Determining When to Use Show-Me Helps and Demos

By Annette Norris Bradford

The availability of powerful yet easy-to-use multimedia tools enables technical writers to consider a powerful new form of embedded user assistance: show-me help. This paper provides an overview of who is currently using show-me help—some current research, some history, and some definitions. It offers some guidance in choosing tools, designing show-me help, and deciding when to include then, concentrating on consideration of your users, potential topics, subsequent releases, and translation. It also suggests how show-me helps can be reused as part of product education and single-sourced into user assistance from the Web.

Introduction

In 1986, I was IBM Information Development's corporate-wide representative to SHARE, a user group for system administrators who keep vital mainframes running "24 x 7." One feature of SHARE is "SCIDS," the nightly welcome party and informal technical exchange. During my first night at SCIDS, I found table after table for attendees selling buttons that represented this group's frustrations: the perceived glacial speed of key applications, the movement toward shipping "OCO" (object code only), and the failure of users to read the documentation before paging the system support people. "RTFM?" I asked the button seller. "What does that mean?" "Read the manual," the button purveyor shot back. (I think the "F" meant "fascinating", or perhaps "factually correct"). I bought one. Here at last, I decided, was someone even more frustrated than I was that users do not read documentation.

As technical writers, we do our best to make our information accurate and usable and even interesting, but secretly we know that the major hurdle to solving system problems with documentation is getting customers to read it. Research conducted 15 years ago found that documentation was the last place that users turned to. The first place was a colleague. Users walked down the hall and knocked on coworkers' doors to ask, "How do you…?" Fierce debates ensued about the impact of this "ask the local expert" strategy on the productivity of the local expert. And technical writers spoke longingly of "capturing" that local expert and making him part of the documentation. Now we can.

The first time I found an expert captive inside a software program was in 1995. Software was shipping on CD-ROM by this time, and storage was becoming less important than usability. I was using a tax preparation program when I hit a snag: an obscure form that I did not know I needed. I looked around for a key to launch online help, but found instead a button that launched a short movie, in which a tax expert explained the form and walked me through completing it. Software reviewers loved this feature. I loved this feature.

No, it is more accurate to say that as a twenty-year veteran of "No, multimedia is too expensive to develop and ship," I was elated with this feature. It meant that I had survived the dark ages when we were constrained from adding multimedia information that we knew our customers needed and wanted because of the cost of developing and shipping it. I had survived primitive, platform-dependent tools that required me to learn tool-specific programming languages to make a simple object like a cursor move across the screen. No more 16-color palettes. No more timing every object on the screen. As Stephen Jobs demonstrated at the rollout of the new generation of Apple operating system with its embedded multimedia suite, it has become possible for a resourceful 12-year-old skateboarder to take commonly available tools and produce a near commercial-quality movie. The technology for creating effective multimedia is now accessible to anyone.

Uses of Multimedia Past and Present

There is a growing body of empirical and anecdotal evidence reinforcing the utility of animations embedded in product help.

Certainly one of this strategy's most respected and vocal advocates is Dr. Ben Shneiderman. In his opening session address to the 2004 STC Conference, Shneiderman said that often textual help is at odds with how we use a system interface and advocated graphic help and task-oriented animations. He described show-me helps that he had created using the TechSmith® Camtasia Studio™ product. He explained that these animations, some as short as 15 seconds, eliminated "how do I" questions by providing a visual, animated walk-through that ran in a Flash player with voice-over, Shneiderman recommended against providing videos because screen capture show me's of the type created with Camtasia or ViewletBuilder produced higher-resolution solutions, (1) though some of our colleagues have used video show me's very effectively. (2)

As early as 1991, Palmiter, Elkerton, and Battett showed that users were faster and more accurate when performing tasks after being shown animated demonstrations rather than textual explanations. (3)

Long before that, instructional designers understood the potential of computers as instruments for education. Let's begin this discussion of show-me help with some history and some definitions.

Some History

The use of multimedia to present computer training information has a long and rich history. By the 1970s, there were viable mainframe-based computer-aided instruction (CAI) systems that large companies like IBM used to deliver internal on-demand service education. The early systems provided text-only monochrome modules that required complex and proprietary programming languages to create and frequent breaks to survive. There was no animation, and even the simplest interactions, such as answering questions, required that the developer have programming skills.

As computers evolved, multimedia tools did too. By the 1980s, the dominant acronym associated with education delivered by computer was CBT—computer-based training. But CBT systems remained largely the domain of trainers, and CBT courses were time-consuming and expensive to produce. As CBT tools moved from mainframes to workstations, developers still wrestled with primitive graphics created with tightly controlled and limited palettes that had to be created on a specific platform and run on a specific platform.

Today, these restrictions have given way to easy-to-use Java-based tools that run in free, widely-available universal players like Macromedia Flash®. Multimedia technology has come of age.

Some Definitions

With this proliferation of programs and technologies, many types of multimedia deliverables have emerged, clouding the terminology picture. I want to define the show-me help in the context of other more common multimedia deliverables.

Tutorial: On the Web, the term " tutorial" most often means a flat file with basic education information in it. I define it as a computer program that uses some combination of modularized, menu-accessible information, animations, graphics, guided demonstrations, simulations, and interactivity to familiarize users with new concepts or products and teach them the basics of use. The purpose is educational. Very structured tutorials are often delivered as computer-based training (CBT).

Demonstrations: Demos are multimedia packages with high visual impact that serve as marketing tools or as orientation information for a new product or concept.

Guided Tour: A cross between a tutorial and a demonstration, a guided tour is meant to teach instead of entertain but does so with less interactivity and more "watch while I do this" design. Guided tours are meant to get users up and running quickly on the premise that people can perform new tasks better if they see them than if they read about them.

How do these differ from show-me help?

• Show-me helps are shorter than traditional tutorials. They teach less and what they teach is focused on a specific task.
• Show-me helps are like demos in that they offer little or no opportunity for interactivity, but unlike demos, their goal is to explain a function, not to sell a new product.
• Show-me helps are most like the guided tours that come with some products or software packages. They teach but with less interactivity than a tutorial, and in smaller, context-specific doses. One scenario from a guided tour, perhaps, might be reused as a show-me help. A number of show me's offered from a menu preceded by some marketing and conceptual information to introduce them might be joined to form a guided tour.
• Show-me helps are embedded, not standalone. This is the biggest distinction between show-me help and other forms of multimedia. A show me appears in the context of the problem it is trying to solve.

Standalone versus Embedded

I have written elsewhere at length about standalone forms of multimedia user assistance (4,5,6,7) and the impact they have on providing a positive user experience.

Embedded multimedia has much more potential benefit to a computer user than standalone because the answer to a user's question is embedded in the context in which the question arose. What is significant about user interfaces that answer our questions, Andrea Ames says in her STC paper "An Introduction to Embedded Assistance," is that "we proactively prevent task failure and keep the user within the task at hand." (8)

The importance of keeping the user within the task at hand was shown effectively by Knabe in his 1995 study of online help. Knabe advocated integrated guidance to overcome such problems as users having to switch from one open application to another (the application they are learning and the help system), dealing with multiple overlapping windows, and mistaking illustrations in the online help with the actual interface. (9)

In IBM, making show-me helps part of embedded user assistance is particularly significant. This shift from standalone multimedia deliverables created by marketing to embedded, integrated extensions of traditional user assistance maintained by technical writers has plan for, develop, and maintain these deliverables.

In the past, when a product needed a demo, a contractor or a department that creates marketing deliverables created it. It was created outside the product development team, so it lacked awareness of product terminology and usually did not follow IBM style. And it was often very expensive to create and not kept up to date. Sometimes the need for multimedia demos was so acute that a marketing or technical person created a "quick and dirty screen cam" and attached an external script. The audience for these quick and dirty demos was usually marketer-to-marketer, but sometimes these demos proved so useful, even in their primitive form, that they were posted on public Web sites.

When Technical Writers Create Multimedia

As technical writers step up and ask for ownership of multimedia deliverables and make them part of the usual product development process, good and bad things happen. But there are also professional rewards to consider that might push you toward trying show-me helps.

Positive Aspects

Good things happen to multimedia deliverables when you get technical writers involved. Multimedia developed by technical writers who are part of the development team:

• Has all the quality in design and execution and rigor that characterizes our other product information.
• Is created using the same terminology and concepts used by the rest of the product.
• Is developed by a team that includes subject matter experts and marketers.
• Is reviewed by a team of product experts
• Is edited so that it is clean, concise, and in the same style as the rest of the documentation.
• Is maintained from release to release as any other part of the technical documentation is maintained.
• Helps product documentation achieve better customer satisfaction metrics. Our experience with show-me helps says that customers like them a lot.

