TELE9752 Group presentations
The papers to be presented are listed on the wiki
Note that the lecturer cannot guarantee the accuracy of content included in student presentations (e.g. there is no time to review presentations before they are delivered, nor to correct each inaccuracy during delivery) so students in the audience should be skeptical about the accuracy of what is said. e.g. in 2010, one group were given a RFC (Internet Request For Comments) to present, and kept referring to RFC as “Remote Function Call”!
Each research presentation group should email their slides to Tim before 4pm Oct 22 2013, so that I can make them available on the course web page, to facilitate the presentations and for future reference by other students
Plan on 12 minutes to present your paper, followed by 3 minutes for questions. 4 papers in the 1st hour, then 3 in each of the 2nd and 3rd hours.
Could each group please email Tim your presentation before class on Tuesday so that I can get them together on the theatre computer before class, so as to minimise time wasted between presentations?
To prepare for exam questions about these presentations, you should have
Tips for presentations:
- a deep understanding of your paper (e.g. as obtained by developing your group's presentation), and
- a broad understanding of the main idea of most of the papers (e.g. as obtained by listening to the presentations),
- a moderately deep understanding of one or two other papers from this list (chosen according to your interests, and obtained by reviewing the presentation and paper).
Useful databases of papers including those about NOC research:
- Summarise the paper in the 2nd slide This should be informative - think about it as if it was the only information you are allowed to know about the paper before you had to go off and spend the rest of your life working on this topic. Don't include platitudes as occur in many abstracts (e.g. "This paper introduces a new technique for network management and shows how it is more effective than existing techniques by using simulation an analysis") but be informative
- Number slides and on the first slide indicate how many slides you plan to present. The numbering helps the audience raise questions after the presentation (“on slide 123 you said...”) and knowing how many slides you plan to present helps the moderator estimate whether you're running out of time. You might want to also include backup slides that you don't plan to present, but which might be handy in answering expected questions.
- Don't overload the audience - remember that yours will be one of 10 presentations made in a 3-hour class, and that the audience understanding of the presentations will be assessed in the final exam.
- You don't have to cover everything in the paper; be selective and do a good job of covering the core of the paper, even if that means omitting some of the minor parts of the paper. (e.g. any proposal(s) made by the paper would likely be core, but descriptions of related work, some esoteric performance evaluations or extensions of the proposal to deal with unusual cases would usually not be as central to the paper).
- Use figures; to capture the attention of the audience and often “ a picture tells a thousand words ”.
- Acknowledge your sources e.g. by citing them in a references slide at the end of your presentation. Since your presentation is about a particular paper, you are free to draw from that paper, but you should cite any other sources.
- You should relate the paper to
- concepts covered in the course, e.g. mention which aspect(s) of FCAPS the paper contributes to, which techniques does the paper use that were covered in the course, does the paper suggest doing things in a way that differs from ways that may be described in the course?
- other papers, e.g. differing opinions about how the problem should be solved, other work by the same authors, follow-on works (e.g. which cite this paper), etc
- Give your own opinion about what you think is good or bad about the paper, e.g. how could it be improved?
- Point at your presentation in a way that is meaningful to the audience, e.g. not with your finger (unless you're touching the screen) because the audience has a different perspective. e.g. use the mouse to point.
Some journals that focus on network management research: