Usability professionals in large software development organizations must either work within the context of an existing software development lifecycle or strive to change the lifecycle to incorporate user-centered methods. The goal of this workshop is to identify best practices for our industry regarding incorporating user-centered methods into software development lifecycles.
A workshop is a closed session. Admission to a workshop requires an approved position paper from you addressing the issues suggested by the coordinator(s). Please send your position paper (which should be roughly 1 to 3 pages) to Lisa Battle, lisa.battle@ssa.gov. Position papers received by April 25 will be accepted or rejected by April 30, in time for you to register before the early registration deadline on May 2. Position papers received by May 24 will be accepted or rejected by May 29, in time for the May 31 registration discount. Papers received after May 24 will be evaluated at the facilitator's discretion.
Participants will be selected based on the quality of submitted position papers containing the following pieces of information:
1. Education, Work Experience, and Interest in Usability / HCI
We have masters’ degrees in cognitive psychology/human factors, human performance technology, and management information systems. Our work focuses on user-experience design, and includes a wide range of experience in implementing user-centered approaches in a wide variety of commercial and government environments. In our current consulting position, we have been responsible for facilitating a transition to a user-centered design focus within a large government agency, prompting a change in the role of the agency’s Usability Center as well as the integration of user-centered design methods within the systems development lifecycle.
2. Description of specific experience with integrating UCD methods into traditional software development lifecycles
We have experience in working in a large software development organization, and helping them to transition from a traditional development approach to a user-centered approach to software development. The existing software development process was based on the waterfall model, which is not iterative or collaborative. Within this organization, we have become involved in shaping the development of two new, more iterative and collaborative lifecycles for software development, as well as leading user-centered design activities for project teams in the context of a traditional software development lifecycle.
We were invited to participate in the establishment of a new software development lifecycle for Internet projects, called the I-PLC. In this new lifecycle, user-centered design became a visible part of the analysis phase for all Internet applications. We established user-centered design as a distinct activity within the analysis phase of the lifecycle, which preceded design and development activities. However, in our more recent work providing input to the standard project lifecycle (SPLC), which is used for all core business applications, we have taken a different approach. Instead of creating a distinct step for user-centered design, we have incorporated user-centered activities within several traditional steps, such as business process analysis and requirements documentation. This approach is in response to the need to more fully integrate user-centered design so that it is not viewed as an additional and separate process.
In our daily work with project teams, we facilitate the use of the lifecycles by leading user-centered design activities. We encourage collaboration among team members from different disciplines, coordinating our activities and deliverables to meet the needs of the project. From this perspective, we have wrestled with resistance to user-centered methods and frequent overlaps between the roles of requirements analysts, user-centered designers, and technical team members.
Through some of our work with technical teams, we have been exposed to the Rational Unified Process (RUP) and the Unified Modeling Language (UML), and have become interested these new approaches to software development. They appear to offer an opportunity to integrate with lifecycles that encourage and accept collaboration and iteration. We have begun to establish pilot projects to allow us to work within this type of lifecycle and see how user-centered methods and outcomes fit into this framework.
3. Overview of the lifecycle you are working with and how you advocate integrating UCD
The IPLC consists of four main phases: planning and analysis, construction, implementation, and maintenance. In the planning and analysis phase, there is a step “Conduct User-Centered Design” which leads to the creation of a paper prototype and a scope agreement for the current release. There is also a step that includes business process analysis, quality assurance, and usability, which occurs during planning and analysis but also on an ongoing basis to provide input to future projects and to create a repository of reusable artifacts such as personas, scenarios, and design patterns. A major disadvantage of this lifecycle is that user-centered design was never fully integrated beyond the planning and analysis phase. There is no opportunity for user-centered designers to review or influence decisions once the project goes into construction. Another disadvantage, which became a strategic challenge for our group, is that this lifecycle has contributed to the perception that user-centered design is a separate step. Advantages of this lifecycle are that it gives user-centered designers control of an important step early in the planning and analysis phase, which allows us to identify critical design features and lead design activities. The attached diagram shows the IPLC and how user-centered design fits into it.
