Your project proposal begins with a statement of concept that expresses a big picture sense of what you wish to accomplish and why. Next, a contribution and benefits statement positions the proposed work in its field and in society. These two can be connected to form the introduction. A prior work section addresses enumerates the basis of the project with needs, resources, and precedents. An evaluation plan says how we will know if the project is successful. A project plan says what will be done when by whom.
Also, briefly state what conference or journal you will submit to. Possibilities include CHI, Pervasive, DUX (Designing the User Experience), Advanced Visual Interfaces, ACM Multimedia (Art Program), IUI (Intelligent User Interfaces), InfoVis.
A good idea for how to develop a proposal in practice is
- Write concept and contribution and benefits.
- Write development portion of project plan.
- Write evaluation plan, and evaluation portion of project plan.
- Now write the prior work section. Make sure it addresses how you will develop and evaluate.
- Write the abstract.
This deliverable (and all others in this final project cycle) must be submitted in ACM Conference Format. Note: this is 2 columns, 9-point type, not large margins. These pages are longer than you might think of a page as being.
Statement of Concept [1/2 page]
So, what are you going to make? Why do you want to make it? What will it reflect? What is the project's function and social role? What processes (e.g., person, social, technological) are important. Why? Develop a 1/2 page statement of concept. In this section, make sure to convince the reader that project is innovative and beneficial. Make the contribution clear. Do this without using any language that sounds like a sales pitch!
Contribution and Benefits[1/2 page]
What good is your invention? What needs will it meet? What problems does it solve? Who will benefit? How? Why would anyone care about it? This is related to the NSF categories of "Broader Impact," and "Intellectual Merit."
Prior Work Analysis [1-2 pages]
Write a thorough and incisive description of the relevant prior work. The prior work collection is an essential component of research and invention. It involves collecting published materials that are relevant to the innovation at hand. There are three kinds of prior work: needs, resources, and precedents. Needs are preexisting conditions that drive the process of innovation. They motivate the relevance of the innovation, answering the question, "Why is this important?" These include stories and statistics about citizen behavior, interview data, and projections about future conditions. Resources are ingredients and raw materials that will be used for constructing the new invention. These include enabling technologies, design methods, processes, and materials. Precedents are points of departure. They are prior methods, systems, and services that are similar to the innovation at hand. Enumeration of precedents, with comparative analysis, serves as a basis for differentiating the new invention from what has been done before.
When you cite prior research make sure to be clear and explicit in describing what in particular is relevant about that work, and how you will apply it in your project.
This is a hybrid scenario, so include relevant research from HCI, multimedia, hypertext, cognitive psychology, graphics, etc (the sciences), and any relevant design and art citations. Use Google, Google Scholar, the ACM Digital Library, CiteSeer, Leonardo Electronic Almanac, Rhizome and TAMU LibCat, as well as the papers already on our reading list. Pay special attention to the last several years of proceedings form whatever conference you plan to submit to. When you find a relevant article, use it to find others through citation chaining. Consider commerical and freelance applications, as well as academic papers. Create a bibliography, with the entries alphabetized and formatted properly.
Evaluation Plan [1 page]
How will we know if your project is sucessful? The goal of evaluation is to establish (prove!?) that you have accomplished what you intended to. Design experiments that will establish this. After you have built your project, you will conduct these experiments later in the project cycle.
We have discussed the fact that many people (especially those who call themselves scientists) strongly associate research with including evaluation processes with clear metrics. How do we know if we've succeeded? While this is not the nature of arts paradigms, which typically depend on intepretive evaluations by experts, to produce creative experience as research, I believe evaluation is necessary. How can it be formulated creatively to be an interesting source of relevant data, rather than deadly dull?
Drawing from approaches we've seen from predecessors such as Smith, Czsiksentmihalyi, Bodker, and others, formulate an evaluation plan.
Project Plan [1 page]
How will you develop the project? Consider both the prototype time period, and the longer term. What software and hardware technologies are involved? What algorithmic methods? What functional modules do you envision? How will they interconnect? What is the system's architecture? What needs to be developed to conduct the evaluation?
Make sure to factor in iterative design. Plan to make lighter-weight prototypes earlier, and more functional ones later. What will the role of formative evaluation be? How will your project evolve? Your plan must consider and account for this.
What will be done by whom, when? What roles is each person playing in the project? What are the milestones and deliverables that each person will be responsible for? When? Include something like a GANT chart. Be very specific in defining each step.