This is the information that was given to participants at the 2016 course on how to write CHI papers.
Please bring some examples from your own recent writing to this course. This can be a thesis abstract, some unpublished papers, or just something you have written recently.
|09:00-10:30||Lecture: Introduction to the Course (Read Interview with Carl Gutwin)|
|16:30-17:45||Writing doctors: Bring your own manuscripts and let’s dissect them.|
Introduction to the CHI PLAY Course
Before we begin, please have a read over the following materials:
- Susanne Bødker, Kasper Hornbæk, Antti Oulasvirta, and Stuart Reeves. 2016. Nine questions for HCI researchers in the making. interactions 23, 4 (June 2016), 58-61. DOI: 10.1145/2949686
- Antti Oulasvirta and Kasper Hornbæk. 2016. HCI Research as Problem-Solving. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI ’16). ACM, New York, NY, USA, 4956-4967. DOI: 10.1145/2858036.2858283
I would also like to reiterate the nine questions for HCI researchers in the making here:
- If you could address just one problem in 10 years, what would it be?
- Are you using your unique situation and resources to the fullest?
- What’s your HCI research genre?
- In one sentence, what is the contribution of your research?
- Is your approach right for your research topic?
- Why is your research interesting?
- Can you fail in trying to answer the research problem?
- Will your work open new possibilities of research?
- Why do you build/prototype?
These questions are really good starting points to give you bearings on your research direction.
I would also like to point you to the results from my questionnaire for CHI researchers about writing. Interestingly in the survey researchers rated the importance of the Introduction and Results section almost equally high (with the results coming out on top) and were not giving as much love to the Discussion section. In my interviews, however, the introduction and discussion were mentioned as important sections. The key ingredients for CHI research papers mentioned in the survey were:
- Clear framing
- Interesting topic
- Clear contribution
- Problem worth solving
- Good problem, motivated by the literature
- A novel and ambitious solution for the problem (e.g., in terms of system, evaluation, data collection)
- Convincing evaluation
- Sound methods
- Considering all relevant implications
- Well structured
- Appropriate language
- A discussion that allows the solution to be transferred to other problem instances and related to what we knew in advance
- Key ingredients are believable answers to a design question that is not obvious or a novel system.
- Clear contribution to the field, sufficient proof for valid claims, usage of a scientifically valid methodology (depending on the type of contribution)
- A clear description of the (usually applied) research problem
- Clearly articulated research question
- A clear description of related work
- A clear description of a valid method for finding the answer to the question
- A clear description of a valid approach to data analysis good discussion with some implications for the design for luck lots of luck in getting a good set of reviewers
- Being clear about the intellectual contribution to HCI research itself
- A clear contribution
- Well-executed user involvement
- Making sure you have enough users, not just testing with students
- Deep and broad referencing
- A well-written abstract
- Clear scope
- Acknowledging the subcommittee/audience you write for
Besides asking for key ingredients in CHI papers, I also asked survey respondents for the main piece of advice for aspiring CHI authors:
- Convince your AC and you have a chance. Reviewers don’t matter in the grand scheme of things. If your paper is even close to getting in in November, a strong rebuttal and an advocate on the AC are all you need.
- Budget a lot of time.
- Picking the most appropriate subcommittee makes a big difference.
- Pick a good problem; know the literature; start early; get feedback; discuss with earlier work; be bold.
- Don’t consider CHI as the only premier venue – in-depth specialized papers have a better place at the specialized SIGCHI conferences. Don’t focus on style, focus on correctness, scientific validity and on a contribution that you think changes or progresses the field significantly.
- Don’t submit 5+ papers a year, make your contribution count. CHI is not an outlet store, it is a scientific conference. (Sometimes, I’m also guilty of doing this)
- Iterate. Don’t leave writing until the last minute.
- Expect to fail (you can always send it to a journal – seriously).
- Make it a good one.
- Do not provide variations of the same thing.
- Choose and do so honestly a few wise reviewers.
The following exercise was done in class with 30 minutes time.
Analyze the Structure of this Abstract
Look at the following abstract and try to understand its structure. I have already outlined the main problem for you, can you find the solution and why it is important and what makes this a contribution to HCI?
