Employees are people too: improving web-based user experience in the workplace

Andrew Sweany, Usability Consultant, Information Technology Services [HREF 1], Monash University [HREF 2], Clayton, Victoria, 3800, Australia.
Email: andrew.sweany@its.monash.edu.au

Abstract

The main purpose of Information Technology (IT) in the workplace is to empower the enterprise’s most important resources – employees. Over the years, web technologies have undoubtedly improved operational efficiency across all types of businesses. A consistent challenge for IT organizations is how to deploy and maintain valuable solutions in a cost-effective manner, leading to the common ‘build or buy’ decision. To address this challenge many IT departments are increasingly purchasing 3rd party applications to minimize or eliminate custom design and development.

The risk of this approach is that decisions regarding which capabilities to deploy and how to deploy them will be based too heavily on technical factors and high-level business need without fully understanding the resulting impact(s) to users. IT departments can technically deliver a capability flawlessly, but if the user’s experience with the solution is poor, benefits will at best be reduced and in the worst case, offset entirely. At the project level, a variety of user centered design (UCD) activities should be performed to allow user considerations to be balanced with technical and business factors.

At the design level, this paper reviews challenges unique to web-based applications and ways to avoid these becoming user experience issues for employees. As publicly facing web applications continue to improve, IT departments will be pressured to provide tools with a comparable level of user experience to employees.

Introduction

Technology companies consider a wide variety of factors and tradeoffs during product design and development. In the conceptual stage, two critical factors that product teams first consider are whether a product is technically feasible and the business case for how it can be marketed. A third category that consumers value (but that product design teams may overlook) is that products are useful, usable, effective, and aesthetically pleasing. As products become more technically complex and the number of competing products increases, the differentiating factor for consumers has increasingly become ease of use. Quite simply, people prefer using products that are easy to use and that allow them to complete activities efficiently
[1, 2].

As little as 10 years ago, the most sophisticated computing technology that people interacted with was often found in the workplace. In fact, for many this was their only exposure to computing technology. However, the explosion of the internet and the rise of the home computer market drastically changed this model. The 1990s was a time when many consumers first began to utilize computers outside of the workplace.

As people began using computers at home companies began to realize that software needed to be simple enough that it could be installed, used and maintained by people of varying skill levels. Nowhere was this more important than the web – after all, in most cases a consumer could evaluate a website’s offerings free of charge and with little time invested. If a person found any aspect of a site unsatisfactory, they could simply move on to any number of competing sites with similar offerings. Because of this, usability has become a differentiating factor on the web, and web usability has, as a whole, slowly but steadily improved [3].

Over time, websites transitioned from being static pages to tools that people directly interacted with to accomplish certain tasks. Before long, people were using the web for a wide range of activities; from sharing video clips of the grandkids to managing their investments. The term ‘web application’ is most generally used to describe the sites that enable interactivity that in the past only existed within the realm of the desktop computer. Although web applications originally borrowed design approaches from desktop applications, the direction of influence is now reversing. As people became more accustomed to web interaction concepts, software development teams have begun to incorporate these concepts into desktop applications. One example of a web concept that has found its way into non-web software is the ability to easily move back and forth across screens, similar to the ‘Back’ and ‘Forward’ buttons within a browser. These and other design trends represent the blending of the traditionally distinct worlds of websites and desktop applications [4].

Other more general trends in web technology, such as the proliferation of higher speed and wireless internet capabilities, continue to change and shape user expectations. As people increasingly have the ability to remain connected outside of the office, both at home and during travel, they have increasingly come to expect that their workplace IT services be accessible from any location where connectivity to the web is available. In a similar manner to the blending of web and desktop applications, the ubiquitous availability of internet access is causing a merging of personal and work related IT use. Therefore, IT departments are now not only challenged with providing easier to use solutions but ones that can be accessed through any available web connection, thus raising new concerns around stability, security, bandwidth required by the application, etc.

Everyday technology surpasses workplace IT

People do not magically alter their expectations of technology when they walk through the office door. Increasingly positive experiences with web-based technologies have helped shape the expectations that users have of software in general. There is ample research demonstrating the correlation between expectations of quality and/or usability and user satisfaction. For example, someone expecting an easy task that turns out to be difficult would be less satisfied than expecting a difficult task that is indeed difficult [5][6]. Stated more plainly, if you expect something to be easy and it turns out to be difficult you will be frustrated. This makes perfect sense; after all, it has become mundane to search the entire Internet on almost any topic and retrieve relevant results but disheartening to search the company intranet and be unable to find a holiday calendar.

