Writing for Computer ScienceReference: Zobel, Justin. "Writing for Computer Science". Springer, London, 2004.
2. Getting Started
- Choosing to explore ideas that seem likely to succeed, or are intriguing, or have the potential to lead to something new, or that contradict received wisdom.
Shaping a Research Project
- Short-term and long-term goal
- Explicitly consider what is needed at the end, then reason backwards.
- Write your Ph.D. thesis to check enough research has yet been done.
Students and Advisors
- Prepare your meetings with your advisor - bring tables of results or list of questions.
Several Characteristics Shared by Successful Research Students
- A willingness to read widely, to explore the field broadly beyond their specific topic, to try things out, and to generally take part in the academic community.
- The enthusiasm to develop their interest in some area, and then ask for advice on how that interest can be turned into a thesis project.
- The ability and persistence to undertake a detailed (and even grueling) investigation of a specific facet of a larger topic.
- The initiative in terms what needs to be done and how to present it, and gradually assume responsibility for all aspects of the research.
- Systematic and organized, and understand the need for rigor, discipline, stringency, quality, and high standards.
- Actively reflect on habits and working practices, and seek to improve themselves and overcome their limitations and knowledge gaps.
- Their work looks plausible, it has the form and feel of high-quality published paper.
The strength to keep working despite some significant failed or unsuccessful activity.
- Getting Stared Checklist
3. Reading and Reviewing
- Papers published in a reputable venue, theses examined at a reputable institution, and books
- The number of papers that a researcher working on a particular project has to know well is usually small, even though the number the researcher should have read to establish their relevance is large.
Finding Research Papers
- Reading about a paper that seems relevant is never a substitute for reading the paper itself. If you need to discuss or cite a paper, read it first.
- Exploring relevant literature: (1) use obvious search terms to explore the Web or use the search engine; (2) visit the websites of research groups and researchers working in the area; (3) follow up the references in promising research papers; (4) discuss your work with as many as people as possible.
- Once there are several versions of the same paper: preprint in an online archive, a conference version, and a journal version. Usually the polished work that has appeared in a journal is the definitive one.
- It is hard to find all relevant work, but finding all significant work is a critical part of doing research.
- A cynical but often accurate rule of thumb is that work that is more than one or two years old and has not been published in a significant venue probably has some serious defect.
- Critical questions list
Developing a Literature Review
- Develop the literature review progressively
Authors, Editors, and Referees
- Originality and validity
Evaluation of Papers
- Check the quality and quanity of references
Content of Reviews
- Is the case for or against the paper convincing?
- Is there adequate guidance for the authors?
Drafting a Review
- Be prapred to change the minds of a paper
- Be constructive
- Be polite
Checking Your Review