Many sports video games contain elements such as running or throwing that are based on real-world physical activities, but the translation of these activities to game controllers means that the original physicality is lost. This results in games where players have limited opportunity to improve their physical skills, where there is little differentiation in people’s physical abilities, and where skills do not change over the course of a game. To explore ways of adding these elements back into sports games, we developed two games with small-scale physical controls for running and throwing — one game was a simple running race, and one was a team-based handball-style game called Jelly Polo. In two studies (three track-and-field tournaments for the running game, and a four-week league for Jelly Polo), we observed the effects of physical controls on gameplay. Our studies showed that the physical controls enabled substantial individual differences in running and passing skill, allowed people to increase their expertise over time, and led to fatigue-based changes in performance during a game. Physical controls increased the games’ challenge, complexity, and unpredictability, and dramatically improved player interest, expressiveness, and enjoyment. Our work shows that game designers should consider the idea of “exertion in the small” as a way to improve play experience in games based on physical activities.Mike Sheinin and Carl Gutwin. 2014. Exertion in the small: improving differentiation and expressiveness in sports games with physical controls. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’14). ACM, New York, NY, USA, 1845-1854. DOI: 10.1145/2556288.2557385
Analyze the Structure of this Introduction
Look at the Introduction from the same paper and try to understand its structure. I have already outlined the main problem for you, again can you find the solution and why it is important and what makes this a contribution to HCI? What parts of the introduction are confusing or misleading?
Many video games contain elements that are based on physical activities (e.g., running, throwing, jumping, or kicking), and sports games in particular are strongly based on physicality. Sports games have many complexities, and players can build up considerable amounts of expertise in them – but the nature of this expertise is usually very unlike that of the original sport. This is because games must translate a real-world physical activity to an action that can be carried out with an Xbox-style game controller, and in the translation, the physicality of the original activity is lost.For example, one main element of many real-world games is bodily movement such as running. Running in many sports video games is translated to a rate-controlled joystick action (i.e., press the stick to move the on-screen character) or a fixed-rate keyboard action (i.e., press and hold the WASD keys to move). This changes the activity of running from a repeated large-muscle action with substantial physical demands, to small fixed movements of the fingers and hand on a controller. In addition, this translation changes a complex multi-degree-of-freedom action to a simpler rate-controlled action where velocity is a function of the system rather than a function of player effort. A second example involves throwing skill. In the real world, passing is a precision skill that requires substantial practice, but in many sports games, passing is translated into an action that is at least partially controlled by the computer (e.g., direction and distance are automatically set or adjusted so that a pass will go to a teammate). These kinds of translations between physical activities and controller actions are present in almost all sports games (and many other avatar-based games as well). However, this approach presents three drawbacks for sports games: There is limited opportunity for expertise development. Although there are many ways for a player to increase their skill in a sports game, there is little opportunity for improvement of basic actions like running or throwing. In contrast, improvement in basic physical skills is a foundation for expertise in real-world sports.There is little differentiation between players in terms of basic actions like running and passing, and thus little opportunity to use these differences in the game. In contrast, success in real-world sports often revolves around individual differences (e.g., taking advantage of a mismatch with an opponent’s physical capabilities, or setting up a team to capitalize on individual strengths and minimize individual weaknesses).Third, the artificial simplicity of controller actions means that there is no change in a player’s physical capabilities over the course of a game. In real-world sports, effort-based factors such as fatigue clearly set apart better players and teams from weaker ones – e.g., many games are won and lost when the team with more endurance takes advantage of the other team’s fatigue. Overall, these drawbacks reduce the richness and realism of sports video games. Although some games can add other types of richness (e.g., difficulty levels for computer players, minigames such as fighting in a hockey game, or ‘manager modes’), the core play experience of a sports game is often limited by these problems. In this paper, we investigate the idea of adding physicality back into controller-based movement in order to add expressiveness and player differentiation back into sports video games. We maintained the basic play environment, with standard controllers and settings, but added two kinds of physical control: impulse-based movement, where each physical action on the controller only moves the character a small amount (similar to taking a single step); and high-precision throwing, where control input has a detailed relationship to the direction and distance of a throw. To test the idea of physical controls in sports video games, we developed two games and ran two studies. The first game was a track-and-field running game called Track and Field Racing (TaFR), where two players race each other in a simulated 100m, 200m, or 400m race. Players controlled the running movement by alternately pressing two keys on the keyboard, as fast as possible, with their first and second fingers. We ran three track meets with TaFR to see whether the physical controls led to individual differences and to performance changes over time. Our study clearly showed both of these effects – physical controls appeared to greatly increase the complexity and unpredictability of the game. Our second game – called Jelly Polo – was a three-on-three top-down ball game loosely based on European Handball. Discrete movement in Jelly Polo involves repeatedly tapping the left joystick of a game controller in the desired direction; precise throwing involves pushing the right joystick in a particular direction and by a particular amount. To explore the effects of these controls on gameplay, we ran a four-week ‘Jelly Polo league’ with four persistent teams of three people. We found substantial evidence that the physical controls changed all three of the issues identified above: first, people’s basic skills in discrete running and precise throwing increased over the four weeks; second, there was considerable individual skill difference across the twelve players, and both people and teams had to adjust to accommodate these differences; and third, fatigue played a major role in the gameplay – it directly affected player speed, and led to novel team strategies and exploitation of fatigue-based mismatches. Our work builds on existing foundations of exertion-based interfaces (e.g., [12, 14, 17]), but looks specifically at ‘exertion in the small,’ with physical actions of hands and fingers. We make four novel contributions: we show that small-scale effort-based control can add considerable complexity and unpredictability to simple actions such as running and movement; we show that providing more expressive input provides increased opportunities for expertise (and differentiation in expertise) for both movement and throwing actions; we show that physical controls can add interest, challenge, and enjoyment to simple team games; we show that fatigue can be a valuable design principle that can dramatically change the way that gameplay evolves, and the ways that teams develop strategy. Overall, our work suggests that designers of games based on physical actions (and sports games in particular), should consider ‘exertion in the small’ as an idea that can improve player experience and player satisfaction.Mike Sheinin and Carl Gutwin. 2014. Exertion in the small: improving differentiation and expressiveness in sports games with physical controls. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’14). ACM, New York, NY, USA, 1845-1854. DOI: 10.1145/2556288.2557385
Now, write your own Title, Abstract, and Introduction
First, brainstorm a CHI paper idea with the group (10 minutes). Use Carl’s four questions to guide you through the process of writing a fictional CHI paper about this research topic that you have in mind:
- What is the real-world problem that we are trying to solve?
- Why is it important to solve this problem?
- What is the solution that we came up with to solve it?
- How do we know that the solution is a good solution to the problem?
Use the same process as many CHI authors:
- Sketch the rough answers to each question into bullet points.
- Get together a maximum of 15 bullet points among all questions.
- Start writing out the bullet points into paragraphs.
- Get together enough for at least 1 page in SIGCHI proceedings format
- Present your introduction to the course.
There is much to be learned about writing with style and elegance. An excellent system to follow is Shani Raja’s 9-Step Editing System when trying to trim down your writing. Have a look at the video below. His nine steps are:
- Quick read-through
- Separate the points
- Make it make sense
- Delete the nonessential
- Make each point unique
- Create sections
- Make the sections flow
- Put points into sections
- Make it read well (simple, clear, and elegant)
What you should definitely avoid in writing your CHI paper are the following things:
- Minimize passive voice. Yes, CHI likes when you say “we” and actively describe what your research team has done.
- Find a rhythm to your writing. Avoid sentences that are too long. You should try and alternate your sentence length in your writing.
- Reduce jargon. Yes, HCI is an interdisciplinary field, but there is still lots of jargon that is used across disciplines and sticks out but not using cutting-edge vocabulary.
- Reduce words that do not contribute to the meaning of a sentence. Be clear and be brief in what you say. Delete adverbs whenever possible.
Identify Bad Writing
Consider the following sentence and try to understand its meaning:
“The move from a structuralist account in which capital is understood to structure social relations in relatively homologous ways to a view of hegemony in which power relations are subject to repetition, convergence, and rearticulation brought the question of temporality into the thinking of structure, and marked a shift from a form of Althusserian theory that takes structural totalities as theoretical objects to one in which the insights into the contingent possibility of structure inaugurate a renewed conception of hegemony as bound up with the contingent sites and strategies of the rearticulation of power.”Butler, J. (1997). Further reflections on conversations of our time. Diacritics, 27(1), 13-15.
A Gary Provost Quote on Good Writing
“This sentence has five words. Here are five more words. Five-word sentences are fine. But several together become monotonous. Listen to what is happening. The writing is getting boring. The sound of it drones. It’s like a stuck record. The ear demands some variety. Now listen. I vary the sentence length, I create music. Music. The writing sings. It has a pleasant rhythm, a lilt, a harmony. I use short sentences. And I use sentences of medium length. And sometimes, when I am certain the reader is rested, I will engage them with a sentence of considerable length, a sentence that burns with energy and builds with all the impetus of a crescendo, the roll of the drums. The crash of the cymbals – sounds that say listen to this, it is important.”Gary Provost