While it is generally believed that the use of IT-provided applications is not voluntary, these mandates are not always effective at forcing employees to utilize capabilities that do not meet their needs. If IT capabilities do not allow employees to complete activities quickly and meet their work goals effectively, a wide array of negative effects can result. At the most rudimentary level, users will simply not be able to work as effectively as possible (or, is some extreme cases, at all), reducing corporate productivity. In cases where new applications are truly disruptive, users may take active steps to maintain their productivity through developing local, non-IT solutions or alternative procedures that bypass the new application entirely. These situations breed a host of potential issues ranging from poor data quality to compromised corporate security. A more subtle but real consequence of poorly implemented IT in the workplace is frustrated employees. Given the number of hours in a day that some employees spend using IT services, this result should not be taken lightly. On the positive side, by providing capabilities that give employees more time to problem solve and meet objectives in innovative ways, the employees’ potential to contribute to the company’s success can be fully realized. For example, by providing capabilities that minimize manual data entry and automate analytics, sales personnel will have more time to establish and foster positive relationships with customers.

Meeting the IT needs of employees

As IT departments are increasingly responsible for providing and maintaining the capabilities that enable employees to compete in rapidly changing business environments, many are concurrently faced with the reality of flat or shrinking budgets. In addition to budget constraints, many other factors, such as changes to business strategy or increasing complexity in the business environment continue to accelerate the need for new IT solutions. Given the number of competing factors, it is easy for management to overlook or downplay user considerations in the decision-making process. To ensure solutions will ultimately meet end user needs, a user centered design (UCD) perspective is critical. A UCD approach means that the needs of end users are incorporated into the design, development and/or deployment of an IT solution. In other words, project teams must frequently ask, “how will this decision affect our users [7]?”

UCD practitioners have utilized a “three-circle" model to convey the importance of evenly balancing user considerations with business and technical factors [8]. As shown in Figure 1, the model indicates that successful products balance business, usage, and technology. The model has proved to be a powerful way to represent the importance of the end user in the traditionally tech-centric world of IT.

Figure 1: The Three-Circle model

At a more applied level, typical UCD activities consist of working with end users throughout the design lifecycle to evaluate and test the User Interface (UI) of a capability prior to deployment [9]. While these activities have been successful in improving tool usability, it has become more and more apparent that many factors beyond usability are important to ensure the successful introduction of new technology into the workplace. For example, if installing an application requires the user to perform several manual steps, the probability of an unsuccessful install increases, potentially hindering employee productivity and resulting in technical support calls. In addition, without adequate self-help functionality, employees may experience ongoing difficulties with completing activities despite interacting with a UI that meets accepted design principles. Therefore, although direct interaction with the capability may be efficient, other activities associated with the product, such as the initial install and ongoing help mechanisms, can lead to an overall negative impression of the solution.

To address these challenges, UCD practitioners have begun utilizing a model that incorporates all touch points a user has with an IT solution. As shown in Figure 2, this comprehensive approach incorporates the entire User Experience (UX) with IT products, including product awareness, acquisition, use, support, and end of life [10]. By understanding and optimizing each of the components of UX, IT departments are better enabled to deploy capabilities that immediately and continually provide value to employees.

Figure 2: The IT User Experience lifecycle

Why users are easily overlooked

Despite the number of user-centric methods available, most IT departments are still overwhelmingly technology focused. If user considerations are so critical, why are they so often overlooked? Some IT investment decisions will always be technology driven: some maintenance investments will always be necessary that are not directly visible to end users (upgrading network capacity or improving the stability of e-mail servers, for example). After all, when the e-mail server goes down, everyone notices, and the effect is visible and immediate. For most IT departments, measures such as ‘percentage of down time’ are the most critical and perhaps only indicators of success. Solely relying on these measures leads to a disproportionate focus on simply making technology available rather than providing technology that is easy to learn or that delivers the most value to employees.