Negative Aspects

Nothing is ever perfect. Here are some potentially negative planning impacts you should consider.

Doing show-me help is a commitment. It cannot be made part of the product information by a zealous multimedia advocate (such as I) who then moves to another job and leaves less enthusiastic people to either maintain it or, worse yet, drop it. Once you provide something to customers, it is hard to take it away. Even if the resources for a project drop, someone on the team has to maintain the multimedia deliverables. Offering show-me helps is an organizational, not an individual, commitment.

Another aspect of this commitment is translation. Multimedia marketing deliverables are seldom translated. No one even asks, "Can it be translated" when marketing commissions multimedia information. But when a multimedia deliverable becomes part of the normal product information delivery process, that it must be translatable is a given if the other product information is translated. So now, we are not just requiring new tools and skills out of our technical writers; we are requiring them from our translation people as well.

Personal and Professional Rewards

Sometimes in the past, when I have jumped into a new technology with a new unfamiliar set of tools and new commitments, I have lived to regret it. And you might be asking yourself about now, "Why would I want to sign up for this?"

The first reason is partly altruistic. We are all looking for ways to improve the experience our users have with our products, and initial customer feedback on show-me helps is very positive. Customers like them. Our success is linked to our customers' success, and as technical writing professionals, we are measured in part by customer satisfaction with our work.

Second, this is not a big stretch for you. You already have the many of the skills you need. We have years of experience with capturing screens, using graphics effectively, understanding and "translating" technical information, and with understanding how the product marketing people and technical evangelists intend to sell the product. Show-me help is a new application for an existing set of skills.

Third, in an uncertain job market, there is no such thing as too many skills. Creating multimedia show me's gives us yet another skill that we can add to our professional portfolios to make us more marketable.

Finally (and don't tell anyone this), it's fun. The tools are neat. The reward seems to exceed the effort. It is a small investment that can reap big rewards. Who doesn't like that?

Choosing a Tool

So you're convinced and you're ready to dive in. Which tool do you select? The current "big three" for creating show me's consisting of animated, commented screen captures are the TechSmith Camtasia Studio product, the Macromedia Robohelp 5 product, and the Qarbon ViewletBuilder product. All three products can create the same kinds of deliverables and have similar features.

My expertise is with the ViewletBuilder tool, and that is the one I will be using in this workshop. Why do I like ViewletBuilder? Simply stated, it is the easiest multimedia tool I have ever seen, and I have seen a lot. It is easy to learn, easy to use, and easy to expand. And the functions meet my needs. With this tool, you can:

• Capture both screens and cursor movement.
• Capture audio with a built-in tool that lets you dictate the size and quality of audio files on the fly at export time.
• Control sizing and palettes easily with context-sensitive toolbars.
• Set timing, demo controls, and modes (interactive or kiosk, for example) on the fly at compile time.
• Import images from multiple formats (SVG 1.0, PNG, GIF, JPEG, BMP, TIFF (6.0 - G3, G4), Flash pix).
• Export images to local and network printers and to these formats: PDF, JPG, TIFF, BMP, and PNG.
• Display your completed show me with the Macromedia Flash® driver program.
• Use an HTML file created by the compile process to launch your show me.
• Develop your show me on many different platforms
• Run your show me on any systems that support Flash without change. Compiled demos are platform independent.

Designing Good Show-Me Help

Good multimedia has to meet basic design criteria. It should:

• Hold the user's interest.
• Be visually attractive.
• Provide real content, not just fluff.
• Be started—or stopped—by the user.
• Have redundant voice and text versions if audio is used.
• Not be the only form of help
• Be short. Somewhere between 30 seconds and 3 minutes seems optimal. At last year's STC, Ben Shneiderman recommended 3 minutes maximum. (1)
• Be launched in context. Most examples of show-me helps that I have seen use a camera icon.
• Use a visual style sheet and writing style sheet for consistent, professional-looking results.

This last point is particularly important. If multimedia deliverables are to move out of the realm of one-shot, short-term solutions into the realm of standard product deliverables, then more rigor in presentation have to be in evidence. Unless we want to unleash the kind of design chaos that desktop publishing systems produced twenty years ago (remember those newsletters with fifteen different fonts on a single page?), we have to have standards.