The new SPLC is still being defined, but it will likely have the same high-level phases. In the planning and analysis phase, there are steps for starting the project, conducting business process analysis, conducting user analysis and design, and documenting high-level requirements. There is no separate step for user-centered design. Instead, user-centered activities are incorporated as part of the steps that occur in the planning and analysis phase. We are trying to structure the outcomes of the user-centered activities to feed into standard deliverables such as the high-level requirements document. We are working to influence the construction phase as well by establishing a change control process that will keep us involved throughout development when adjustments are made to the original high-level requirements and design. Our goal is to influence the standard project lifecycle so that user-centered methods are fully integratedso that they are not thought of as a separate step, but simply as the way software is developed.
4. Obstacles/challenges encountered and strategies that have been successful in dealing with them
Obstacles and challenges:
· Lack of recognition that old processes need to change in order to benefit fully from user-centered design methods. The organization sees user-centered design as valuable and expects it to be used in all high-priority projects. However, they expect it to be done without recognizing that the traditional software development processes have to change in order to benefit from the methods. In software development processes, many decisions that should be influenced by the outcomes of user-centered design are made very early, before user-centered design occurs. Applying user-centered methods under these circumstances significantly limits the value that the methods can provide and prohibits the full integration of the methods.
· The perception that user-centered design is an additional step that lengthens the software development lifecycle. User-centered design is viewed as a separate and additional step in the life cycle. As long as it is viewed as a separate step, it will appear to lengthen the timeframes.
· The relationship between user-centered design activities and traditional requirements definition is problematic. Requirements are traditionally formalized and signed off before design activities begin. This is contrary to user-centered methods, in which information is gathered iteratively, validated and refined through user testing of design prototypes. The documentation of requirements should occur as one outcome of user-centered design activities, after the prototypes have been refined.
· Handoffs between roles may cause a completed interface design to be altered during development. Because our activities are not integrated beyond the planning and analysis phase, decisions are made during development without involving the original design team.
Strategies for dealing with obstacles and challenges:
We are addressing these challenges by focusing on integration at three different levels in the organization.
· At the project team level, we are working closely with requirements analysts and developers to ensure that the outcomes of user-centered design activities are merged into their deliverables. This addresses the perception that there are additional activities involved in user-centered design, and helps to change the traditional requirements definition activities.
· At the level of the systems organization, we are working closely with the process improvement team to document a new formalized standard project lifecycle that is user-centered and to integrate our methods so that they are not viewed as separate. We are providing instructions and templates for user-centered activities which will be provided to project teams as part of the standard lifecycle. We are also working to incorporate change control during development so that user-centered designers will be included in decision-making.
· At the executive level, we are looking for a champion who can assist in establishing Agency-wide support and commitment to the changes necessary.
5. What you would like to accomplish in the workshop
We would like to share our experiences with others who are also trying to integrate user-centered design. We hope to identify best practices in our industry for integrating our methods with software development lifecycles, with the intention of progressing towards a time when user-centered methods are the way that all software is designed.
The facilitators will create a list of best practices for successful integration of user-centered design into the software development lifecycle. If a preferred lifecycle is identified, the facilitators will illustrate it. If desired, the presenters can create a rudimentary poster for display.
The facilitators will publish the results as an article in User Experience.
The facilitators will create a discussion group and create a list of topics for follow-on workshops.
We have masters’ degrees in cognitive psychology/human factors, human performance technology, and management information systems. Our work focuses on user-experience design, and includes a wide range of experience in implementing user-centered approaches in a wide variety of commercial and government environments. In our current consulting position, we have been responsible for facilitating a transition to a user-centered design focus within a large government agency, prompting a change in the role of the agency’s Usability Center as well as the integration of user-centered design methods within the systems development lifecycle.