoft Windows OS contains interface libraries that make it very easy to create applications such as Microsoft Word but difficult to create very much else. Since most computer users are familiar with standard OS conventions, it is important to learn them inside and out to understand what users know and expect. However, when it comes time to actually design a product, we do not hesitate to disregard these conventions if we think of a better interface for the task at hand. The graphical design and layout of an interface is also constrained by using the default OS libraries. Instead of a plain vanilla user experience, more companies, including Play, are tailoring the appearance of each program to the current task.
In support of this creative freedom, it is important that product development teams have software programmers who create custom user interface libraries that go beyond the basic operating system fare to make the overall experience better. Such libraries go beyond OS provisions with improvements such as arbitrarily shaped user controls and windows, alternate graphical rendering styles, and even 3-D animation capabilities. This provides CHI designers with a greater freedom to fully explore the possibilities of each application. A side benefit of adhering to a custom interface library is that it can make the process of porting software to other operating systems simpler, since in programming, the OS is bypassed in favor of the custom library.
Play's product design process is oriented around rapid prototyping. Both programmers and designers are encouraged to simply try various ideas and methods via rough, quick approximations. Feedback on these implementations is obtained from the rest of the product team and potential end users, and then another prototype is created. We found that extensively designing products on paper prior to implementation is usually a waste of time and effort. Our process is quite informal, since we've found too much formality tends to kill passion, limit creativity, and slow development to a crawl. Designers are free to experiment with various concepts, as often half-baked ideas are saved and used in later projects to great effect.
Play takes a holistic approach to product design, taking into account software, hardware, user interface, product box, documentation, tutorial content, advertisements, and Web presence. Each customer's overall product experience is a composite of their individual experiences with all of these elements, so it makes sense to design the entire product as one large experience. For instance, the light, humorous tone of the Gizmos written documentation drove similar language in the advertising, on-line help system, and product packaging, providing a unified feel to the product.
We believe the best job satisfaction for designers is an extremely personal acknowledgment given upon the completion and shipment of a product. Relying on external factors such as magazine reviews and customer comments is dangerous, as it offers competitors and those of ill intent a quick way of destroying confidence and halting momentum. Although it is to some extent human nature to desire and enjoy the praise of others, depending on their approval is a flimsy structure on which to stand. We encourage development team members to use the products they have created and thereby verify their accomplishments to themselves.
Product interface design continues to change at a fast pace, as more powerful hardware allows for the creation of more compelling experiences. While previous computer software required only standard 2-D layout skill, Play's designers are now required to be fluent in many disciplines, from three-dimensional modeling and rendering to digital audio design. The breadth of skills required will continue to increase as new technologies add more possibilities and continue to enrich computer-human interaction.
One of the primary goals of Gizmos 98 was to improve the end user experience for Windows 98 users. Many parts of Windows 98 remain basically unchanged since Windows 3.1, including its accessories, multimedia tools, and included entertainment software. Mass-market PC hardware capabilities such as 24-bit graphic display cards and stereo sound cards were not being used to full effect in Windows 98. We decided to create a set of tools, including calculators, calendars, clocks, card games, and other programs that took full advantage of the multimedia PC. Where applicable, these tools would mimic real-world equivalents in every possible way.
Making the interface appear as a real gadget was the first goal. Early attempts to photograph and digitize pictures of real-world objects were quickly abandoned in favor of using advanced 3-D animation software. This made it possible to design interfaces that looked photorealistic while still maintaining full creative control over their shape, appearance, and functionality. Accurate lighting cues such as highlights and shadows were calculated by the 3-D render engine, adding to the illusion of real-world objects. User controls, such as buttons, were rendered at each possible state (such as selected and deselected), and the resulting bitmap pieces saved for use at run-time. This allowed, for example, the calculator buttons' shadow to disappear when pressed, adding to interest and realism.
Environmental sound effects were used in subtle ways. Actual calculators, clocks, and other objects were digitized, performing various functions, and these sound samples were digitally cleaned up and used in the software. Only real-world sounds were used and only at a low volume level as another realistic cue. Based on early user testing, the use of sound effects was made an option since some users loved them and some did not.
The creation of this interface was typical of Play products, requiring a combination of graphic and audio design, end user testing, and new custom software code to create the desired effect. Because the tools in Gizmos look, sound, and behave almost identically to their real-world counterparts, most people feel comfortable using them right away.
©2000 ACM 1072-5220/00/0200 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2000 ACM, Inc.