How do you do this? Unfortunately, simply creating and publishing standards and expecting users to obey them is usually ineffective. It is better to have the tool enforce consistency. For example, when you start a new screen project, ViewletBuilder lets you apply an HTML-style cascading style sheet to the demo presentation. In this first screen of the "New Project" wizard, you see the drop-down for defining the style sheet.

These style sheets might also be used to make textual characteristics, such as typeface and style, match the style of the traditional textual help information, further contributing to the seamlessness of the product information. Even though various tools create the parts of our embedded user assistance, the differences in media can be less jarring if you adhere to the same style guidelines for both.

Deciding to Use Show-Me Help

As an information planner, you make the decision to include show-me help. Here are some points to consider.

• Think about who your users are.
• Think about tasks in visual terms.
• Consider subsequent releases.
• Decide whether or not the show-me help will be translated.

Think about who your users are

Not all users benefit from, or will even use, show-me help. Kang, Plaisant, and Shneiderman found that different types of applications benefit from different styles of help. "For example," they write, "a complex CAD system to be used daily by an engineer will require elaborate tutorials and detailed online reference manuals. On the other hand, a web application used once or twice by a novice may benefit from a video demonstration." The challenge is finding the most effective method to help all users learn an interface. (10)

At the 2001 STC conference, Christina Downs and Anne Jackson related that while at Carnegie-Mellon University, they experimented with providing adaptive help that could "evaluate a user's skill level and select and display appropriate versions of the help." I experimented with this model in a tutorial that I created in 1984 when I created four unique paths through my tutorial based on whether my users were familiar with mainframes, with personal computers, with both, or with neither, and the technique was easy to implement and well-received. It kept my users from seeing parts of the tutorial that told them things they already knew. I agree with Downs and Jackson that while "most online help systems present a ‘one-size-fits-all' solution, we can do better with thoughtful design of our user assistance. (11)

There is also strong anecdotal evidence to suggest that novice users and casual users benefit the most from show-me helps. I like the classification of users that technical communicator Craig Marion used in his 2003 STC paper to describe the users of a Strohl System™ tool. One class of users, he explained, is made up of administrators who use the tool frequently and "typically learn to value the system's rich functionality. The other class is the majority of users. They use the system infrequently…and find themselves frustrated by the same rich functionality." (12)

I think the same can be said of users of show me's. The experts who use a system or program every day don't need show-me help, but those of us who use a system or a program or even a function within a program infrequently can benefit from embedded show-me helps. Thus I would say the users who benefit most from show-me helps are novice users and casual users.

You are a novice only once, but you are a casual of the obscure but rich functionality of a product forever. I have read that 80% of users access only 20% of the function of a product. But below that 20% is another perhaps 40% of function that many users access but access very infrequently. These are the tasks that send us searching through documentation, and these, I think, are the best candidates for show-me helps. A Cooper U course I took a couple of years ago said to design for "the perpetual intermediate." (13) Casual users are perpetual intermediates using obscure functions.

Think about tasks in visual terms

What kinds of topics lend themselves to show-me helps? A Tivoli colleague (14) summarized topics that are good candidates for show me's in this way:

• Topics that are difficult to document in words. Some tasks can be simple to do, but are difficult to describe.
• Topics that heavily involve visual design, such as Drag and Drop artifacts, Free Layout, and Visual editors.
• Topics that we know users have trouble completing. This customer feedback to identify these topics can be gathered through analysis of support document or user testing.

Consider subsequent releases

When I wrote earlier in this paper about the negative aspects of having technical writers create multimedia deliverables, I emphasized that adding show-me help is a commitment. Once you provide something to customers, it is hard to take it away, even if the resources for a project drop.

Do you need to consider translation?

As noted earlier, planning for translation is also an issue. In IBM, our translation centers support a stable and finite set of tools. Writers creating required product deliverables must use a supported tool or negotiate a special use arrangement.

When multimedia was the province of marketing, then these deliverables were considered extra and translated only if a certain country requested the translation and agreed to pay for it.

But with these deliverables as part of the "official" product deliverables and created by IBM information developers who usually work with translation-center sanctioned tools, there are different expectations.

Some products have taken the explanations for their multimedia presentations directly from the text of the online help, so that even though the show me was not translated, the same words are found elsewhere in the help. There is also some work under way to have ViewletBuilder added to the list of translation-center approved tools.