Furthermore, the effects of deploying unusable business solutions, such as lost productivity or a high volume of tech support calls, more easily go unnoticed. Few enterprises have the type of metrics and measurement systems in place to capture these impacts and accurately convert them into financial terms. Although there may be loose references to training and customer support costs within most Total Cost of Ownership (TCO) equations, the estimates typically used do not come close to truly accounting for the productivity loss and frustration that occurs as a result of deploying solutions that do not meet user’s needs. Even more difficult to detect and quantify is the result of deploying a capability that doesn’t provide anticipated value and as a result is underutilized, leading to non-approved tools and procedures. In essence, productivity loss and other negative consequences of unusable or under utilized applications do not have the same ‘voice’ as technical or business concerns. However, the undeniable truth is that the value (and potential return) of investing in a new capability can only be realized if end users adopt and utilize a new capability to more effectively complete business objectives [11].

The risk of 3rd party solutions

As more and more solutions become commercially available, many IT departments have begun to minimize or eliminate the custom development of web applications. Although the initial benefits of purchasing a solution appear promising (quicker deployment, easier upgrades, etc.) it further increases the risk of downplaying users as the most critical component of successful IT deployments. For example, cost and schedule goals are two of the most critical success criteria for project teams and this will undoubtedly hold true in the future. However, one potential risk associated with focusing too heavily on hitting schedule and cost goals is the minimization of activities perceived to slow velocity or increase spending. On 3rd party solution deployments, the pressure to quickly deploy often overshadows activities such as usability testing and quality assurance. Project teams also assume that the vendor has completed these testing activities during the original development and further testing would therefore be redundant. Lastly, usability testing may be seen as unnecessary because of the assumption that a purchased application is entirely inflexible and must be accepted ‘as is.’ In reality, many-third party capabilities have a significant amount of configuration and deployment flexibility and thus require the same types of testing (usability, system performance, etc.) traditionally performed on more customized applications.

With 3rd party solutions, users can no longer provide ‘blue sky’ requirements, a process where all functionality requests are captured and later prioritized for development. Although this method has its own detriments, it can provide end users a sense of involvement in the development process, improving solution adoption rates and easing the overall transition effort. Third party solutions will likely be significantly different from the customized applications familiar to users. In addition, because the benefit of purchasing a solution result from avoiding or minimizing customized development, IT has a limited ability to subsequently design in enhancements if a capability does not meet users’ expectations. Therefore, due to the increased difficulty of making mid-course development corrections, user considerations become even more critical to the success of 3rd party solution deployments than they were with the more flexible ‘build from scratch’ approach.

Deploying purchased solutions often means that organizations and end users must be more willing to adapt to the capability vs. developing or customizing the capability to meet user wishes. In many cases, adapting to the architecture and functionality of a 3rd party solution requires a significant amount of business transformation. Most often this means changing existing business processes (who does what and when, how communications happen, how decisions are made, etc.) to match how a given capability architecture is designed or can be configured. Companies routinely underestimate the difficulty of managing these business transformations and their impact on individual workers. Even a well-planned change in a business process can disrupt the work routines of experienced employees, and it can have a significant short-term impact on effectiveness and productivity. Poorly planned changes can lead to the overall failure of the business transformation efforts, resulting in significant financial loss [12].

Case Study: Overlooking the user

There are various cases from the IT world where the implications of user impacts were not considered, degrading the full benefits of deploying a 3rd party capability. One such experience involved an enterprise-wide deployment of a 3rd party web application used by managers to post job openings and used by potential candidates to apply for jobs. Obviously, the ability to attract and retain talented employees is critical for any business – therefore, it is essential that potential employees can easily search for, locate, and apply for positions. In addition, this particular application was critically important to a highly paid user population with a low tolerance for time wasting activities i.e. hiring managers. Although this scenario would suggest that usability would be a critical factor for a successful deployment, most usability related activities were bypassed in an attempt to deploy the new solution as quickly as possible.

For this particular deployment a ‘zero configurations’ approach was utilized. In other words, when deploying the capability into production, all available functionality in the purchased system was activated by default, with no consideration of whether the functionality was useful to employees or not. It quickly became apparent that certain functionality in the new solution was redundant with other applications they had been traditionally used to complete the overall business process. For example, replying to a candidate by e-mail and scheduling a time to meet with had to be performed within the new system instead of their existing e-mail and calendaring tool. Mangers now had to use the new system to track job queries, replies, and whether a follow-up conversation with an applicant had been scheduled. Effectively this meant forcing users to deal with an additional inbox and calendar to manage the job openings they oversaw. This redundant functionality resulted in frustration and significant increases in the time it took to complete tasks.

Additional issues beyond interacting with the web application itself compounded the aggravation. Users found that online help resources were not useful in addressing questions, resulting in a considerable number of technical support calls. Subsequent survey results and user feedback pointed out that pre-deployment communications about what was changing and where to get training on these changes were not adequate. In summary, by not accounting for how the new capability differed from existing work practices, a host of issues ensued, affecting employees’ ability to complete activities across the corporation.

The project team, working with a user centered design expert, conducted user research to identify root causes and solutions to the variety of problems that users of new application were experiencing. The first user research activity included several rounds of usability tests to observe employees attempting to complete tasks with the new system. As reported by employees, the new system forced users to perform sub-tasks in a manner and sequence that was vastly different from what they were accustomed to. The formal usability testing approach allowed the specific pain points to be captured and communicated back to the development team in a much more structured format than simply relying on subjective qualitative feedback from users.

With a better understanding of how users approached their tasks, the project team set out to build an optimized flow of screens that better matched the managers’ natural workflow. The simplest yet effective changes were configuration only, meaning that they did not require any code customization. Rather, these changes simply consisted of turning on or off functionality and fields that employees needed or found useful for the hiring process. These changes alone dropped the average task completion times from 45 min. to 23.5 min. In addition, several code modifications were implemented at the user interface level to address critical usability issues, and those that were not addressed through modification were channeled back to the vendor to improve future releases. Unfortunately, the time and effort to implement these changes were significant, eroding the intended benefits of deploying the 3rd party solution.

On future deployments, the training team shifted their focus from “quantity” (e.g., percentage of users trained) to “quality” (effectiveness) to verify that end users were adequately prepared for upcoming changes. Lastly and most importantly, multiple parties were involved in analyzing and improving the overall user experience, including stakeholders, future end users, technical support personnel, training personnel, and user readiness representatives. Although the initial deployment provided some hard lessons, the result was positive as it helped to increase focus on the importance of user considerations when purchasing, configuring, and deploying 3rd party capabilities [13].

Avoiding user experience issues

After discussing the risk of overlooking user considerations, the natural question is how one avoids this risk in the first place. At a minimum, it is important to ensure that decision makers place an adequate focus on the end user by employing a user centered design perspective. In the end, it does no good to spend significant time and money to deploy a solution that employees cannot effectively use unless it undergoes a costly round of reworks. It is critical that IT departments work with stakeholders and end users of web based solutions to understand their expectations regarding usability. For example, in a university setting, students will likely have different expectations than director level staff, and their expectations will likely differ from administrative assistants. Do your users expect to be able to ‘walk and use’ the new system or do they accept that some training will be necessary? A small amount of effort to understand your users early in the design, develop and/or simply deploy lifecycle will greatly reduce the overall risk of the final solution coming up short.

Once users’ expectations are accurately understood, the next step is to ensure that the solution will meet these expectations. One way to ensure the end solution is ‘good enough’ is to establish criteria for usability and other aspects of quality (for example, system response time) and incorporate them into the purchase decision. This means that a passive demonstration of functionality by a vendor will not be sufficient. Instead of solely relying on a binary data point (i.e. the product includes a desired feature or it does not), it is more useful to create realistic goals that relate to your end users tasks (for example, on average, employees should be able to request leave within 2 min.). Therefore, it is important to analyze who will be impacted and how to determine the best way to optimize the user experience of each deployment.

If different 3rd party options are being explored, purchasers should ask vendors to provide the results of usability testing they performed during design and development. If a vendor is reluctant to provide such data it likely indicates that they have not adequately performed standard usability activities. Ultimately, this does not bode well for a company’s employees, as they will be the ones who struggle to use it. Even if a 3rd party vendor is willing to describe their usability activities, it is essential to perform additional usability testing within the context of your business. This is important because you and your 3rd party vendor have different priorities. Whereas a vendor is striving to provide a generic solution that can meet the requirements of many potential customers, any single client will have specific requirements that, in all likelihood, will not be fully met by the default solution. Therefore, usability testing can assist in deciding whether the solution is close enough to meet the needs of employees to be acceptable.

Making this decision requires that the usability vs. customization trade-off needs to be priced. For example, a web application has the office supplies order functionality you are interesting in but your group of pilot users complains that it is difficult to use. Usability tests conducted with this sample population of your employees indicates a 30% failure rate – that is, three out of every 10 employees will not be able to place an order correctly. An analysis by the deployment team suggests that it will take approximately 20 hours of coding and testing to address the issues. Armed with this type of data, stakeholders can now make an educated decision of whether the 20-hour investment is worthwhile. If the entire business has 30 employees, the investment is probably not warranted, and perhaps training targeted for this task is the answer. If your employee pool is 5000, the answer is likely to be quite different.

It is also important to usability test the final product prior to deployment, regardless of whether customization has occurred or not. Pre-deployment usability testing of the fully configured product, loaded with real data and used in the production environment, helps to identify and highlight issues that can be used to improve the overall user experience. As mentioned earlier, many considerations beyond tool usability are important when deploying a solution. These additional aspects of the overall user experience include training, communication, online help, and post deployment technical support. Usability testing results can be used to direct resources to the most valuable efforts. For example, usability testing may show that the order functionality is difficult to use, but technically it may be very difficult to redesign. If redesigning the solution is not feasible then it will be necessary to train users on how to complete the difficult task. At the very least, if a task is known to be difficult and this is communicated to employees, they will be better able to approach the challenge and their level of frustration will be reduced.

Common issues with web applications in the workplace

Many potential user experience issues are universal across all types of products, but at the implementation level there are issues that are rather unique to web applications. An even more specialized subset is web applications used within the workplace. In general, applications found within the workplaces are often less polished, both from a visual and usability perspective, and than their consumer facing equivalents. In addition, the nature of workplace tasks often means that web applications in the workplace are designed to be more data-entry intensive. Lastly, the simple fact that the workplace represents a smaller user population in a more technically controlled environment reduces the scope and severity of some potential usability issues that may exist for publicly facing applications. With all that being said, the following issues are those that continue to be troubling for employees using web applications in the workplace.

1. The ‘Back’ and ‘Forward’ buttons: users of the web have become accustomed to quickly moving back and forth between pages. Place an application within a browser rather than static pages and this behavior continues. This freedom of movement sets the stage for a user to lose data populated in a form, spreadsheet, etc. Provide a separate ‘Back’ link or button within the application itself that automatically saves any data entered on a page.
2. Annoying log-ins: employees can use many different web applications across the course of a day. If possible, tie applications together with a single sign-on or ensure applications are compatible with browser functionality that offers to remember user name, password, or both.
3. Time outs: for security reasons, web applications in the workplace often utilize a time out mechanism where a user is automatically logged off after a certain period of inactivity. The problem is that the workplace is full of distractions. Therefore, only use timeouts if it is truly necessary, provide ample warnings and enable auto-save to avoid data loss.
4. Restricted interaction: Many applications that were traditionally on the desktop are now web based. Unfortunately, when this transition is made, the ability to mimic common desktop interaction concepts, such as copy/paste, dynamic column resizing, and drag and drop may be limited or impossible. Explore different technologies, platforms, coding languages, etc. to enable the interaction types your users are familiar with and appreciate.
5. Rendering issues: Even within a single enterprise, there may be several different types and versions of browsers in use. Add to this a widening variety of display sizes and resolutions, and the ability to disable pop-ups, cookies and scripting, and it becomes difficult to know whether your users are seeing the user interface of the application the way it was intended to appear. Investigate the hardware and software that employees are likely using, test to ensure the web application appears correctly on a number of these different configurations prior to deployment, and observe users interacting with the application in the real world to verify it renders correctly.
6. Page refresh: One potential issue with pages refreshing is the difficulty of knowing what has actually been ‘refreshed.’ If users are working with a screen full of data, it is difficult to notice small but potentially critical changes to data, even if the user initiated the refresh after an action. Use high color contrast or movement to ensure that any changes are easily noticed. Better yet, use AJAX or similar web development techniques that provide real time responsiveness within the application.

