Smartdollars: research

Showing posts with label research. Show all posts

Monday, December 5, 2011

Is Kinect Doomed to Casual Gaming?

Weird News Dance Central by Harmonix appears to be one of the better Kinect launch titles (some say the best).

My Last Exams Ever (Hopefully)

Weird News This week I've been very busy with my written comprehensive exams.

CRA-W Grad Cohort: Research Proposals

Weird News What exactly is a research proposal and how do you prepare a good one? The following advice, given at the 2011 CRA-W Grad Cohort, is mainly intended for your PhD thesis proposal, but can apply to other research proposals as well.

What is a PhD Proposal?

Not the end product but rather the brainstorming and planning part of the process
A succinct write-up of your proposed research goals, strategies, justifications, and contributions
Allows you to get feedback from your committee
Helps you focus

General Process

Use your abstract to recruit your committee members
Write the proposal and iterate with your supervisor (length will depend on department)
Give the proposal to the committee after your supervisor agrees it's ready
Prepare a lot, reduce nervousness
Present the proposal to committee
Expect hard questions to follow

Makeup of the Committee

Supervisor
Faculty in topic area
Faculty outside topic area (great for high level feedback)
External member (possibly in your area - good chance for networking)

Role of the Committee

Gives you guidance, feedback, and eventually reference letters
They make sure you know your stuff at the defence, that you have thought about the issues, that the work will be novel and important enough for a PhD, and that your techniques are sound

Goals of Proposal: Be Sure You Clearly Answer These Questions

What problem are you studying?
Why is it important?
What results have you achieved so far and why do they matter?
How substantially different is your approach from prior work?
How will you systematically evaluate your results?
What do you need to complete the work?

Before You Start Writing

Figure out what you want to accomplish
Write a succinct thesis statement or hypothesis
Discuss your ideas with others
Present parts of thesis at seminars, conferences, etc
Think about 3-4 major contributions or papers
Formulate your contributions in writing
Think about your audience (they are not aware of the prior work, your skills, etc), and provide the appropriate background, terminology, and so on
Figure out how the state of the art relates to your problem

Presentation Tips

Attend other proposal/thesis defences
Thank the committee
Introduce yourself and your background
Practice!!
Be polite about interruptions for questions
Ask someone to take note of the questions asked
Remember that it's ok to not have an answer to a question, and to ask for help and direction

Techniques for Telling Nonlinear Stories

Weird News At the beginning of October, I'll be participating in a digital narrative workshop for the GRAND NCE called 'Experiencing Stories with/in Digital Games.' The following abstract describes a round table talk I'll be giving as part of the student gathering:

Non-linear fiction ranges from the use of static plots with events presented in a non-chronological way to interactive story worlds where users make choices that affect the outcome of the story. A wide range of media can be used to implement works of non-linear fiction, including novels, film, storytelling engines, and games. In all of these cases, there are multiple techniques available for designing and telling a non-linear story.

We have compiled a taxonomy for these techniques with broad categories for strategies for explaining non-linearity, structuring stories as graphs, relying on emergent behaviour, creating character driven plots, and designing data-driven interactive worlds. Each of these categories breaks down further with classic examples from all types of media and with approaches reported on in academic literature. For instance, a strategy to explain a story’s non-linearity is to attribute it to time travel (The Legend of Zelda: Orcarina of Time) or to a hallucinating or brain-damaged character (Memento). Many examples use a simple branching structure to change the story based on player choices (Choose Your Own Adventure) while others rely on game data to adjust how other characters in the story react to you (Fallout 3).

In this talk, we will present our taxonomy and frame it in the context of story in games. We will discuss what techniques have been used in existing games, and which have not yet been employed. We will consider why games have not used certain techniques, and use this to suggest how they might do so in the future.

I'm currently hammering out a potential thesis research plan for myself, and this work will contribute toward that.

Game Papers at SIGGRAPH 2011 (Part 1 of 2)

Weird News Following up on my general impressions of SIGGRAPH I wanted to write about a few of the games paper presentations I enjoyed seeing at the conference.