Some translation centers do not recapture the screens in our books, but instead translate only the explanation of them. Using this model, if the text for a show-me help could be exported from the tool, called "on the fly" through the use of variables, and re-imported from an external text or XML-format file, then a show-me help could meet current translation guidelines. I do not know of a multimedia tool with this capability. ViewletBuilder lets you export text to a file. This makes it easier to edit and review. But this text currently cannot be re-imported.

Use and Reuse

Embedding show-me helps inside of user assistance is not the only exposure that these highly-useful animated information snippets can receive. In November 2004, IBM rolled out the IBM Education Assistant Web site.

The IBM Education Assistant Web site (http://www-306.ibm.com/software/info/education/assistant/) integrates narrated presentations, show-me demonstrations, tutorials, and resource links to help customers successfully use IBM software products. This site is an excellent example of how the assistance provided in show-me helps can have a life beyond its role inside a product's embedded user assistance.

An extension of this idea is to source the "latest and greatest" on the Web. In her 2000 article in Intercom, Cheryl Lockett Zubak pointed out the trend of moving help onto the Web where the Help button in a dialog box checks the Web for existence of help on the Web and displays this help if it is found (otherwise, it displays the local version). (14) This hybrid approach enables help developers to update show me's even after the product has shipped if the product code changes or the show me shipped with the product is confusing. Many products today (including ViewletBuilder) use this approach to providing the best, most up-to-date help.

Conclusion

The technology for creating effective show me help is now accessible to anyone. Technical writers are in the best position to create effective show me's. We have experience creating help systems, we have graphic experience, and we have the infrastructure support for creating, inspecting, distributing ,and maintaining customer information. Now that tools as simple to use a ViewletBuilder are available, conditions are right for technical writers to move into the delivery of this fun, very useful information delivery medium.

Notes

(1) Shneiderman, Ben, "Opening Session," Society for Technical Communications National Conference, May 2004.

(2) Barron, Rick, Steve Hix, Paul Lorence, and Jenny R. Redfern, "Creating Multimedia Hardware Procedures with ShowMe™ How, Proceedings of the Society for Technical Communications, 1999.

(3) Palmiter, S, J. Elkerton, and P. Baggett, "Animated demonstrations versus written instructions for learning procedural tasks: A preliminary investigation," International Journal of Man-Machine Studies, 34, 687-701, as quoted in Kang, Plaisant, and Shneiderman.

(4) Bradford, Annette, "Conceptual Differences between Online Information and the Printed Page," Technical Communication, 3rd Quarter, 1984, pp. 13-16.

(5) Bradford, Annette, "Writing Training Information: A Comparison of Books, Online Tutorials, and Videotape," Journal of Technical Writing and Communication, V.17 No. 2 (1987), pp.115-127.

(6) Bradford, Annette, "Presenting Information on a Computer Screen: A Decision Process for the Information Planner," The Human Factors Society Bulletin V.30, No. 11 (November 1987), pp.1-3.

(7) Bradford, Annette, "A Planning Process for Online Information," Effective Documentation: What We Have Learned from Research, ed. Steve Dohney-Farina (Cambridge, MA: Massachusetts Institute of Technology Press, 1988).

(8) Ames, Andrea L., "An Introduction to Embedded Assistance," Proceedings of the Society for Technical Communications, 2002.

(9) Knabe, K., "Apple guide: a case study in user-aided design of online help," Conference Companion on Human Factors in Computer Systems, May 1995, as quoted in Kang, Plaisant, and Shneiderman.

(10) Kang, Hyunmo, Catherine Plaisant, and Ben Shneiderman, "New Approaches to Help user Get Started with Visual Interface: Multi-Layered Interfaces and Integrated Initial Guidance," found at http://www.ils.unc.edu/govstat/papers/dg-kang-final.pdf.

(11) Downs, Christina M. and Anne F. Jackson, "Interactive Help: Adapting Content for Multiple Users," Proceedings of the Society for Technical Communications, 2001.

(12) Marion, Craig, "From Online Help to Integrated User Assistance: One Company's Journey Beyond the Online Help Paradigm," Proceedings of the Society for Technical Communications, 2003.

(13) Cooper U, "Interaction Design Practicum," December 9-13, 2002, Durham, NC: www.cooper.com.

(14) I am indebted to Tivoli colleague Elizabeth Poggemeyer for these recommendations.

(15) Zubek, Cheryl Lockett, "Trends in HTML-Based Help Design," Intercom, January 2000, pp. 15-19.

Used with permission from the Society for Technical Communication, Arlington, VA U.S.A.