Discussion

Beyond benefits to the individual and the enterprise, a more significant focus on delivering end user value has major implications for the IT industry as a whole. Just as functionality alone is becoming less of a driver of product sales [14], simply providing technology to employees will not provide optimal return on IT expenditures. In many cases, the technology now available to consumers meets or exceeds the capabilities provided by IT departments, both in terms of functionality and user experience. As people become more accustomed to products and services that deliver a good user experience, the greater the expectation that IT departments follow suit and do the same.

Younger people entering the workforce will not have the historical context of pre-IT days. They have grown up in a world where the internet, e-mail and web applications have always existed. From a software perspective, they are likely much more accustomed to smooth, real time interaction enabled with AJAX than clunky, inflexible desktop application created with C++. In regards to expectations, it means that they will likely be much less patient with frustrating IT in the workplace. Given the amount of time that employees spend using IT services, it is not far fetched to believe that the quality of these services does and will continue to considerably affect overall employee satisfaction.

As publicly facing web applications and web-based technologies continue to focus on improving user experience, employees of all ages will continue to alter their expectations of what ‘good’ IT means. A competitive advantage that IT organizations can deliver is greater business value through positive user experience. Moving forward, it will be increasingly insufficient for IT departments to solely focus on deploying technology more cheaply. IT organizations will maximize the business value they deliver by directly addressing and solving the issues their end users and business organization have in achieving their goals and objectives. Focusing on getting the optimal user experience designed into or provided by the IT solution will help drive a greater overall return on investment than can be achieved by focusing primarily on total cost of ownership.

In summary, all technology design and deployment efforts face constraints. Whether deploying a web based IT solution or marketing a new consumer product, project teams face the competing pressures of providing useful and usable technology while balancing time, cost, and other limitations. Regardless of the context, however, the importance of delivering end-user value to ensure adoption (and sales) remains consistent. Pre-selection and pre-deployment user experience activities allow decision makers to more wisely explore available sources and ultimately to deploy the standardized solution that most closely meets employees’ needs.

References

[1] Bias, R. G., and Mayhew, D. J, Eds., Cost-justifying Usability. Academic Press, Inc., 1994.

[2] Belz, S. M., "Development of a New Photographic Experience,” Ergonomics in Design, vol. 14, no. 4, Fall 2006.

[3] Nielsen, J. and Loranger, H., Prioritizing Web Usability. New Riders Publishing, 2006

[4] Fowler, S. and Stanwick, V., Web Application Design Handbook. Morgan Kaufman Publishing, 2004.

[5] Oliver, R.L. (1977). Effect of expectation and disconfirmation on postexposure product evaluations: An alternate interpretation. Journal of Applied Psychology, 62(4), (480-486).

[6] Rich, A., and McGee, M., (2004), Expected usability magnitude estimation, Proceedings of the Human Factors and Ergonomics Society Annual Meeting.

[7] Abras, C., Maloney-Krichmar, D. and Preece, J. (2004). User-centered design. In Bainbridge, W. (Ed.) Encyclopedia of Human-Computer Interaction. Thousand Oaks: Sage Publications.

[8] Simmons, E., “The Usage Model: Describing Product Usage during Design and Development,” IEEE Software, vol. 23, no. 3, pp. 34–41, May/Jun, 2006.

[9] Norman, D. A. and Draper, S. W., Eds., User Centered System Design, Erlbaum, Hillsdale, New Jersey, 1986.

[10] Sward, D., “Measuring the Business Value of Information Technology: Practical Strategies for IT and Business Managers,” Intel Press, ISBN: 0-9764832-7-0, 2006.

[11] Goodhue, D. L. and Thomson, R. L., “ Task Technology Fit and Individual Performance,” MIS Quarterly, pp. 213–236, 1995.

[12] Burlton, R.T., Business Process Management, Sams Publishing, Indianapolis, 2001.

[13] Sweany, A. and Gomez, M. “Bringing the Voice of Employees into IT Decision Making,” Intel Technology Journal, Volume 11, Issue 1, 2007.

[14] “Entertaining Research”, September, 2004, at http://www.hpl.hp.com/news/2004/july_sept/experience.html

Hypertext references

HREF1: http://www.its.monash.edu.au/
HREF2: http://www.monash.edu.au/

Copyright

Andrew Sweany © 2008. The authors assign to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The authors also grant a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers and for the document to be published on mirrors on the World Wide Web.