Like the final presentation, the goal of the final paper is to communicate the intellectual merit and broad impact significance of your project. The introduction begins with a problem statement that articulates what your team has worked on, and why. Next, address research contributions in terms of design, intellectual merit, implications, and broad impact. The expected length is 8 - 10 pages in the appropriate conference format (see below).
The research paper is an argument for its contributions. Research contributions should be reiterated and further explained in detail throughout the paper. Create a coherent story that connects the dots of the ideas presented on the first page. The first page is most crucial. It must sufficiently interest the reader, or they will stop reading!
Prepare your paper for an appropriate research conference venue. We've talked about the following:
- T-Cards: ACM Designing Interactive Systems (DIS)
- Disaster City AR Tour: ACM Ubicomp
- Spring Game: SIGGRAPH Games
- Trans-Surface Photo Sharing: ACM Tabletops and Interactive Surfaces
- Installation Art: ACM Multimedia (Interactive Art Track)
Mine your research notebooks for material. Return to your proposal. Revise the framing you set out with to match your experience. Restate the orientation that shines through intact. Derive new goals and results that have emerged through your design process.
Writing a research paper is an iterative process. Think about the research contributions at each iteration. Make changes across the entire document as necessary (including the title). Consistency of language is key!
Figures and Captions: Illustrate your paper clearly. Use labeled diagrams and flow charts to depict the major components and flow of control in your system's architecture and user experience. Use photos and other compelling visuals to emphasize critical aspects of your project. Write clear captions for each figure. Each caption must be self-explanatory, summarizing important aspect(s) of your research. Each figure, with its caption, should stand on its own. In addition, each figure must be referenced clearly in the text (with its Figure number), in context. Make sure that figures appear on the same page where they are referenced or, at the very least, on an earlier page (do not put figures on pages after they are referenced).
Title: The title should be catchy. It should crystallize the research contribution into a succinct nugget, motivating interest.
Abstract: Since it is an overview of the paper and contribution as a whole, the abstract is often written last. The maximum length is 2 paragraphs.
Introduction: Begin the Introduction with a contextualized problem statement. What conceptual space did you set out to work in? What problem(s) did you set out to solve? Why? How? What is the significance? What conditions and needs motivate the work? Go back to the structure for the first paragraph of the proposal to understand how to organize your Introduction.
Follow this with an overview of your approach, and an overview of the paper. Present the objective of your research, and your principal hypothesis. Also, address broad impact in this section.
Background: Identify the relevant prior work -- ingredients, and precedents. Ingredients are pre-existing building blocks, like hardware, algorithms, protocols, documenst, and API specifications, which you are using to build your work. Precedents are examples of previous projects that are similar to yours. When you are reading the precedent papers, look for models of how to structure your own paper, as well as for relevant ideas. Clearly state how each citation contributes to what you are doing, and how your work differs from (builds on) it.
Design: What has motivated your design? How has your iterative design process proceeded? What design did you arrive at? Include aspects of hardware, software, interaction, and experience. Describe the created system, its components, the flow of control between components, and the flows that involve users. How has your design evolved from initial conception, through prototypes and user studies? What options have you considered? How have you selected from among design choices? What has been the role of data collected in studies? How has your concept guided the process? How has your concept evolved through the process?
Evaluation: Define your study apparatus and design. Present data clearly to provide evidence for your argument and establish the validity of your hypotheses. Use quantitative and qualitative data as appropriate. Qualitative evidence, such as quotes and stories about users' experiences, can round out the validation of quantitative data.
Analyze and present the data clearly and specifically, highlighting your significant results. Develop visualizations to present the most important aspects of the data (bar graphs, trend lines, ...). Make sure that the scales of the axises are appropriate. Choose appropriate formats: a conversation between users should not be written as a paragraph, but as a series of sentences on separate lines. Chart style should fit the data and be free of extraneous illustration.
Discussion / Implications for Design: This classic payoff for research is a distillation of the intellectual merit. Generalize what you have learned into principles that can inform others. These principles can emerge from formative and summative aspects of the design process and data. The implications for design are ultimate results of your work, the ones that can help other researchers and developers. Refine significant methods and techniques. How have you solved the problem? What can be learned from the solution(s)? What new understanding has your project developed? How?
Conclusions and Future Work: Restate the results of your work. What field(s) does the research impact? How have you combined disciplinary perspectives and methodologies in an innovative way? How does the project advance knowledge and understanding within its own field or across different fields? What new knowledge have you created? What new understanding have you developed? How does the project suggest and explore creative and original concepts? Recapitulate your contributions. Develop their significance. Let us know what your next steps would / will be.
References: Make sure to cite all relevant publications here. Use the proper formatting, as per the conference proceedings template. Order your references by author's last name. Be consistent and thorough!
You are encouraged to bring your writing to The Writing Center while you are working on it, before you submit. Here are some other resources about how to write a research paper:
- Saul Greenberg, U Calgrary, How to Write a Research Paper and Thesis.
- Bill Griswold, How to Read and Evaluate Technical Papers.
- Alka Agrawal, Science, The All-Important Research Talk: Learning How to Do It Better.
- Harvard Writing Project, Strategies for Essay Writing and Student's Guides to Writing.
- Harvard Writing Project, A Student�s Guide to Writing in the Life Sciences.
- William Strunk, The Elements of Style.
Full Semester DVD + Paper: Put together a DVD with all of your semester's primary deliverables: proposal, plan, research notebook, video (uncompressed and compressed]), and final paper. Include documentation of your lightweight prototype if it is not too difficult. For the research notebook entries, make a folder for each member of your group. In addition to the DVD, turn in a print out of your paper. Also, please post your video some place like YouTube, and send the link to the class group.
The co-evaluation rubric is here. Fill in entries for yourself and all members of your team.