This page describes the requirements for the individual effort portion of the group project.

Part 1: Initial paper exploration

Each student will individually look through (at least) the last three years of each of the International Sympoisum on Wearable Computers (ISWC), the International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp), and the International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI). These proceedings can be found in the ACM Digital Library; ISWC and UbiComp are listed under the Publication tab here and MobileHCI is under the Publication tab here. If you have a favorite conference you’d like to add to this list, you may, but the papers you choose must match the theme of the class and no more than three may come from conferences other than those listed here.

When looking through these conferences, don’t read the papers (yet). Look instead for things that catch your eye; in particular, you can look at:

  • the session title (e.g., Wrist and hand interaction I);
  • the paper title (e.g., Nail+: sensing fingernail deformation to detect finger force touch interactions on rigid surfaces);
  • the abstract for papers; and/or
  • the video for papers (if there is a video, it can be found on an individual paper’s page).

From these three years of three conferences, you will choose nine papers (you can do one per conference/year, but don’t have to) that look, at first glance, interesting to possibly replicate or extend.

You will send me these nine papers via Slack, in a Post. Your post must clearly number the papers 1–9, and consist of:

  • the paper title;
  • the paper authors;
  • the conference the paper came from;
  • a link to the paper’s page on the ACM Digital Library;
  • 1–2 sentences about what you find interesting about the paper.

Note that there are different kinds of papers; in particular, you may notice “study” papers and “implementation” papers. To draw from my own publications, Quickdraw is a study paper, looking at how quickly people can access their mobile devices; Nenya, on the other hand, is primarily an implementation paper (with a study component), discussing how I created a ring to control mobile devices.

Part 2: Paper choice and summarization

Next you will download the PDFs for each your nine papers and skim the paper. You can read this brochure about efficiently reading papers, but the relevant points for skimming, pulled from that link and modified slightly, are:

  • Read the introduction. Pay attention to the contributions the authors are trying to make. Sometimes they explicitly state this.
  • Read the section headings.
  • Look at the images and any tables or graphs and read the captions.
  • Read the conclusions.

From these nine papers, you will:

The summary should be several paragraphs long, and contain the following parts:

  • the title, authors, conference, and link to the paper (the same format as in your nine paper list);
  • a clear explanation of what the goal of the authors was, including the contributions they were trying to make;
  • an explanation of what method(s) they used to make their contributions, including what (if any) implementation was involved;
  • what limitations, problems, or other issues there are with the research;
  • and how you might replicate or extend their work during the course of the semester in order to either address the limitations or to confirm the correctness of their research.

Example

Here’s an example of what such a summary might look like:

Quickdraw: the impact of mobility and on-body placement on device access time.
Daniel Ashbrook, James R Clawson, Kent Lyons, Thad E Starner, and Nirmal Patel. CHI ‘08: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, New York, New York, USA 2008 pp. 219-222.
http://portal.acm.org/citation.cfm?doid=1357054.1357092

The goal of the authors was to empirically determine how long it would take people to access devices that were placed at different points on the body. In particular, they investigated a device in the pocket, in a hip holster, and on the wrist.

The authors performed a study where they had people either stand still or walk around, and timed how long it would take participants to respond to an alarm on the device. They implemented software on a phone to run the study, including presenting an alarm sound, measuring response times, and detecting when the phone was removed from the pocket by using the device’s camera.

The research is at this point quite a few years old. Mobile phones have gotten much less bulky, screens have become capacitive rather than resistive, and we now have smartwatches. Hip holsters for mobile phones are less common than they were in 2007 when this study was run. All of these factors mean that the findings in this study may no longer reflect the access times for real devices today.

I propose to replicate this study in order to determine whether the measurements from this paper still apply today. Doing so will involve implementing similar software to that used in this paper, and designing and running a similar study. I propose to use a modern Android-based mobile phone and an Android Wear-based smartwatch to write the study software. Because hip holsters are less common, I propose to skip that part of the study.

Grading

The individual effort portion of your grade is worth 12%.

Part 1 is worth 5% of your final grade; the criteria for this part are:

  • You send me the correct number of papers from the correct conferences, as described above (1%).
  • You adhere to the format and provide all of the information requested for each of the papers (3%).
  • Your writing is clear and comprehensible without significant spelling or grammar errors (1%).

Part 2 is worth 7% of your final grade; the criteria for this part are:

  • You send me the correct number of papers from the correct conferences, as described above (1%).
  • You adhere to the format and provide all of the information requested for each of the papers (5%).
  • Your writing is clear and comprehensible without significant spelling or grammar errors (1%).